Engineering Mathematics I

Engineering
Mathematics-I
Prof. C. Nahak,
Prof. J. Kumar
Prof. S. Kumar,
ENGINEERING MATHEMATICS – I (3+0)
Course Developer
Prof. C. Nahak, Prof. J. Kumar & Prof. S. Kumar
Mathematics Department, IIT Kharagpur, Kharagpur 721 302)
AgriMoon App AgriVarsha App

App that helps the students to gain the App that helps the students to All Agricultural
Knowledge about Agriculture, Books, News, Competitive Exams IBPS-AFO, FCI, ICAR-
Jobs, Interviews of Toppers & achieved peoples, JRF, SRF, NET, NSC, State Agricultural
Events (Seminar, Workshop), Company & exams are available here.
College Detail and Exam notification.
Index
Lesson Name Page No
Module 1: Differential Calculus
Lesson 1: Rolle’s Theorem, Lagrange’s Mean Value Theorem, 5-13
Cauchy’s Mean Value Theorem
Lesson 2: Taylor’s Theorem / Taylor’s Expansion, Maclaurin’s 14-19
Expansion
Lesson 3: Indeterminate forms ; L’Hospital’s Rule 20-29
Lesson 4: Limit, Continuity of Functions of Two Variables 30-46
Lesson 5: Partial and Total Derivatives 47-61
Lesson 6: Homogeneous Functions, Euler’s Theorem 62-68
Lesson 7: Composite and Implicit functions for Two Variables 69-77
Lesson 8: Derivative of Higher Order 78-84
Lesson 9: Taylor’s Expansion for Function of Two Variables 85-90
Lesson 10: Maximum and Minimum of Function of Two Variables 91-96
Lesson 11: Lagrange’s Multiplier Rule / Constrained Optimization 97-104
Lesson 12: Convexity, Concavity and Points of Inflexion 105-111
Lesson 13: Curvature 112-122
Lesson 14: Asymptotes 123-132
Lesson 15: Tracing of Curves 133-144
Module 2: Integral Calculus
Lesson16: Improper Integral 145-152
Lesson17. Test for Convergence 153-163
Lesson 18: Rectification 164-174
Lesson 19: Volume and Surface of Revolution 175-185
Lesson 20: Double Integration 186-196
Lesson 21: Triple Integration 197-204
Lesson 22: Area & Volume using Double and Triple Integration 205-213
Lesson 23: Gamma Function 214-222
Lesson 24: The Beta Function 223-228
Module 3: Ordinary Differential Equations
Lesson 25: Introduction 229-233

Lesson 26: Differential Equations of First Order 234-239
Lesson 27: Linear Differential Equation of First Order 240-245
Lesson 28: Exact Differential Equations of First Order 246-251
Lesson 29: Exact Differential Equations : Integrating Factors 252-258
Lesson 30: Linear Differential Equations of Higher Order 259-264
Lesson 31: Linear Differential Equations of Higher Order 265-270
Lesson 32: Linear Differential Equations of Higher Order (Cont.) 271-276
Lesson 33: Method of Undetermined Coefficients 277-282
Lesson 34: Method of Variation of Parameters 283-288
Lesson 35: Equations Reducible to Linear Differential Equations 289-294
with Constant Coefficient
Lesson 36: Methods for Solving Simultaneoous Ordinary Differential 295-300
Equations
Lesson 37: Series Solutions about an Ordinary Point 301-306
Lesson 38: Series Solutions about an Ordinary Points (Cont.) 307-312
Lesson 39: Series Solutions about a Regular Singular Point 313-318
Lesson 40: Series Solutions about a Regular Singular Point (Cont..) 319-324
Module 4: Vector Calculus
Lesson 41: Introduction 325-332
Lesson 42: Gradient and Directional Derivatives 333-337
Lesson 43: Divergence and Curl 338-341
Lesson 44: Line Integral 342-347
Lesson 45: Green’s Theorem in the Plane 348-352
Lesson 46: Surface Integral 353-357
Lesson 47: Stokes’s Theorem 358-362
Lesson 48: Divergence Theorem of Gauss 363-367
Lesson 1
Rolle’s Theorem, Lagrange’s Mean Value Theorem , Cauchy’s Mean Value

Theorem
1.1 Introduction
In this lesson first we will state the Rolle’s theorems, mean value theorems and
study some of its applications.
Theorem 1. 1 [Rolle's Theorem]: Let be continuous on the closed interval
and differentiable on the open interval . If , then there
exists at least one number in such that .
Proof: Assume . If and , then we
consider instead of . Since is continuous on it attains
its bounds: Let and be both maximum and minimum of on . If
, then is throughout i.e., is constant on
for all in . Thus at least one such that .
Suppose . If varies on then there are points where
or points where Without loss of generality assume and the
5 www.AgriMoon.Com
Rolle’s Theorem, Lagrange’s Mean Value Theorem, Cauchy’s Mean Value Theorem
function takes the maximum value at , so that . It is to be noted
that if , , which is a contradiction. Now as
is the maximum value of the function, it follows that ,
both when and .
Hence,
when
when . Since it is given that the derivative at exists, we get
when and when . Combining the two
inequalities we have, .
Note: Rolle’s theorem shows that b/w any two zero’s of a function there
exists at least one zero o i.e., clearly is continous on [-1,1]
Example 1: Verify the Roll's theorem for .
6 www.AgriMoon.Com
Solution:
(i) , (ii) is differentiable on , so all conditions of
Roll's theorems are satisfying. Hence implies and
Example 2: in .
Solution:
, is continuous. But is not differentiable at .
Note that , for which is differentiable. As , for
and , for .
Example 3: Show that the equation , has only one real root
Solution:
is an odd degree polynomial, hence it has at least one
real root as complex roots occurs in pair.
Suppose two real roots such that , then on , all
properties of Roll's theorem satisfied, hence , such that ,
7 www.AgriMoon.Com
But , a contradiction to
Rolle’s therorem. Hence the equation has only one real root.
1.2. Mean Value Theorems
Theorem 1.2 [Lagrange's Mean Value Theorem]: If a function is
continuous on , differentiable , then there exists at least one point ,
such that . Hence Lagrange's mean
value theorem can be written as
, where .
Geometrical Representation: If all points of the arc there is a tangent line,
then there is a point between and at which the tangent is parallel to the
chord connecting the points and .
1.2.1 Cauchy's Mean Value Theorem
Cauchy's mean value theorem, also known as the extended mean value theorem,
is the more general form of the mean value theorem.
Theorem 1.2 [Cauchy's Mean Value Theorem]: It states that if functions
and are both continuous on the closed interval , and differentiable on the
8 www.AgriMoon.Com
open interval and then there exists some , such
that
Note 1: Cauchy's mean value theorem can be used to prove L'Hospital's rule.
The mean value theorem (Lagrange) is the special case of Cauchy's mean value
theorem when .
Note 2: The proof of Cauchy's mean value theorem is based on the same idea as
the proof of the mean value theorem
1.2.2 Another form of the statement: If and are derivable in
and for any , then there exists at least one
number such that
Example 4: Write the Cauchy formula for the functions ,
on .
9 www.AgriMoon.Com
Solution:
Clearly iff
, . . Hence
i.e., implies , so .
1.2.3 The Intermediate Value Theorem It states the following: If is
continuous on , and is a number between and , then there is a
such that .
1.2.4 Applications of the Mean Value Theorem to Geometric properties of

Functions.
Let be a function which is continuous on a closed inteval and assume
has a derivative at each point of the open interval . Then we have
1. (i) If , is strictly increasing on .

2. (ii) If , is strictly decreasing on
.
3. (iii) If , is constant.
10 www.AgriMoon.Com
Intermediate value Theorem for Derivatives: If exists for ,
with then for any number between and there is a
number where .
Application: If exists with , on any interval then has a
differentiable inverse, there.
Converse of Rolle’s theorem : - (need not true).
Example 1.5 Let be continuous on and differentiable . If
such that , does it follow that ?
Solution:
No: Take for example on , implies .
But and .
Example 1.6 Show that
11 www.AgriMoon.Com
Solution:
Let on , By mean value theorem ,
But , and , for all . Hence
Example 1.7 Show that for all .
Solution:
Let on . By mean value theorem =
but for all . Hence the results.
Questions: Answer the following question.
1. Verify the truth of Rolle’s theorem for the functions
(a) f ( x) = x 2 − 3 x + 2 on [1,2]
(b) f ( x) =( x − 1)( x − 2)( x − 3) on [1,3]
(c) f ( x) = sin x on (a) [0, π ]
2. The function f ( x) = 4 x 3 + x 2 − 4 x − 1 has roots 1 and -1. Find the root of the
derivative f ′( x) mentioned in Rolle’ s throrem.
3. Verify Lagrange’s formula for the function f ( x=

) 2 x − x 2 on [0,1].
4. Apply Lagrange theorem and prove the inequalities
(i) e x ≥ 1 + x (ii) ln(1 + x) < x ( x > 0)
12 www.AgriMoon.Com
(iii) b n − a n < nb n−1 (b − a ) for (b > a )
sin x
5. Using Cauchy’s mean value theorem show that lim =1
x →0 x
Keywords: Rolle’s Theorem, Lagrange's and Cauchy’s mean value;

L'Hospital's rule; Intermediate value.
References
Thomas, W. Finny. (1998). Calculus and Analytic Geometry, 6th Edition,
Publishers, Narsa, India.
R. K. Jain, and Iyengar, SRK. (2010). Advanced Engineering Mathematics, 3 rd

Edition Publishers, Narsa, India.
Widder, D.V. (2002). Advance Calculus 2nd Edition, Publishers, PHI, India.
Piskunov, N. (1996). Differential and Integral Calculus Vol I, & II, Publishers,
CBS, India.
Suggested Readings
Tom, M. Apostol. (2003). Calculus, Volume II Second Editions, Publishers,

John Willey & Sons, Singapore.
13 www.AgriMoon.Com
Lesson 2
Taylor's theorem / Taylor’s expansion, Maclaurin’s expansion
2.1 Introduction
In calculus, Taylor's theorem gives us a polynomial which approximates the

function in terms of the derivatives of the function. Since the derivatives are
usually easy to compute, there is no difficulty in computing these polynomials.
A simple example of Taylor's theorem is the approximation of the exponential

function near .
The precise statement of the Taylor’s theorem is as follows:
Theorem 2.1: If is an integer and is a function which is times
continuously differentiable on the closed interval and times
differentiable on the open interval , then
14 www.AgriMoon.Com
Here, n denotes the factorial of , and is a remainder term, denoting the
difference between the Taylor polynomial of degree n and the original function.
The remainder term depends on and is small if is close enough to .
Several expressions are available for it. The Lagrange form is given by
where
If we put , Taylor's formula reduces to Maclaurin's formula.
where lies between and .
Notes
• In fact, the mean value theorem is used to prove Taylor's theorem with the
Lagrange remainder term.
• The Taylor series of a real function that is infinitely differentiable in a
neighborhood of a real number , is the power series of the form
15 www.AgriMoon.Com
• In general, a function need not be equal to its Taylor series, since it is

possible that the Taylor series does not converge, or that it converges to a
different function.
• However, for some functions , one can show that the remainder term
approaches zero as approaches . Those functions can be expressed
as a Taylor series in a neighbourhood of the point and are called analytic.
Example 2.1 Show that
Solution:
Here , , So
. But for and
16 www.AgriMoon.Com
Example 2.2 . Find the Taylor series expansion of
Solution:
for , we have
17 www.AgriMoon.Com
Example 2.3 : Find if
Ans: .
Questions: Answer the following questions.
1. Expand in power of x − 2 of the polynomial x − 5 x + 5 x + x + 2.

4 3 2
2. Expand in power of x + 1 of the polynomial x + 2 x − x + x + 1.

5 4 2
3. Write Taylor’s formula for the function y = x when a = 1, n = 3.
4. Write the Maclaurin formula for the function y = 1 + x when n = 2.
5. Using the results of above problem, estimate the error of the approximate
1 1
equation 1 + x ≈ 1 + x − x 2 when x = 0.2.
2 8
6. Write down the Taylor’s expansion for the function f ( x) = sin x about the point
π
a= with n = 4.
4
x x2 x
7. Applying Taylor’s theorem with remainder prove that 1 + − < 1 + x < 1 + if
2 8 2
x > 0.
8. Applying Maclaurin’s theorem with remainder expand
(i) ln (1 + x) (ii) (1 + x) .
m
18 www.AgriMoon.Com
Keywords: Taylor’s Formula, Taylor’s Series, Maclaurin Formula and Series.
References
W. Thomas Finny (1998). Calculus and Analytic Geometry, 6th Edition, Publishers,
Narsa, India.

CBS, India.
Suggested Readings
Tom M. Apostol. (2003). Calculus, Volume II Second Editions, Publishers,John

Willey & Sons, Singapore.
19 www.AgriMoon.Com
Lesson 3
Indeterminate forms ; L’Hospital’s Rule
3.1 Introduction
Consider the following limits and
In the first limit if we put we will get and in the second limit if we
“plugged” in infinity we get (recall that as goes to infinity a polynomial
will behave in the same fashion that it’s largest power behaves). Both of these
are called Indeterminate form.
3.1.1 Indeterminate forms
First limit can be found by the factorizing the numerator cancelling the common
factor. That is
The second limit can be evaluated as:
20 www.AgriMoon.Com
However what about the following two limits. and , This
first is a indeterminate form, but we can’t factor this one. The second is an
indeterminate form, but we can’t just factor an out of the numerator. Does
there exists some method to evaluate the limits? The answer is yes. By
(L'Hospital's Rule).
Suppose that we have one of the following cases,
or
where can be any real number, infinity or negative infinity. In these cases we
have,
21 www.AgriMoon.Com
Theorem 3.1: Suppose the functions and in , satisfy the
Cauchy Theorem and , then if the ratio has a limit as
, there also exists , and .
Proof.: On the interval take some point . Applying the Cauchy's
mean value theorem we have
where is a number lies between and . But it is given that
and so
..................(1)
If , then , since lies between and . Suppose if
, by (1) exists and is equal to . Hence
22 www.AgriMoon.Com
and, finally,
Note 3.1: The theorem also holds for the case where the functions and
are not defined at , but . We can
make them to be continuous at by redefine ,
, since does not depend on whether the
function and are defined at .
Note 3.2: If and the derivatives and satisfy the
conditions that we imposed by the theorem on the functions and ,
then applying the L'Hospital rule , and so forth.
23 www.AgriMoon.Com
Note 3.3: If , but , then the theorem is applicable to the
reciprocal ratio , which tends to zero as . Hence, the ratio tends to
infinity.
Example 3.1:
Note 3.4: The L'Hospital rule is also applicable if and
Put , we see that as and therefore , and
. Applying the L'Hospital rule to the ratio
, we find that
24 www.AgriMoon.Com
which proves the results.
We also stated in earlier that if both and approaching infinity as
(or , the L'Hospital rule is also applied.
Example 3.2: Find (
Solution:
Taking derivative both numerator and denominator five times we obtain: Ans: 3
Other Indeterminate forms :
The other indeterminate forms reduce to the following cases. (a) (b) (c)
(d) (e) .
25 www.AgriMoon.Com
(a) Let , , it is required to find
i.e. the indeterminate form . Now
or If &
which is ( )- form or one can write
( )- form
Example 3.3
26 www.AgriMoon.Com
b) Let , , it is required to find
. Put . Taking logarithms of both sides of it, we
have
(by the continuity of ) and if .
Similarly we can find the Indeterminate form
Example 3.4: Solution: Put ,
So .
Example 3.5: Find the
Ans: 1
27 www.AgriMoon.Com
Example 3.6 Using Taylor's formula compute
Ans: 1
Evaluate the following limits :
1.
2.
3.
4.
5.
6.
7.
28 www.AgriMoon.Com
8.
Ans.: 1. 1, 2. -2, 3. Limit does not exist, 4. 2, 5. 0, 6. , 7. & 8. 1
Keywords: Indeterminate forms ; L’Hospital’s Rule.
References
W. Thomas, Finny (1998). Calculus and Analytic Geometry, 6th Edition,
Jain, R. K. and Iyengar, SRK. (2010). Advanced Engineering Mathematics, 3

rd Edition Publishers, Narsa, India.
CBS, India.
Suggested Readings
Tom M. Apostol (2003). Calculus, Volume II Second Editions, Publishers,John

29 www.AgriMoon.Com
Lesson 4
Limit, Continuity of Functions of Two Variables
4.1 Introduction
So far we have studied functions of a single (independent) variables. Many

familiar quantities, however, are functions of two or more variables. For
instance, the work done by the force and the volume of the rigid
circular cylinder ( ) are both functions of two variables. The volume of
a rectangular solid ( ) a function of three variables. The notation for a
function of two or more variables is similar to that for a function of single

variable.
Example 4.1: (two variables)
Example 4.2: (three variable)
A function of two variables is a rule that assigns a real number to
each ordered pair of real numbers in the domain of . The range of is
the set of all values of the function: where .
30 www.AgriMoon.Com
In concrete terms: A function is usually just a formula involving the
two variables and . For every and we put in, we get a number out. The
set of all we allowed to put into the function is called the domain of the
function. Usually the domain is unspecified, and then the domain is the set of all
we can put into the formula for and not get square roots of negatives, or
division by zero, or some such. i.e.,the domain is usually the set of all we
can put into the function without getting an undefined expression.

This is the natural domain. The range is simply all the numbers we can “hit”
by putting all from the domain into the function.
Example: 4.3: Let . The domain is the disk of radius
7, centre at origin. Now will be bigger if ar each smaller. So
is biggest when . This is . Now the smallest value
can achieve is 0, when (which happens, for example when
and ). If , could not be defined. Hence the
range is [0,7].
31 www.AgriMoon.Com
Definition. The graph of a function of two variables is the set
where is the domain of . That is, the graph is the surface in 3-
dimensinal Euclidean Space .
4.1.1 A contour curves or level curves
A contour curve for a function is a trace of the surface
parallel to the -plane. That is, let for some number , and plot
in the -plane.
The domain of a function of two variables , which is denoted from
now onwards is the set of all points in the -plane for which is
defined. For example, means that is the set of points
such that is greater than .
Example 4.4. Determine the domain of
32 www.AgriMoon.Com
Solution:
Since the argument of must be positive, the domain of is the set of
points for which the denominator is not equal to 0. However,
means that . In set notation this is written as .
Most of the sets in the -plane we encounter will be bounded by a closed
curve.
As a result, we define an open region to be the set of all points inside of but not
including a closed curve, and we define a closed region to be the set of all
points inside of and including a closed curve.
Equivalently, a point is said to be a boundary point of a set S if any circle
centered at contains both points inside of and outside of S, and
correspondingly, a set S is open if it contains none of its boundary points and

closed if it contains all of its boundary points.
Example 4.5. Determine if the domain of the following function is open or

closed.
33 www.AgriMoon.Com
Solution:
To begin with, the quantity cannot be negative since it is under the
square root. Thus, the domain of is the set of points that satisfy
or .
That is, the domain is the set of points inside and on the circle of radius 3
centered at the origin, which we write as .
Moreover, the domain is a closed region of the -plane since it contains the
boundary circle of radius 3 centered at the origin.
We say that a region S is connected if any two points in can be joined by a
curve which is contained in S:
4.1.2 Functions of Space and Time
Functions of two variables are important for reasons other than that their graph
is a surface. In particular, a function of the form u(x,t) is often interpreted to be
a function of x at a given point in time. For example, let's place an xy-
coordinate system on a violin whose strings have a length of l, If u(x,t) is
considered the displacement of a string above or below a horizontal line at a
point and at a time , then y = u(x,t) is the shape of the string at a fixed time t.
34 www.AgriMoon.Com
Likewise, u (x, t) might represent the temperature at a distance xfrom one end of
the rod at time .
4.1.3 Limits and Continuity
Now we will extend the properties of limits and continuity from the familiar
function of one variable to the new territory of functions of two or more
variables.
Let us recall limit of function of single variable: Let be a function defined on
an open interval containing (except possible at ) and let be a real number.
The statement means that for given , there exists a
such that , whenever .
In less formal language this means that, if the limit holds, then gets closer
and closer to as gets closer and closer to .
Consider the following limits.
Good job if you saw this as “limit does not exist” indicating a vertical
asymptote at .
35 www.AgriMoon.Com
This limit is indeterminate. With some algebraic manipulation, the zero factors
could cancel and reveal a real number as a limit. In this case, factoring leads
to……
The limit exists as approaches 2 even though the function does not exist. In
the first case, zero in the denominator led to a vertical asymptote; in the second
case the zeros cancelled out and the limit reveals a hole in the graph at .
The concept of limits in two dimensions can now be extended to functions of

two variables.
Definition 4.1 Let f be a function of two variables defined on an open disc

centered at i.e., , except polssible
at , and let L be the real numbers Then
36 www.AgriMoon.Com
if given such that
whenever < .
Graphically for any point in the disc with radius , the value
lies between and .
Example 4.6 Let .
For the limit of this function to exist at (-1,3), values of must get closer to 13
as points on the -plane get closer and closer to (-1,3).
. For proof we have to go back to epsilon and delta.
Example 4.7 Verifying the limit by definition
Solution:
We have to show that whenever . Now
. Let .
37 www.AgriMoon.Com
Example 4.8. Show that .
Solution:
Now
Put , whenever .
Example 4.9.
Solution:
To show that , whenever
Now
. Set .
For a single variable function we have has two direction i.e.,
38 www.AgriMoon.Com
But in case of function of two variables the ,
approaches to in infinitely many directions.
Example 4.10: Test whether exists.
Solution:
Let on the line . So
As depend on , so the limit does not exist.
Example 11: Solution: Let , , implies
. The limit becomes .
Definition of Continuity of a Function of Two Variables
39 www.AgriMoon.Com
A function of two variables is continuous at a point in an open region if
is equal to the limit of as approaches . In limit
notation:
Give Definition
The function is continuous in the open region if is continuous at every
point in .
The following results are presented without proof. As was the case in functions
of one variable, continuity is “user friendly”. In other words, if is a real
number and and are continuous functions at then the functions below
are also continuous at :
40 www.AgriMoon.Com
The conclusions indicate that arithmetic combinations of continuous functions

are also continuous —that polynomial and rational functions are continuous on
their domains.
Finally, the following result asserts that the composition of continuous functions
are also continuous. If is continuous at and is continuous at ,
then the composition function is continuous at
and
Example 4.12 Find the limit and discuss the continuity of the function
Solution:
. The function will be continuous when
Example 4.13. Using and show that the function is
continuous at origin.
41 www.AgriMoon.Com
Solution:
Set and ( is fixed). Then
. Take .
Example 4.14. Is it possible to define at so that is
continuous?
Solution:
Note that
where and . If we define , is continuous
every where.
42 www.AgriMoon.Com
Example 4.15. Show that the function is continuous if we define
Solution:
Discontinuity possible only at . Note with and , from
for small , that ; hence limit at exists and is
0.
Property 1: If a function is defined and continuous in a closed and
bounded domain , then there will be at least one point in such that
And at least one point such that
43 www.AgriMoon.Com
We call as the maximum value of the function and
is the minimum value of the function. This result states that a
function which is continuous on a closed and bounded domain has a
maximum and minimum.
Property 2: If has both maximum and minimum and respectively,
let , then such that .
Corollary to property 2.
If a function is continuous in a closed and bounded domain and
assumes both positive and negative values, then there will be a point inside the
domain at which the vanishes.
1. Find , if it exists.
2. Show that
3. Prove that .
4. Find
44 www.AgriMoon.Com
5. Find , if it exists.
6. Test for continuity
7. Find the and discuss the continuity of the
function at .
8. Find the
and discuss the continuity of the function at .
Example 1: Let for and for
. Is it continuous at or can we make continuous by
redefining ? (Hint: not possible)
Example 2: Is it possible to extend to the origin so that the
resulting function is continuous? (Hint: not possible)
Keywords: Limit, Continuity, Maximum and Minimum values.
References
45 www.AgriMoon.Com
Jain, R. K. and Iyengar, SRK. (2010), Advanced Engineering Mathematics, 3

CBS, India.
Suggested Readings

46 www.AgriMoon.Com
Lesson 5
Partial and Total Derivatives
5.1 Introduction
Let , we denote as the partial derivative of with respect to and
define as
and similarly
Example 5.1: Given , find the partial derivative of and
Solution:
, .
The partial derivatives of a function of any number of variables are determined

similarly. Thus if
47 www.AgriMoon.Com
Partial and total derivatives
Informally, we say that the values of and at the point denote
the slope of the surface in the - and -directions, respectively.
Example 5.2: Find the slopes of the surface given by
at the point in the -direction and the -direction.
Solution:
5.1.2 Differentiability for Functions of Two Variables
We begin by reviewing the concept of differentiation for functions of one

variable. We define the derivative in case of function of single variable.
Let and let be an interior point of . Then is differentiable at
means
48 www.AgriMoon.Com
or equivalently
exists. The number is called the derivative of at .
Geometrically the derivative of a function at is interpreted as the slope of the
tangent line to the graph of at the point .
Extending the definition of differentiability in its present form to functions of

two variables is not possible because the definition involves division and
dividing by a vector or by a point in two dimensional space is not possible. To
carry out the extension, an equivalent definition is developed that involves
division by a distance. The limit statement can be rewritten as
So the following definition is equivalent to the original one.
49 www.AgriMoon.Com
Let and let be an interior point of . Then is differentiable at
means there is a number, , such that
One way to interpret this expression is that tends
to 0 faster than and consequently is approximately equal to
. The equation is the equation of
the line tangent to the graph of at the point . So is
approximated very well by its tangent line. This observation is the bases for
linear approximation.
Using this form of the definition as a model it is possible to construct a

definition of differentiability for functions of two variables.
50 www.AgriMoon.Com
Definition 5.1. Let and let be an interior point of .
Then f is differentiable at means there are two numbers,
and such that
The vector
or
is called the derivative of at the point . Interpret this definition as
requiring that the graph of has a tangent plane at the point .
In fact it is easy to get an equation for this tangent plane. It is

. In
, the same symbol is use for two different purposes. First
51 www.AgriMoon.Com
as a subscript where it denotes the variable of differentiation and second as the

first coordinate of a point in . Strictly speaking such a dual use of one symbol
is improper, but this is so common as to be acceptable. In the general case, the

derivative is a vector in space and it is computed by computing all of the first
order partial derivatives. As in the case of functions of one variable,

differentiability implies continuity.
For functions of one variable if the derivative, , can be computed, then is
differentiable at . The corresponding assertion for functions of two variables is
false, as we know existence of partial derivative does not mean the function of
two variable is continuous. We might suspect that if is continuous at
and the first order partial derivatives exist there, then is differentiable at
but that conjecture is false as the following example shows.
Example 5.1. Let if and .
Solution:
So if were differentiable at (0, 0),we would have that
as and . That is
52 www.AgriMoon.Com
. But if the limit is computed along the path , we
get .
The natural question to ask then is under what conditions can we conclude that
is differentiable at . The answer is contained in the following theorem.
Theorem 5.1. Let and let be an interior point of . Suppose
all of the first order partial derivatives of exist in a open disk about
and are continuous at . Then is differentiable at .
Example 5.2. Show that the function is differentiable
everywhere in its domain.
Solution:
The domain of is all of except for the origin. We shall show that has
continuous partial derivatives everywhere in its domain (that is, the function
is in ). The partial derivatives are and . Since each of
and is the quotient of continuous functions, the partial derivatives are
53 www.AgriMoon.Com
continuous everywhere except the origin (where the denominators are zero).
Thus, is differentiable everywhere in its domain.
We know that if a function is differentiable at a point, it has partial derivatives

there. Therefore, if any of the partial derivatives fail to exist, then the function
cannot be differentiable. This is what happens in the following example.
Example 5.3: Consider the function . Is it differentiable at
the origin.
Solution:
Let us find the partial derivatives if they exist at . Now
Since the limit does not exit so does not exit. Similarly we can show
also does not exist. Thus cannot be differentiable at the origin.
54 www.AgriMoon.Com
In Example 5.3 the partial derivatives and did not exist at the origin and
this was sufficient to establish non differentiability there.
In the following example even if both of the partial derivatives, and
, exist is not differentiable at (0, 0).
Example 5.4: Consider the function . Show that the partial
derivatives and exist, but that is not differentiable at (0, 0).
Solution:
Now
and similarly . Suppose the function is differentiable at (0,0),
i.e.,
That is
55 www.AgriMoon.Com
If this limit exists, we get the same value no matter how and approach 0.
Suppose we take . Then the limit becomes
But this limit does not exist, since small values for will make the fraction
arbitrarily large. Thus, this function is not differentiable at the origin, even
though the partial derivatives and exist.
In summary if a function is differentiable at point, then it is continuous there.

Having both partial derivatives at a point does not guarantee that a function is
continuous there.
Theorem 5.1 :
56 www.AgriMoon.Com
Here subscript 1 and 2 denote the partial derivative with respect to its first and
second argument, respectively. The proof is given in Lesson 7.
5.1.2 Total Differential
Definition 5.2 (Total Differential) For a function of two variables,
if and are given increments and, then the corresponding increment of is
The differentials and are independent variables; that is, they can be given
any values. Then the differential , also called the total differential, is defined
by
Example 5.5: If , find the differential .
Further, if changes from 2 to 2.05 and changes from 3 to 2.96, compare the
values of and . Which is easier to compute or ?
Solution:
By definition,
57 www.AgriMoon.Com
Putting , , , and , we get
The increment of is
Notice that but is easier to compute.
5.2 Total derivative: In the mathematical field of differential calculus, the term
total derivative has a number of closely related meanings.
The total derivative of a function, , of several variables, e.g., , , , etc., with
respect to one of its input variables, e.g., , is different from the partial
derivative. Calculation of the total derivative of with respect to does not
assume that the other arguments are constant while varies; instead, it allows
the other arguments to depend on . The total derivative adds in these indirect
58 www.AgriMoon.Com
dependencies to find the overall dependency of on . For example, the total
derivative of with respect to is
Consider multiplying both sides of the equation by the differential .
The result will be the differential change in the function . Because
depends on , some of that change will be due to the partial derivative of with
respect to . However, some of that change will also be due to the partial
derivatives of with respect to the variables and . So, the differential is
applied to the total derivatives of and to find differentials and , which
can then be used to find the contribution to .
Example 5. 6: Find the total derivative of ,
Solution:
, , .
59 www.AgriMoon.Com
1. Test the differentiability of
2. Find the total differential of
3.
4. Find at where
Keywords: Partial Derivative, Differential, Total Differential
References

60 www.AgriMoon.Com
CBS, India.
Suggested Readings

61 www.AgriMoon.Com
Lesson 6
Homogeneous Functions, Euler's Theorem
6.1 Introduction
A polynomial in and is said to be homogeneous if all its terms are of same
degree. For example,
is homogeneous. It is easy to generalize the property so that functions not

polynomials can have this property.
Definition 6.1
A function is homogeneous of degree in a region D iff, for
and for every positive value , . The number is +ve, -
ve, or zero and need not be an integer.
Example 6.1 . Here ; is any quadrant
without the axes.
Example 6.2
62 www.AgriMoon.Com
This function is homogeneous of degree 0; is first and third quadrant without
the axes.
Example 6.3 .
This function is not homogeneous.
Theorem 6.1 [Euler's Theorem] Let is a homogeneous function of
degree in (region) and and are continuous in . Then
for all .
Proof. Now differentiate partially with respect to , we
obtain
Chain rule :
Finally set .
Example 6.4 If . Then show that
63 www.AgriMoon.Com
Proof. Let .
is not homogeneous function , but is
is homogeneous function of degree .Therefore
But ,
Hence
.............................
Differentiating partially w.r.t. , we have
..................(6.2)
64 www.AgriMoon.Com
Differentiating partially w.r.t. , we have
..................(6.3)
Multiplying by , by and adding, we have
Example 6.5 If . Then find
65 www.AgriMoon.Com
Ans.: ).
Example 6.6 (1) If , show that .
Solution:
Here is not a homogenous function but is a homogenous
fucntion of degree 2
i.e.,
or
(2) If show that .
Solution:
is not homogenous function, but is a homogenous function of degree 3 in
By Euler’s theorem, we have
66 www.AgriMoon.Com
i.e.,
1. If , show that
2. If , show that
3. If , prove that
4. If , then find the value of
Keywords: Homogeneous Function, Euler’s Theorem, Parial Derivatives
References

CBS, India.
67 www.AgriMoon.Com
Suggested Readings

68 www.AgriMoon.Com
Lesson 7
Composite and Implicit Functions for Two Variables
7.1 Introduction
The chain rule works for functions of more than one variable. Consider the
function where and , and and are
differentiable with respect to , then
Suppose that each argument of is a two-variable function such that

and , and that these functions are all differentiable.
Then the chain rule would look like:
If we consider above as a vector function, we can use vector notation

to write the above equivalently as the dot product of the gradient of and a
derivative of :
Partial and Total Increment: We consider a function , increase the

independent variable by (keeping fixed), then will be increased: this
69 www.AgriMoon.Com
Composite and Implicit Functions (Chain Rule) for Two Variables
increase is called the partial increment with respect to which we denote as

, so that
Similarly we define . If we increase the argument by and by , we

get a new increment , which is called the total increment of and defined
by
It is noted that total increment is not equal to the sum of the partial increments,
. Let us assume that has continuous partial derivatives
at the point under consideration. Express in terms of partial
derivatives. To do this we have
and using Lagrange mean value theorem separately
(where lies between and and between and ). As partial

derivatives are continuous it follows that
70 www.AgriMoon.Com
Where the quantities and approach zero as and

approach zero.
Now we will derive the total differential of composite function.
Theorem 7.1:
We use this formula for the composite function
Example 7.1:
71 www.AgriMoon.Com
We can generalize this results. If is a function of four

arguments and each of them depends on and , then
If a function , where , , depend on a single independent

variables : , , , then is actually a function of one
variable only.
Hence,
This formula is known as the formula for calculating the total derivative (in
contrast to the partial derivative ).
Example 7.2: Find and of , , .
Solution:
72 www.AgriMoon.Com
, , , , , .
So
In these expressions, we have to substitute and for and

respectively.
Example 7.3: Find the total derivative of ,
Solution:
, , .
73 www.AgriMoon.Com
7.1.1 Let us find the the total differential of the composite function
and and , we know the total differential
Now substitute the expression and defined in the above composite
function, after simplification we obtain
Where and
Example 7.4: Find the total differential of the composite function ,

, .
Solution:
74 www.AgriMoon.Com
7.2 Composite and implicitly Functions:
Let some function of be defined by the equation . We shall

prove the following theorem.
Theorem 7.2 Let a function of be defined implicitly by the equation
where , , are continuous in the domain containing the point
, which satisfies (7.1), also at the point . Then
Proof. Given is a function of two variables , and and is again a

function of so that is a composite function of . Its derivative with respect
to is
75 www.AgriMoon.Com
As is considered as a function of alone, which is identically zero. So we

have
which implies
Example 7.5: An equation is given that connects and
find .
Solution:
, , , by the above theorem
we obtain .
1. Find at where
2. If ,
, then show that
3. .
4. Find , when
76 www.AgriMoon.Com
5. Find , when , any
constant, .
Keywords: Chain Rule, Composite Function
References

Jain, R. K. and Iyengar, SRK. (2010). Advanced Engineering Mathematics, 3 rd

CBS, India.
Suggested Readings

77 www.AgriMoon.Com
Lesson 8
Derivative of Higher Order
8.1 Introduction
Derivative of higher order of composite function may be computed by the

principles given in Lesson 7. As an example, let us compute three drivatives of
order two for the function We assume that three
functions along with partial derivatives are continous upto order 3. First let us
consider the higher order partial derivatives.
8.1.1 For , we assume that the three fucntions
Differentiating again, remember that are themselves composite
functions.
78 www.AgriMoon.Com
We omit the arguments in these fucntions to have space. If we admit that
then it is easily shown that .
8.1.1 Higher-order partial derivatives As is true for ordinary derivatives, it is

possible to take second, third, and higher order partial derivatives of a function
of several variables, provided such derivatives exist.
It is not true in general
Example 8.1 Let ; for and .
Solution:
We have
79 www.AgriMoon.Com
Hence
So
80 www.AgriMoon.Com
i.e., .
8.1.2 Partial Derivatives of Higher Order (Equality of and ).
If possesses continuous second order partial derivatives and ,
then
Note: Existence of partial derivatives does not ensure continuity of a function.
Example 8.2 Let ; for and .
Solution:
But is discontinuous at .
81 www.AgriMoon.Com
Example 8.3 If , show that
Solution:
i.e., .
Example 8.4 If , show that
Solution: We have and . Now
So
82 www.AgriMoon.Com
Example 8.5 Let , where and are assumed
to have continuous second partial derivatives, show that .
Solution:
1. For , find ,
2. Find , if by not eleminating and
3. Show that the functions z = φ ( x 2 − y 2 ) , where φ (u ) is a differentiable function,

∂z ∂z
satisfies the relationship y + x = 0.
∂x ∂y
dy
4. Find the derivatives of the functions represented implicitly
dx
2 2 2
(i ) sin ( xy ) − e − x y = 0 (ii ) xe + ye − e = 0 (iii ) y = x (iv) x + y = a
xy 2 y x xy x y 3 3 3
5. If r = xφ ( x + y ) + yψ ( x + y ) , show that
83 www.AgriMoon.Com
∂ 2r ∂ 2r ∂ 2r
− 2 + = 0.
∂x 2 ∂x∂y ∂y 2
( φ and ψ are twice differentiable function.)
1
6. If u = [φ (ax + y ) + φ (ax − y )] , show that
y
∂ 2u a 2 ∂  2 ∂u 
= ⋅ y .
∂x 2 y 2 ∂y  ∂y 
Keywords: Higher order derivatives, higher order partial derivatives
References
Jain,R. K. and Iyengar,SRK. (2010). Advanced Engineering Mathematics, 3 rd

CBS, India.
Suggested Readings

84 www.AgriMoon.Com
Lesson 9
Taylor's expansion for function of two variables
9.1 Introduction
Let which is continuous, together with all its partial derivatives up
to -th order inclusive, in some neighborhood of a point . Then like
a function of single variable we can represent as sum of an -th degree
polynomial in power of and and some remainder. We consider
here in case and show that has of the form
where are independent of and , and is the remainder, and
it is very similar to function of single variable.
Let us apply the Taylor formula for function of the variable assuming
to be constant.
85 www.AgriMoon.Com
Taylor's Expansion for Function of Two Variables
where , . We expand the functions ,
, in a Taylor's series in powers of
where ,
where ,
where , . Substituting expression (3), (4) and (5)
into formula (2), we get
86 www.AgriMoon.Com
arranging the numbers as given in (1), we have
This is the Taylor's formula for . The expression
87 www.AgriMoon.Com
This is called the remainder. If we denote , , and
, becomes
Example 9.1: Find the remainder of the function given by
Solution:
Where is given by
88 www.AgriMoon.Com
Questions: Answer the following question.
π π
1. Expand z = sin x sin y in powers of ( x − ) and ( y − ) . Find the terms of
4 4
the first and second orders and R2 (the remainder of second order).
2. Let f ( x, y ) = e sin y . Expand f ( x + h, y + k ) in powers of h and k and also find

x
R2 .
89 www.AgriMoon.Com
3. Expand x y + sin y + e in powers of ( x − 1) and ( y − π ) through quadratic terms

2 x
and write the remainder.
4. Expand x 3 − 2xy 2 in Taylor’s Theorem about a = 1 , b = −1 .
5. Show that for 0 < θ < 1 ,
1
e ax sin by = by + abxy + [(a 3 x 3 − 3ab 2 xy 2 ) sin (bθy ) + (3a 2bx 2 y − b 3 y 3 ) cos(bθy )]e aθx .
6
Keywords: Taylor’s polynomial
References


CBS, India.
Suggested Readings

90 www.AgriMoon.Com
Lesson 10
Maximum and Minimum of function of two variables
10.1 Introduction
We say that a function has a maximum (local) at a point if
for all points sufficiently close to the point .
A function of two variables has a absolute maximum (global maximum) at a

point if for all points on the domain of the
function.
Analogously we say that a function has a minimum (local) at a
point if
for all points sufficiently close to the point . Similarly we define
absolute minimum (global minimum).
91 www.AgriMoon.Com
Maximum and Minimum of Function of Two Variables
The maximum and minimum of a function are called extrema of the function;
we say that a function has an extremum of a given point if it has a maximum or
minimum at the given points.
Example 10.1. The function contains a minimum
at .
Solution: As for all and i.e.,
Example 10.2 The function
Solution:
For , , . Now for , .
So , . i.e., , is a maximum point
of .
Necessary Conditions for an Extremun: If a function attains an
extremum at and , then each first partial derivative
either vanishes for these values or does not exist.
92 www.AgriMoon.Com
This result is not sufficient for investigating the extreme points, but permits
finding these values for cases in which we are sure of the existence of a
maximum or minimum. Otherwise more investigation is required.
Example 10.3. Consider the function
Solution:
The function has partial derivatives as , which vanish at
and . But this function has neither maximum nor minimum at
and , since it takes both negative and positive values. Points at
which (or does not exist) (or does not exist) are called critical
points of the function . Thus if a function has an extreme point this
can occur at the critical point. Converse may not true.
For investigation of a function at critical points, let us establish sufficient

conditions for the maximum of a function of two variables, which can be
generalized to functions of more than two variables also.
Theorem 10.1: Let a function have continuous second partial derivatives on
an open region containing a point for which and .
Let
93 www.AgriMoon.Com
or
1. If and , then has a local minimum at .
2. If and , then has a local maximum at .
3. If , then has neither a local minimum nor a local maximum at .
4. The test is inconclusive if . (Additional investigation is required)
Proof follows from Taylor's theorem.
Note that if , then and must have same sign. This
means that can be replaced by .
Example 10.4 Find the extreme point of
94 www.AgriMoon.Com
Solution:
solving we obtain . i.e., or . So and
are the critical points. , , .
i.e., has neither minimum nor maximum at critical point . Hence is
a saddle point. We will consider the critical point
and , we conclude that has a maximum at
References
95 www.AgriMoon.Com

CBS, India.
Suggested Readings

96 www.AgriMoon.Com
Lesson 11
Lagrange's Multiplier Rule / Constrained Optimization
11.1 Introduction
We presents an introduction to optimization problems that involve finding a

maximum or a minimum value of an objective function subject to a
constraint of the form .
Maximum and Minimum. Finding optimum values of the function
without a constraint is a well known problem in calculus. One would normally use
the gradient to find critical points (gradient ( ) vanishes). Then check all
stationary and boundary points to find optimum values.
Example 1.
has a critical/ stationary point at (0,0).
97 www.AgriMoon.Com
The Hessian: A common method of determining whether or not a function has an

extreme value at a stationary point is to evaluate the hessian of the function of
variables at that point. where the hessian is defined as
A square matrix of order n × n is said to be positive definite if its leading principal

minors are all positive.
For =2, we have
Second Derivative Test: The Second derivative test determines the optimality of
stationary point according to the following rules:
∂2 f ∂2 f ∂2 f ∂f ∂f
Let= A,= B= , 2 C , and = = 0 at the point ( x, y ) , then
∂x 2
∂x∂y ∂y ∂x ∂y
98 www.AgriMoon.Com
1. If A > 0 and AC − B 2 > 0 at the point

( x, y ) , then f has a local minimum at ( x, y ) .
2. If A < 0 and AC − B 2 > 0 at the point

( x, y ) , then f has a local maximum at ( x, y ) .
3. If AC − B 2 < 0 at ( x, y ) , then ( x, y ) is a
saddle point of f .
4. If AC − B 2 =
0 , further investigation is
required.
In the above Example 1,
Therefore has a minimum at (0,0) as and determinant of the matrix
is .
11.1.1 Constrained Maximum and Minimum
When finding the extreme values of subject to a constraint , the
stationary points found above will not work. This new problem can be thought of
as finding extreme values of when the point is restricted to lie on the
99 www.AgriMoon.Com
surface . The value of is maximized (minimized) when the
surfaces touch each other,i.e , they have a common tangent for line.
This means that the surfaces, gradient vectors at that point are parallel, hence,
The number in the equation is known as the Lagrange multiplier.
11.2 Lagrange multiplier method
The Lagrange multiplier methods solves the constrained optimization problem by

transforming it into a non-constrained optimization problem of the form:
))
or . Then finding the gradient and Hessian as was done above will
determine any optimum values of .
Suppose we want to find optimum values for the following:
Example 11.2: subject to .
Then the Lagrangian method will result in a non-constrained function.
100 www.AgriMoon.Com
. The gradient for this new function is
Solving , we obtain , and .
The Hessian matrix at the stationary point
Since the solution , minimizes
subject to with
Example 11.3: Find the rectangle of parameter l which has maximum area i.e.,
Maximize subject to
Solution:
i.e., i.e., .
, so that the rectangule of maximum area is a square.
Example 11.4 Find the shortest distance from the point (1,0) to the parabola
, i.e., Minimize subject to .
Now
If then , from
Hence
Hence
i.e.,
Hence the only solution is and the required distance is unity.
Questions: Answer the following question
1. Determine the maximum value of the -th root of a product of numbers
provided that their sum is equal to a given number . Thus the
problem is stated as follows: it is required to find the maximum of the function

subject to , , for all .
References

CBS, India.
Suggested Readings

Lesson 12
Convexity, Concavity and Points of Inflexion
12.1 Introduction
In the plane, we consider a curve , which is the graph of a single-
valued differentiable function .
Definition 12.1: We say that the curve is convex downward bending up on the
interval if all points of the curve lie above the tangent at any point on the
interval. Or when the curve turns anti-clock wise we call it is convex downward
(concave upward) (see Fig. 1).
Fig.1. (Convex downward/Bending up)
Definition: We say that a curve is convex upwards for bending down on the
interval if all points of the curve lie below the tangent at any point on the
interval. Or when the curve turns clock-wise we say it is convex upward

(concave downward) (see Fig. 2).
Fig. 2. (Convex upward / Bending down )
The curve has a point of inflexion at , at which the curve changes from convex
upwards to convex downwards and vice-versa.
Theorem 1: If for all points of an interval , , the curve
on this interval is convex upward. If , the curve is convex
downward.
If is convex upward on .
If is convex doward on .
P •
Fig. 3. (Inflexion point)
Example 12.1: Find the ranges of values of for which the curve
is convex downwards, convex upwards, and
also determine the point of inflection.
Solution:
Now on the interval , , , hence . If ,
, i.e., and . Hence for , . Now on
the interval , . Hence the curve is convex downward on the interval
and . Convex upwards on . The curve has inflection points
at and as changes sign. At , and at , .
i.e., and are two points of inflection of the curve.
Example 12.2: Determine the intervals where the graph of the function is
convex downward and convex upward of
Solution:
Hence,
Then . For , , , the graph is
convex downward. For , , , the graph is convex
upward. There is no inflection point, since is not defined when .
Example 12.3: Determine the intervals where the graph of the function is
convex downward and convex upward of ,
Solution:
, and . So, for
and for , , , the graph is convex
downward. For , , , and the graph is convex upward.
There is an inflection point at . There is no inflection point at , the
graph is convex downward for .
Example 12.4: Find the point of inflection of the curve ,
Solution:
, . if , or . i.e.,
or . Now changes sign from negative to positive as passes
through 1 and changes sign from positive to negative as passes through .
Thus and are two points of inflection of the given curve.
Example 12.5: What conditions must the coefficients satisfy for the
curve to have points of inflection?
Solution:
has a point of inflection iff the equation
has different real roots. i.e., discriminant
is positive. i.e. .
1. Determine all the inflexion points of
2. Determine all the inflexion points of
3. Determine all the inflexion points of for
4. Sketch the curve . Determine the inflexion points. Compare with
graph of .
5. Determine the inflexion points and the intervals of convex downward /

bending up and convex upward / bending down for the following curve
6.
7.
8.
9. Sketch the curve
10.Pint of inflexion of .
Keywords: Convex up, Convex down, Inflexion Point.
References

CBS, India.
Suggested Readings

Lesson 13
Curvature
13.1 Introduction
Curvature measures the extent to which a curve is not contained in a straight line.
It curvature measures how curved the curve is. We have heard the comparison of
bending or curvature of a road at two of its points. The curvature of a straight line
is zero. It also measures how fast the tangent vector turns as a point moves along
the curve.
Fig.1.
Let be a fixed point on the curve. Let arc , and arc , so that
arc . Let , be the angles which the tangents at and make
with some fixed line (say - axis). denotes the angle formed by these tangents.
The symbol also denotes the angle through which the tangent turns from and
through a distance . will be large or small, as compared with ,
depending the degree of the sharpness of the bend. This suggests the following
definitions:
Curvature
The curvature of the curve at is defined as .
The reciprocal of curvature is the radius of curvature.
Length of Arc as a Function, Derivative of Arc.
Let be the equation of a given curve on which we take a fixed point .

Let and be the variable points on the curve with arc
and arc so that arc .
∆y
A
N
S
P ∆x
Fig. 2.
, taking limit both sides we have
Curvature
Radius of Curvature: Cartesian Equations
We define the absolute value of as the curvature and denote it by .
Consider the curve , we note that and, therefore,
Differentiating this with respect to , we have
As , we have
Hence , where ,
Note: If , the radius of curvature, , is positive or negative according as
is +ve or -ve i.e., accordingly as the curve is convex downward or convex upward.
But we consider is +ve here. Curvature is zero at point of inflection. Since is
independent of the choice of -axis and -axis, interchanging and , we see that
, is given by
Curvature
Curvature- parametric Equation
Given , . .
Hence the curvature
Curvature- polar Equation
Let be the given curve in polar co-ordinates. Now its cartesian

coordinates are of the form , . i.e., ,
. Now
and
substituting the latter expressions in the previous parametric-form, we have
We know
Curvature
numerator becomes
To check we can observe that
denominator becomes
Hence
Curvature
The radius of curvature is
Example 1: Determine the radius of curvature of the curve
Solution:
Hence
* We know
numerator becomes
Curvature
denominator becomes
Hence
Example 2: Find the radius of curvature of
Ans.:
Example : Find the radius of curvature of , for ,
Ans.:
Example : Find the curvature of the hyperbola at .

Solution:
Curvature
and . So
When , .
Example 3: For what value of is the radius of curvature of smallest?
Solution:
, and radius of curvature is . Then
etting , we find , , . As the second
derivative at this point is positive, is the point which gives the smallest
radius of curvature.
Example 4: Find the radius of curvature at any point on the curves:
Solution:
Curvature
implies
i.e. .
Example : Find the radius of curvature at the origin of the curve
Solution:
which implies .
Example 5: Find the curvature of the cycloid , at

an arbitrary point .
Curvature
Solution:
, , , . Using this
parametric formula , we obtain
When ,

1. Find the curvature of the curve at the point (a,b) and (a,0)
2. Find the curvature of the curve at the point (2,0)
3. Find the curvatur e of the curve at the point (3,4)
Questions: Find the radius of curvature of the following curves at the

indicated points.
4. at the point (4,8)
5. at the point
6. at the point (1,0)
Curvature
7. at the point
8. Find the point of the curve at which the radius of curvature is minimum.
Ans.: 1. , 2. , 3. , 4. , 5. 29, 6. , 7. 1 & 8.
References

Jain, R. K. and Iyengar, SRK. (2010), Advanced Engineering Mathematics, 3 rd

CBS, India.
Suggested Readings

Lesson 14
Asymptotes
14.1 Introduction
A straight line is called an asymptote to a curve (fig.1), if the distance
distance from a point of to approaches to zero as recedes to infinity.
Roughly speaking, a straight line is said to be an asymptote of a curve if it comes

arbitrary close to that curve (but never touches the curve).
14.1.1 Asymptotes of Functions: If the graph of a function has an asymptote ,
then we say that the function has an asymptote . A function can have more than
one asymptote. If an asymptote is parallel with the -axis, we call it a vertical
asymptote. If an asymptote is parallel with the -axis,
we call it a horizontal asymptote. All other asymptotes are oblique asymptotes.
Asymptotes
M P ( x, y )
d δ
Fig. 1
Vertical Asymptotes
A straight line is a vertical asymptote to the the curve if
or . Consequently, to find vertical
asymptotes one has to find values of such that when they are approached by
the function , the latter approaches infinity. Then the straight line is a
vertical asymptote.
Example 14.1: The curve has a vertical asymptote , since as
Asymptotes
Example 14.2: The curve has infinite number of vertical asymptotes at
for as when .
Example 14.3: The curve has no vertical asymptote at as
14.2 Horizontal Asymptotes
A line is a horizontal asymptote of a function iff or
, with .
Examples 14.4: The curve has horizontal asymptote as
. So, is a horizontal asymptote of the function .
14.3 Oblique Asymptotes/Inclined Asymptotes
Let the curve have an inclined or oblique asymptote (fig.1) whose
equation is .
Asymptotes
Here and are unknown real numbers to be determined. Let be the
perpendicular distance of any point on the curve to the line .
Hence, . Now as . Hence, . i.e.,
, hence
So .
Example 14.5: Find the asymptotes to the curve
Solution:
When , , and , , hence the straight line is a
vertical asymptote of the above curve.
Asymptotes
Next to find the asymptotes of the form , i.e., the inclined asymptote.
Hence is an inclined asymptotes to the given curve.
Example 14.6: Find the oblique asymptotes to the curve
Solution:
14.3.1 Tutorial Discussion
• An asymptote is a straight line which acts as a boundary for the graph of a

function.
Asymptotes
• When a function has an asymptote (and not all functions have them) the
function gets closer and closer to the asymptote as the input value to the
function approaches either a specific value a or positive or negative infinity.
• The functions most likely to have asymptotes are rational functions
• Vertical asymptotes occur when the following condition is met:
The denominator of the simplified rational function is equal to 0.
Remember, the simplified rational function has cancelled any factors common to
both the numerator and denominator.
e.g., Given the function
The first step is to cancel any factors common to both numerator and denominator.
In this case there are none.
The second step is to see where the denominator of the simplified function equals
0. implies .
The vertical line is the only vertical asymptote for the function. As the
input value to this function gets closer and closer to -1 the function itself looks
and acts more and more like the vertical line .
Asymptotes
Example 14.7
First simplify the function. Factor both numerator and denominator and cancel any
common factors.
The asymptote(s) occur where the simplified denominator equals 0. i.e., .
The vertical line is the only vertical asymptote for this function. As the input
value to this function gets closer and closer to 3 the function itself looks more
and more like the vertical line .
Example 14.8 If
Factor both the numerator and denominator and cancel any common factors.
In this case there are no common factors to cancel.
Asymptotes
The denominator equals zero whenever either or . Hence this
function has two vertical asymptotes, one at and the other at .
5. Horizontal Asymptotes
Horizontal asymptotes occur when either one of the following conditions is met
(you should notice that both conditions cannot be true for the same function).
• The degree of the numerator is less than the degree of the denominator. In this
case the asymptote is the horizontal line .
• The degree of the numerator is equal to the degree of the denominator. In this
case the asymptote is the horizontal line where is the leading
coefficient in the numerator and is the leading coefficient in the denominator.
When the degree of the numerator is greater than the degree of the denominator
there is no horizontal asymptote.
Example 14.9
then there is a horizontal asymptote at the line because the degree of the
numerator 2 is less than the degree of the denominator 3.
Asymptotes
This means that as gets larger and larger in both the positive and negative
directions ( ) and ( ) the function itself looks more and more like the
horizontal line
Find the vertical asymptotes, horizontal asymptotes and inclined asymptotes for
each of the following functions Problems:
Exercises:
Find the asymptotes of the following curves:
1.
Solution: Vertical: Horizontal: Inclined: none
2.
Solution: Vertical: Horizontal: none Inclined:
3. Ans.
4. Ans.
Asymptotes
5. Ans. No asymptotes
6. Ans.
7. Ans.
8. Sketch the function
Keywords: Asymptotes, horizantal, vertical and inclied asymptotes.
References

CBS, India.
Suggested Readings
Tom M. Apostol, (2003). Calculus, Volume II Second Editions, Publishers,John

Lesson 15
Tracing of Curves
15.1 Introduction
Now we use some mathematical techniques to trace curves and graphs of

functions much more efficiently. We shall especially look for the following
aspects of the curve.
1. Intersection with the coordinate axes.
2. Critical points
3. Regions of increase
4. Regions of decrease
5. Maxima and minima (including local ones)
6. Behaviour as x becomes large positive and large negative.
7. Values of x near which y becomes large positive or large negative.
8. Regions where the curve is convex up or down.
9. Asymptotes of the curve
10. Find whether the curve is symmetric
15.2 Behaviour as x becomes very Large
Suppose we have a function f defined for all sufficiently larger numbers. Then
we get substantial information concerning our function by investigating how it
behaves as x becomes large.
For example, sin x oscillates between -1 and +1 no matter how large x is.
Tracing of Curves
However, polynomials do not oscillate. When f(x) = x2 as x becomes large

positive. So does x2. Similarly with the function x3, or x4 (etc.). We consider this
systematically.
Example 15.1 Consider a parabola,
y = ax2 + bx + c, with a ≠ 0.
There are two essential cases, when a > 0 or a < 0. We have the parabola which
looks like in the figure
y = ax2 + bx + c y = ax2 + bx + c
a>0 a<0
We look some numerical examples.
Example 15.2 Sketch the graph of the curve
y = f(x) - 3x2 + 5x – 1
We recognize this as a parabola.
5 1
f ( x)= x 2 (−3 + − ),
x x2
when is large positive or negative, then x2 is large positive and the factor on
the right is close to -3. Hence f(x) is large negative. This means that the
parabola has the shape as shown in figure.
Tracing of Curves
5
We have f ′( x) =−6 x + 5 . Thus f ′( x) = 0 iff x = . There is exactly one
6
2
5  5  25
critical point. We have f   =−3   + −1 > 0
6 6 6
The critical point is a maximum, because we have already seen that the parabola
bends down.
The curve crosses the x-axis exactly when
−3 x 2 + 5 x − 1 =0
−5 ± 25 − 12 5 ± 13
=x =
−6 6
Hence the graph of the parabola looks as on the figure.
5− 13 5 5+ 13
6 6 6
Bending down or convex upward
The same principle applies to sketching any parabola.
(i) Looking at what happens when x becomes large positive or negative tells us
whether the parabola bends up or down.
Tracing of Curves
(ii) A quadratic function
f(x) = ax2 + bx + c with a ≠ 0
has only one critical point, when
f ′( x)= 2ax + b= 0
−b
So x =
2a
Knowing whether the parabola bends up or down tells us whether the critical
−b
point is maximum or minimum, and the value x = tells us exactly where this
2a
critical point lies.
(iii) The points where the parabola crosses the x-axis are determined by the
quadratic formula.
Example 15.3. (Cubics) Consider a polynomial
f(x) = x3 + 2x – 1, find f(x) when x → ±∞ . We have
We can write it in the form
 2 1
x3 1 + 2 − 3  and, when x → + ∞ means f ( x) → + ∞
 x x 
Example 15.4. (a) Consider the quotient polynomials like
x3 + 2 x − 1
Q( x) = 3
2x − x + 1
1
Here if x → ± ∞ , then Q( x) → .
2
Tracing of Curves
x3 − 1
Example 15.4(b) Consider the quotient Q( x) = 2
x +5
Here lim Q ( x) = + ∞ and lim Q ( x) = − ∞

x →+∞ x →−∞
The meaning of the above limit is that there is no number which is the limit of
Q(x) as x → + ∞ or x → − ∞ .
We can now sketch the graphs of cubic polynomials symmetrically.
Example 15.5 Sketch the graph of f(x) = x3 – 2x + 1
1. If x → + ∞ then f ( x) → + ∞
If x → − ∞ then f ( x) → − ∞
2. We have f ′(=
x) 3x 2 − 2
f ′( x) =0⇔ x=± 2
3
The critical points of f are x = + 2 and x = − 2 .

3 3
3. Let g (=
x) f ′(=
x) 3 x 2 − 2 . Then the graph of g is a parabola which is given
as
− 2 2
3 3
Graph of g ( x) = f ′( x)
Tracing of Curves
Therefore, f ′( x) > 0 ⇔ x > 2 and x < − 2 , where g(x) > 0 and f is

3 3
strictly increasing on the intervals x ≥ 2 and x ≤ − 2 .

3 3
Similarly f ′( x) < 0 ⇔ − 2 < x < 2 where g(x) < 0, and f is strictly

3 3
decreasing on this interval . Therefore − 2 is a local maximum for f, and

3
2 is a local maximum.
3
4. f ′′( x) = 6 x , and f ′′( x) > 0 iff x > 0 and f ′′( x) < 0 iff for x > 0, therefore f
is bending up ( convex downward ) for x > 0 and bending down ( convex
upward ) for x < 0. There is an inflection point at x = 0.
Putting all this together, we find that the graph of f looks like this
graph of f(x) = x3 – 2x + 1
2
− 3
Example 15.6 Sketch the graph of the curve.
y = - x3 + 3x - 5
1. When x = 0 , we have y = -5. With general polynomial for degree ≥ 3 there is
in general no simple formula for those x such that f(x) = 0, so we do not give
explicitly in the intersection of the graph with the x – axis.
2. The derivative is f ′( x) =
−3 x 2 + 3
Tracing of Curves
f ′( x) =0⇔ x=±1
The graph of f ′( x) is given by
f ′′( x) = − 6 x , f ′′(1) = − 6 , f ′′(−1) =

6 , f ′′( x) > 0 iff x < 0 and f ′′( x) < 0 iff
x > 0. x = 0 is an inflection point x = 0.
f is strictly decreasing ⇔ f ′( x) < 0
⇔ x < −1 and x > 1
f is strictly increasing ⇔ f ′( x) > 0
⇔ −1 < x < 1 .
Therefore f has a local minimum at x = -1 and local maximum at x = 1.
Putting all this information together, we see that graph of f looks like this
-1 1
graph of f(x) = - x3 + 3x + 5
Example 15.7 Let f(x) = 4x3+2. Sketch the graph of f.
Tracing of Curves
Solution:
Here we have f ′( x=
) 12 x 2 > 0 ∀ x ≠ 0 . There is only one critical point, when
x = 0. Hence the function is strictly increasing for all x, and its graph looks like
f ′′( x) =
24 x > 0 for all x > 0
f ′′( x) < 0 for x < 0
Convex downward
convexup
Example 15.8 Sketch the graph of f(x) = 4x3 + 4x .
Solution:
f ′( x)= 3 x 2 + 4 > 0 ∀ x
f ′′( x) =
6 x > 0 for x > 0
f ′′( x) < 0 for x < 0
Tracing of Curves
So the graph looks like
Convex downward
Convex upward
In both the above examples x = 0 is an inflection point.
15.3 Rational Functions
We shall now consider quotient of polynomials.
Example 15.9 Sketch the graph of the curve
x −1
=y f=
( x)
x +1
1. When x = 0, we have f(x) = 1. When x = 1, f(x) = 0.
2
2. The derivative is f ′( x) =
( x + 1)
2
It is never zero, so the function has no critical points.
3. The denominator is a square and hence is always positive, whenever it is

defined, i.e., for x ≠ −1 . Thus f ′( x) > 0 for x ≠ −1 . The function is not
defined at x = -1 and hence derivative also is not defined at x = -1, i.e., f(x) is
increasing in the region x < -1 and is increasing in the region x > -1
Tracing of Curves
f(x)
x=0
f(x)=1
x=1
f(x)=0
4. There is no region of decreasing.
5. Since the derivative is never zero, there is no relative maximum or minimum.
−4
6. The second derivative is f ′′( x) = .
( x + 1)
3
There is no inflection point since f ′′( x) ≠ 0 for all x where the function is
defined. If x < -1, (x + 1)3 < 0, and f ′′( x) > 0 , f(x) is bending up or convex
downward. If x > -1, then x+1 > 0 ⇒ (x+1)3 > 0. So f ′′( x) < 0 i.e., f(x) is
bending down (convex upward).
 1
x 1 − 
x −1
( x) = lim  =
x
7. As x → ∞ , f(x) → 1 f= 1
x + 1 x→∞  1 
x 1 + 
 x
when x → −∞ , f(x) → 1
Tracing of Curves
8. As x → −1 , the denominator approaches 0 and the numerator approaches -2.

If x approaches -1 from the right so x > -1, then the denominator is +ve and the
x −1
numerator is negative. Hence the function is negative , and is large
x +1
negative. Putting all these information we get the graph looks like the given
figure.
EXERCISES
Sketch the following curves, indicating all the information stated in the
examples etc.
x2 + 2
1. y =
x −3
x −3
2. y =
x2 + 1
3. =
y x4 + 4 x
4. =
y x8 + x
5. f ( x) = x 4 + 3 x 3 − x 2 + 5
x2 − 1
6. y =
x
7. Show that a curve y = ax 3 + bx 2 + cx + d with a ≠ 0 has exactly one

inflection point.
Keywords: Curve tracing, increasing, decreasing, convex up, convex down.
References
Tracing of Curves

CBS, India.
Suggested Readings

Module 2: Integral calculus
Lesson 16
Improper Integral
16.1 Introduction
Integral with infinite limits. Let a function f(x) be defined, positive and continuous
for all values of x such that a ≤ x < ∞ . Consider the integral
b
I (b) = ∫ f ( x) dx
a
a b x
O
Fig. 1
This integral is meaningful for b > a. This integral varies with b and is continuous
function of b. Let us consider the behavior of this integral when b → +∞ (Fig. 1).
Definition 16.1 if there exists a finite limit
b
lim ∫ f ( x) dx
b→∞ a
Improper Integral
Then this limit is called the improper integral of the function f(x) on the interval
[ a, +∞ ] and is denoted by the symbol
+∞
∫a
f ( x) dx
Thus, by definition, we have
+∞ b
∫ a
f ( x) dx = lim ∫ f ( x) dx
b→∞ a
b
In this case it is said that the improper integral exists or converges. If ∫a
f ( x) dx
+∞
as b → +∞ does not have a finite limit, one say that ∫a
f ( x) dx does not exist or
diverges.
If f ( x) ≥ 0 , the geometrical meaning of the improper integral can be seen as if the

b
integral ∫a
f ( x) dx expresses the area of region bounded by the curve y = f(x), the
x – axis and the ordinates x = a, x = b, it is natural to consider that the improper

+∞
integral ∫ a
f ( x) dx expresses the area of an unbounded ( infinite ) region lying
between the curve y = f(x), x = a and x-axis.
We similarly define the improper integrals of other infinite intervals:
a a
∫−∞
f ( x) dx = lim
α →−∞ α∫ f ( x) dx
Improper Integral
+∞ c c
∫=
f ( x) dx
−∞ ∫−∞
f ( x) dx + ∫
−∞
f ( x) dx
The latter equation should be understood as if each of the improper integrals on the
right exists, then, by definition, the integral on the left also exists (converges).
+∞ dx
Example 16.1: Evaluate the integral ∫ 0 1 + x2
Solution:
By the definition of improper integral we find
+∞
dx b dx b
π
∫0 1 + x 2 b→∞ ∫0 1 + x 2 b→∞
−1
= =
lim =
lim tan x
0 2
Note that this integral expresses the area of an infinite curvilinear trapezoid crosses
x –axis as x → ∞ .
+∞ dx
Example 16.2: Evaluate ∫−∞ 1 + x 2
Solution:
+∞ dx dx
0 +∞ dx
∫−∞ 1 + x 2
= ∫−∞ 1 + x 2 ∫0 1 + x 2
+
π
The 2nd integral is equal to (see example 1)
2
Improper Integral
Compute the First Integral:
π
lim tan x= lim ( tan −1 0 − tan −1 =
b)
0
0dx 0 dx
∫−∞ 1 + = lim ∫ −1
=
x 2 b→−∞ b 1 + x 2 b→−∞ b b→−∞ 2
+∞dx π π
Hence, ∫−∞ 1 + x 2
= + =π
2 2
In many cases it is sufficient to determine whether the given integral converges or

diverges, and to estimate its value. The following theorems, which we give without
proof, may useful in this respect.
Theorem 16.1: Let f and g be continuous function on the interval [a, ∞) with
o ≤ f ( x) ≤ g ( x) ∀ a ≤ x < ∞ .
+∞ +∞
If ∫a
g ( x) dx converges then ∫
a
f ( x) dx also converges, and
+∞ +∞
∫a
f ( x) dx ≤ ∫
a
g ( x) dx
Theorem 16.1: The integral of a discontinuous function:

c
The integral ∫
a
f ( x) dx of the function f(x) discontinuous at a point c is defined as
follows:
Improper Integral
c c −ε
∫
a
f ( x) dx = lim ∫
ε →0 + a
f ( x) dx
If the limit on the right exists, the integral is called an improper convergent
integral, otherwise it is divergent. If the function f(x) is discontinuous at x = a of
the interval [a,c] then by definition ,
c c
∫ a
f ( x) dx = lim ∫
ε →0 a +ε
f ( x) dx
If the function f(x) is discontinuous at some point x = x0 inside the interval [a,c] ,
we put
c x0 c
∫=
a
f ( x) dx ∫
a
f ( x) dx + ∫ f ( x) dx
x0
If both the improper integrals on the right hand side of the equation exist.
1 dx
Example 16.3 Evaluate ∫0
1− x
1 dx 1−ε dx
Solution: ∫0
= lim ∫
1 − x ε →0 0 1 − x
dx
1−ε
(1 − x ) 2 dx
−1
= lim ∫
ε →0 0
1−ε
(1 − x )
1
2
= −
− 12 + 1
0
Improper Integral
1−ε
=
−2 1 − x
0
=
lim − 2
ε →0
( ε −1 = 2 )
1dx
Example 16.4: Evaluate the integral ∫−1 x 2 .
Solution:
Since inside the interval of integration there exist a point x = 0, at which the
integrand is not continuous, we express the integration as:
1 dx 0−ε dx 1 dx
∫−1 x 2 ε →0 ∫−1 x 2 ε →0 ∫0+ε x 2
= lim + lim
−ε dx 1 dx
= lim ∫
x 2 ε →0 ∫ε x 2
+ lim
ε →0 −1
−ε 1
1 1
=lim − − lim
ε →0 x ε →0 x
−1 ε
 1   1
=− lim  − + 1 − lim 1 − 
ε →0
 ε  ε →0  ε 
 1   1
But − lim  − + 1 =∞ and − lim 1 −  =
∞ i.e., the integral diverges on [-1,0]
ε →0
 ε  ε →0
 ε
as well as on [0,1] .
Hence the given integral diverges on the entire interval [-1, 1].
Improper Integral
It should be noted that if we had evaluated the given integral without paying
attention to the discontinuity of the integrand at point x = 0, the result would have
dx −1
1
1 1 
1
been wrong as ∫ 2 = =
− −  = −2
−1 x x −1  1 −1 
For determining the convergence of improper integrals of discontinuous functions

and for estimating their values, one can refer Lesson 17. These integrals are
discussed in details in Lesson 17 also.
c dx c dx
∫a (c − x ) p
, also ∫
a ( x − a) p
c dx
It is easy to verify that ∫a (c − x ) p
converges for p < 1 and diverges for p ≥ 1.
Same applies also to 2nd integral.
EXERCISES
Evaluate the following improper integrals:
1 dx
1. ∫
0
1 − x2
∞
2. ∫
0
e − x dx
∞ dx
3. ∫
0 a + x2
2
Improper Integral
1 dx
4. ∫
0
1 − x2
1
5. ∫ ln x dx
0
π π
Ans.: 1. 1, 2. 1, 3. (a > 0) , 4. & 5. 1
2a 2
Keywords: Improper Integrals, Positive Function, Area Of the Region.
References

CBS, India.
Suggested Readings
Tom M. Apostol (2003). Calculus, Volume II Second Editions, Publishers, John

Lesson 17
Tests for Convergence
17.1 Introduction
In this Lesson the convergence of Improper Integrals is studied.
Definition 16.1 if there exists a finite limit
b
lim ∫ f ( x) dx
b→∞ a
Then this limit is called the value of the improper integral of the function f(x)
on the interval [ a, +∞ ] and is denoted by the symbol
+∞
∫a
f ( x) dx
Thus, by definition, we have
+∞ b
∫a
f ( x) dx = lim ∫ f ( x) dx
b→∞ a
In this case it is said that the improper integral exists or converges. If

b +∞
∫a
f ( x) dx as b → +∞ does not have a finite limit, one say that ∫a
f ( x) dx
does not exist or diverges.
If f ( x) ≥ 0 , the geometrical meaning of the improper integral can be seen as if

b
the integral ∫a
f ( x) dx expresses the area of region bounded by the curve y =
f(x), the x – axis and the ordinates x = a, x = b, it is natural to consider that the
+∞
improper integral ∫a
f ( x) dx expresses the area of an unbounded ( infinite )
region lying between the curve y = f(x), x = a and x-axis.
We similarly define the improper integrals of other infinite intervals:
a a
∫−∞
f ( x) dx = lim ∫
α →−∞ α
f ( x) dx
+∞ c c
∫=
f ( x) dx
−∞ ∫−∞
f ( x) dx + ∫
−∞
f ( x) dx
The latter equation should be understood as if each of the improper integrals on

the right exists, then, by definition, the integral on the left also exists
(converges).
+∞ dx
Example 16.1 Find out at which p the integral ∫1 xp
converges and at which
it diverges.
a b
Solution:
Since (when p ≠ 1)
b
dx 1 1− p 1
( b1− p − 1)
b
∫1 x p 1 − p
= = x
1− p
1
We have
∫1 x p b→+∞ 1 − p ( b − 1)
+∞ dx 1 1− p
= lim
Consequently, with respect to like this integral we conclude that if p > 1, then
+∞ dx 1
∫1
=
x p p −1
, and the integral converges.
+∞ dx
If p < 1, then ∫ 1 xp
= ∞ and integral diverges.
+∞ dx +∞
When p = 1, ∫1 x p
= ln x 1 = ∞ , and the integral diverges.
+∞ dx
Note: We call the p-integral ∫1 xp
converges for p > 1, and diverges for p ≤ 1
which is in the comparison test of improper integral used.
In many cases it is sufficient to determine whether the given integral converges

or diverges, and to estimate its value. The following theorems, which we give
without proof, may useful in this respect.
Theorem 17.1. Let f and g be continuous function on the interval [a, ∞) with
o ≤ f ( x) ≤ g ( x) ∀ a ≤ x < ∞ .
+∞ +∞
If ∫ a
g ( x) dx converges then ∫ a
f ( x) dx also converges, and
+∞ +∞
∫a
f ( x) dx ≤ ∫
a
g ( x) dx
+∞ dx
Example 17.2 Investigate the integral ∫
1 x 2 (1 + e x )
for convergence.
Solution:
It will be noted that when 1 ≤ x
1 1
< 2
x (1 + e ) x
2 x
+∞
+∞ dx 1
And ∫1 x2
=
−
x1
=
1
+∞ dx
Consequently, ∫ 1 x 2 (1 + e x )
converges, and its value is less than 1. Hence
+∞ dx
∫1 x 2 (1 + e x )
converges.
+∞
Theorem 17.2. If for all x( x ≥ a ),0 ≤ g ( x) ≤ f ( x) holds true and ∫a
g ( x) dx
+∞
diverges, then the integral ∫
a
f ( x) dx also diverges.
Example 17.3 Find out whether the following integral converges or diverges.
+∞ x +1
∫1
x 3
dx
Solution:
x +1 x 1
We note that > =
x3 x3 x
+∞ dx 1
But ∫
1
x
1
2
= ∞ as p=
2
< 2 . Hence the given integral is divergent.
In the above two theorems we considered improper integrals of nonnegative

functions. For the case of a function f(x) which changes its sign over an infinite
interval we have the following result.
+∞
Theorem17.3. If the integral ∫a
f ( x) dx converges, then the integral
+∞
∫a
f ( x)dx also converges.
In this case, the later integral is called an absolutely convergent integral.
+∞
Definition 17.1: An integral ∫
a
f ( x)dx converges conditionally if and only if
+∞ +∞
∫a
f ( x)dx converges but ∫ a
f ( x) dx is not convergent.
+∞ sin x
Example 17.3 Investigate the convergence of the integral ∫
1 x3
dx .
Solution:
sin x 1 +∞ 1
Here,
x3
≤
x3
. But ∫
1 x3
dx convergent as p = 3.
∞ sin x
Therefore, the integral ∫
1 x3
dx also converges.
17.2 The Integral of a Discontinuous Function
A function f(x) is defined and continuous when a ≤ x < c , and either not defined
or discontinuous when x = c. In this case, one cannot speak of the integral
c
∫a
f ( x) dx as limit of integral sums, because f(x) is not continuous on [a, c] and
for this reason the limit may not exist.
c
The integral ∫a
f ( x) dx of the function f(x) discontinuous at a point c is defined
as follows:
c c −ε
∫
a
f ( x) dx = lim ∫
ε →0 a
f ( x) dx
If the limit on the right exists, the integral is called an improper convergent
integral, otherwise it is divergent. If the function f(x) is discontinuous at x = a of
the interval [a, c] then by definition,
c c
∫ a
f ( x) dx = lim ∫
ε →0 a +ε
f ( x) dx
If the function f(x) is discontinuous at some point x = x0 inside the integral [a, c]
, we put
c x0 c
∫=
f ( x) dx
a ∫a
f ( x) dx + ∫ f ( x) dx
x0
If both the improper integrals on the right hand side of the equation exist.
1 dx
Example 17.4 Test the convergence of the integral ∫−1 x 2 .
Solution:
Since inside the interval of integration there exist a point x = 0, at which the
integrand is not continuous, we express the integration as:
1 dx 0−ε dx 1 dx
∫−1 x 2 ε →0 ∫−1 x 2 ε →0 ∫0+ε x 2
= lim + lim
−ε dx 1 dx
= lim ∫
x 2 ε →0 ∫ε x 2
+ lim
ε →0 −1
−ε 1
1 1
=lim − − lim
ε →0 x ε →0 x
−1 ε
 1   1
=− lim  − + 1 − lim 1 − 
ε →0
 ε  ε →0  ε 
 1   1
But − lim  − + 1 =∞ and − lim 1 −  =
∞ i.e., the integral diverges on [-
ε →0
 ε  ε →0
 ε
1, 0] as well as on [0,1] .
Hence the given integral diverges on the entire interval [-1, 1].
It should be noted that if we had evaluated the given integral without paying
attention to the discontinuity of the integrand at point x = 0, the result would
dx −1
1
1 1 1 
have been wrong as ∫ 2 = =
− −  = −2
−1 x x −1  1 −1 
This is impossible (Fig. 3)
x=0
1
y= 2
x
(-1, 1) (1, 1)
y=0
-1 -1
Fig. 3
Note: If the function f(x), defined on the interval [a, b], and has finite number
of discontinuity points a1, a2,…., an within the interval,
b a1 a2 b
Then ∫a
f (=
x) dx ∫
a
f ( x) dx + ∫ f ( x) dx + ..... + ∫ f ( x) dx
a1 an
If each of the improper integrals on the right side of the equation converges then
b
∫a
f ( x) dx is called convergent but if even one of these integrals diverges, then
b
∫a
f ( x) dx too is called divergent.
For determining the convergence of improper integrals of discontinuous

functions and for estimating their values, one can frequently make use of
theorems similar to those used to estimate integrals within infinite limits.
Theorem 17.3. Let f(x) and g(x) be continuous functions in [a,c] except at x = c
and at all points of this interval the inequalities g(x) ≥ f(x) hold and
c c
∫a
g ( x) dx converges, then ∫a
f ( x) dx also converges.
Theorem 17.4. Let f(x) and g(x) be continuous functions on [a,c] except at x =
c and at all points of this interval the inequalities f(x) ≥ g(x) ≥ 0 hold and
c c
∫a
g ( x) dx diverges, then ∫ a
f ( x) dx also diverges.
Theorem 17.5. Let f(x) be a continuous function on [a, c] except at x = c , and

c
the improper integral ∫a
f ( x) dx of the absolute value of this function
c
converges, then the integral ∫
a
f ( x) dx of function of itself also converges. We
frequently come across the improper integral of the following types.
c dx c dx
∫a (c − x ) p
, also ∫
a ( x − a) p
c dx
It is easy to verify that ∫ a (c − x ) p
converges for p < 1 and diverges for p ≥ 1.
Same applies also to 2nd.
1 dx
Example17.5 Does the integral ∫0
x + 4x 3
converge?
Solution:
The integrand is discontinuous at x = 0.
1 1
Now ≤
x + 4 x3 x
dx 1 1 dx
∫
1
The improper integral ∫ 0
x
1
2
as
2
< 1 exists and hence
0
x + 4 x3
also
exists.
EXERCISES
Test the convergence of the following improper integrals:

∞
1. ∫ 0
x sin x dx
∞ dx
2. ∫
1
x
∞ dx
3. ∫−∞ x 2 + 2 x + 2
1 dx
4. ∫ 0
1
x3
2 dx
5. ∫ 0 x3
6. Let b > 2. Find the area under the curve y = e −2 x between 2 and b. Does this
area approach a limit when b → ∞ . If so what limit?
∞
7. Can an improper integral ∫
a
f ( x) dx ever be transformed onto a proper
integral by a change of variable?
3
Ans.: 1. The integral diverges, 2. The integral diverges, 3. π , 4. , 5. The
2
1 1 −4 1 1 1
integral diverges, 6. − e −2b + e , yes e −4 & 7. Yes, =
f ( x) = , x .
2 2 2 x2 t
Keywords: Convergence, absolutely convergence, comparison test.
References

CBS, India.
Suggested Readings

Lesson 18
Rectification
18.1 Introduction
The method of finding the length of the arc of the curve of is called the
rectification. Let y = f ( x ) be a differentiable function defined on [a, b] with
a < b and assume that its derivative is continuous. Our aim is to determine the
length of the curve described by the graph. The main idea behind this is to
approximate the curve by small line segments and add these up.
(x1,f(x1)
(x2,f(x2) (x4,f(x4)
(x3,f(x3)
a = x0 x1 x2 x3 x4 = b
Fig .1
We consider a partition of the interval [a, b] . a = x0 ≤ x1 ≤ x2 ≤ x3....... ≤ xn = b

In figure 1 take n = 4 for simplification.
For each xi we have on the curve ( xi , f ( xi ) ) . We draw the line segments between
two successive points. The length of such a segments the length of the line
between
Rectification
( xi , f ( xi ) ) and ( xi+1, f ( xi+1) ) ( xi +1 − xi )2 + ( f ( xi +1 )− f ( xi ) ) -------------

2
is equal to
(1)
( f ( xi+1)− f ( xi ) ) =( xi+1− xi ) f ′(ci )
By mean value theorem, we conclude that
xi ) ( xi +1− xi ) f ′(ci ) , where ci ∈( xi , xi +1 )

f ( xi +1 ) − f (=
Hence (1) becomes now
( xi +1− xi )2 + ( xi +1− xi )2 f ′(ci )2
= ( xi +1− xi )2 1 + f ′(ci )2  -------------- (18.2)
Hence the form of the line segment is
n −1
∑ 1 + f ′(ci ) 2 ( xi +1 − xi ) ----------------- (18.3)

i =0
Now as f ′( x) is continuous function. So is H ( x=) 1 + f ′( x)2 . So we can write

n −1
eqn. (3) as ∑ H (c ) ( x
i i +1 − xi )
i =0
Since H ( x) is continuous on [a, b] , H (ci ) satisfies the inequalities:
min H ≤ H (ci ) ≤ max H

[ xi , xi+1 ] [ xi , xi+1 ]
Rectification
i.e., H (ci ) lies between the minimum and the maximum of on the
interval [ xi , xi +1 ] . Thus the sum we have written down lies between a lower sum
and an upper sum for the function H . We call such sums as Riemann sums.
This is true for every partition of the interval.
We know from basic integration theory that there is exactly one number lying
between every upper sum and every lower sum, and that number is the definite
interval. Therefore it is reasonable to define:
Length of our curve between a and b
2
b  dy  b
∫a  dx  ∫a 1 + f ′( x)  dx ---------(18.4)
= + = 2
1 dx
Similarly for x = φ ( y) and φ ′( y) are continuous on [a, b] , then the length of our
curve between a and b is
2
b  dx  b
∫a  dy  ∫a 1 + φ ′( y) dy
= + = 2
1   dy
Example18.1 Find the length of the arc of f ( x) = x on [0, 4].

3
2
Solution:
Rectification
3
As f , f ′( x) = x are both continuous on [0, 4], the length of the arc or length
1
2
2
2
3 1
4 4
9
of curve L = ∫
0 2 
∫
1 +  x 2  dx =
0
1 + x dx ,
4
9
Let 1 + x =
t , when x=0, t=1,
4
x=4, t=10
4 10 10
9 4 1 4 2 3 8  32 
∫
0
1 + x dx = ∫ t 2 dt = × × t 2 =
4 91 9 3 1 27 
10 − 1

Example 18.2 Find the length of the curve y=x2 between x =0 and x =1.
Solution:
From the definition above, we see that the integral is
1 1
∫ ∫ 1 + 4 x dx set u =2x, du = 2dx

1 + (2 x) dx =
2 2
0 0
When x=0, u=0, x=1, u=2
1
12
Hence ∫ 1 + 4 x 2 dx = ∫ 1 + u 2 du ---------------------------- (18.5),
0 2 0
b
We can find the integral ∫ 1 + x 2 dx for b > 0, as
0
1 1
( ) ( ) ( ) 
2 −2
1
 b + b + 1 + 2ln b + b 2 + 1 − b + b 2 + 1
2

4 2 2
Rectification
1 1
( ) ( ) 12 ( 2 + 5 ) 
2 2 −2
So ∫ 1 + u 2 du= 2+ 5 + 2ln 2 + 5 −
0 4  2 
1 1
( ) ( ) ( ) 
2 1 −2
Hence (18.5) becomes: 2+ 5 + 2ln 2 + 5 − 2+ 5
8  2 2 
18.2 Length of Parameterized Curve
There is one other way in which we can describe a curve. Suppose that we look
at a point which moves in the plane. Its coordinates can be given as a function
of time t. Thus, we get two functions of t, say
x = f(t), y = g(t),
We may view these as describing a point moving along a curve. The functions f
and g give the coordinates of the point as function of t.
=
Example 18.3 cosθ , y r sin θ . Then
Let, x r=
( x, y ) = ( r cosθ , r sin θ ) is a point on the circle.
( cos θ ,sin θ )
θ
r =1
As θ increases, we view the point as moving along the circle in anticlockwise

direction. The choice of letter θ really does not matter and we could use t
instead. In particular, the angle θ is itself express as a function of time. For
example, if a bug moves around the circle with uniform (constant) angular
speed, then we can write θ = ω t , where ω is constant.
Then x = cos(ω t ) , y = sin(ω t ) .
Rectification
When ( x, y ) is described by two function of t as above, we say that we have a
parameterization of the curve in terms of parameter t.
This describes the motion of a bug around the circle with angular speed ω . Note
that the parametric representation of a curve is not unique. For
example x = r sin θ , y = r cosθ also represents a point on the circle.
We shall now determine the length of a curve given by a parameterization.

Suppose that our curve is given by
x = f (t), y = g (t), with a ≤ t ≤ b
and assume that both f, g have continuous derivatives. With eqn (18.4) it is very
reasonable to define the length of our curve (in parametric form) to be
b
= ∫ f ′(t ) 2 + g ′(t ) 2 dt .
b
l a
a
Observe that when a curve is given in usual form y = f(x) we can let
t = x = g (t) and y = f (t).
This shows how to view the usual form as a special case of the parametric form.
In that case g ′(t ) = 1 and the formula for the length in parametric form is seen to
be the same as the formula we obtained before for a curve y = f(x) .It is also
convenient to put the formula in the other standard notation for the derivative.
We have
Rectification
dx dy
= f ′(t ) and = y′(t )
dt dt
Hence the length of the curve can be written in the form
2 2
 dx   dy 
b
= ∫   +   dt
b
l a
a  dt   dt 
Without loss of generality let

s(t) = length of the curve as function of t.
Thus we may write
2 2
 dx   dy 
t
=s (t ) ∫
a
  +   dt
 dt   dt 
This gives
2 2
ds  dx   dy 
=   +  = f ′(t ) 2 + g ′(t ) 2
dt  dt   dt 
Sometimes one writes symbolically
(ds)2 = (dx)2 + (dy)2
To suggest the Pythagoras theorem i.e.,
Rectification
2
ds  dy 
= 1+  
dx  dx 
Example 18.4 Find the length of the curve x = cos t , y = sin t between
=t 0,=t π
Solution:
The length is the interval
∫ ( sin t ) + ( cos t ) dt
2 2
=π
If we integrate between 0 and 2π we would get 2π . This is the length of the

circle of radius 1.
Example 18.5 Find the length of the curve x = et cos t , y = et sin t between t
=1 and t = 2.
Solution:
2 2
 t ′  t ′
2
= ∫ ( e cos t )  + ( e sin t )  dt
2
l 1
1
2
=∫ (−e sin t + e cos t ) + (e cos t + e sin t ) dt
t t 2 t t 2
2
= ∫
1
(e 2t sin 2 t + e 2t cos 2 t + e 2t cos 2 t + e 2t sin 2 t ) dt
Rectification
2
= ∫ e dt
2= 2 e 2 − e 
t
Example 18.6 Find the length of the curve

x = cos3 θ , y = sin 3 θ for 0 ≤ θ ≤ π 2
Solution:
dx
= 3cos 2 θ (− sin θ )
We have
dθ
dy
= 3sin 2 θ cosθ
dθ
Hence,
π
2
π
=l0 2
∫
0
9cos 4 θ + 9sin 4 θ cos 2 θ dθ
π
2
= 3∫ cos 2 θ sin 2 θ dθ
0
π
2
= 3∫ sin θ cosθ dθ as sin θ , cosθ > 0 for 0 ≤ θ ≤ π 2

0
Hence
π π π
2
π 32 3 2
3
3∫ sin θ cosθ dθ =
2 ∫0
l0 =
2
sin 2θ dθ =
− cos 2θ =
0
4 0 2
EXERCISES
Rectification
Find the length of the following curves:
1
=
1. y ln x , ≤x≤2,
2
2. y = 4 − x 2 , − 2 ≤ x ≤ 2 ,
1 x
=
3. y (e + e − x ) between x = 1 and x = -1
2
=
4. y ln cos x , 0 ≤ x ≤ π3 ,
5. Find the length of the circle of radius r .
6. Find the length of the curve x = cos3t, y = sin3t between t = 0 and t = π

7. Find the length of the curve x = 3t , y = 4t – 1 , 0 ≤ t ≤ 1 .
8. Find the length of the curve x = 1- cos t , y = t- sin t , 0 ≤ t ≤ 2π .
9. Using exercise (9), find the length of the curve r = sin 2 θ2 from 0 to π .
4+2 5   17 + 4  4
Ans.: 1.
5
+ ln   , 2. 2 17 + ln 
1
(
 , 3. e − , 4. ln 2 + 3 , )
2  1+ 5   17 − 4  e
5. 2π r , 6. 3, 7. 5, 8. 8 & 9. 2
Keywords: Rectification, length of curve, parametric form,
References

Rectification
CBS, India.
Suggested Readings

Lesson 19
Volume and Surface of Revolution
19.1 Introduction
Volume of Revolution: We start our applications with volumes of revolutions.

Our aim is to find the lengths, areas and volumes of the standard geometric
figures.
Let y = f ( x) be continuous function of x on the interval with [a, b] with (a < b) .

Assume that f ( x) ≥ 0 ∀ x∈ [a, b] . If we revolve y = f ( x) around axis, we
obtain a solid, whose volume we want to compute.
xi xi +1
Take a partition of [a, b] say a= x0 ≤ x1 ≤ x2 ≤ x3 ≤ .....xn =

≤ b
Let ci be a minimum of f on the interval [ xi , xi +1 ] and di be the maximum of f in

that interval. Then the solid of revolutions is that small interval lies between a
small cylinder and a big cylinder. The width of these cylinders is xi +1 − xi and the
radius is f (ci ) for the small cylinders and f (di ) for the big cylinder. Hence the
volume of revolutions, denoted by V satisfies the inequalities
n −1 n −1
∑ π f (ci )2 ( xi +1 − xi ) ≤ V ≤ ∑ π f (di )2 ( xi +1 − xi )
=i 0 =i 0
b
It is therefore reasonable to define this volume to be V = ∫ π f ( x)2 dx
a
If we revolve the curve around x = φ ( y) around y − axis and

d
φ ( y) ≥ 0 ∀ y ∈ [c, d ] , we define the volume to be V = ∫ π f ( y)2 dy
c
If the curve be expressed by = (t ), y φ (t )

x f=
b t2
V = π ∫ y dx = π ∫ (φ (t ) ) f ′(t ) dt where t1 , t2 are values of t that corresponds to

2 2
a t1
x = a and x = b respectively.
Example 19.1: Compute the volume of the sphere of radius 1.
Solution:
We take the function y= 1 − x 2 between 0 and 1. If we rotate this curve around

x − axis , we shall get half the sphere. Its volume is therefore
1
1
x3 2
∫ π (1 − x2
) dx =π ( x − ) =π
0 3 0 3
2 4
So the volume of full sphere is 2 × π = π
3 3
Example 19.2: Find the volume obtained by rotating the region between y = x3
and y = x in the first quadrant around the x − axis .
y = x2
y = x2
The graph of the region is given on the figure.
As x3 =x ⇒ x( x 2 − x) =0⇒ x=0, x =±1 , for first quadrant we take 0 ≤ x ≤ 1.

The required V volume is equal to the difference of the volume obtained by
rotating y = x and y = x 2 .
Let=
f ( x) x=
, g ( x) x 3 . Then
1 1
V = π ∫ f ( x) dx −π ∫ g ( x)2 dx
2
0 0
1 1
= π ∫ x 2 dx −π ∫ x6 dx
0 0
π π
= −
3 7
1
Example 19.3: (Volume of Chimneys) . Consider the function f ( x) = .
x
a 1 b
a
1
Let 0 < a < 1 . The volume of revolution of the curve y = between x = a and
x
1
dx
x = 1 is given by ∫ π = π ln x a = −π ln a ,
1
a
x
As a → 0 , ln a becomes very large negative, so that − ln a becomes very large

positive, and the volume becomes arbitrary large. The above figure illustrates
the chimney.
In this computation, we determined the volume of a chimney near the y − axis .

We can also fixed the volume of the chimney going off to the right, say
1
between 1 and a number b >1. Suppose the chimney is defined by y = . The
x
volume of revolution between 1 and b is given by the integral

1
b b
dx
∫  x  ∫ π=
1
π =
 
dx
0x
π ln b ,as b → ∞ we see that this volume becomes arbitrary
large (divergent integral)
But we are interested to find finite volume for the infinite chimney.
1
Example 19.4: Compute the volume of revolution of the curve y = between
x4
a and 1. Find the limit as a → 0
Solution:
1
The volume of revolution of the curve y = between x = a and x = 1
x4
1
1
1 −1
1 1
is given by the integral ∫ π 1
π∫
dx =x 2 dx =×2 x2π 2π 1 − a 
=
a a
x2 a
When a → 0 limit becomes 2π
Example 19.5 Find the volume of a cone whose base has a radius r , and a
height h , by rotating a straight line passing through the origin around the
x − axis
Solution:
y r
y= x
h r
x
h
r 1
The equation of the straight line is y = x . Slant height is y = . Hence the
h x2
2
h
r  r2 h 2 π r 2 h3 1 2
volume of the cone is ∫ π =
 x  dx π= 2 ∫
x dx × =π r h
 
2
0 h h 0 h 3 3
19.2 Surface of Revolution
Let y = f ( x) be a positive continuously differentiable function on an interval

[a,b]. We wish to find a formula for the area of the surface of revolution of the
graph of f around the x − axis , as given in the figure
a b
We shall see that the surface area is given by the integral
S = ∫ 2π y 1 +   dx
b
dy
a  dx 
The idea again is to approximate the curve by line segments. We use a partition
a = x0 ≤ x1 ≤ x2 ≤ x3....... ≤ xn = b
f ( xi +1 )
f ( xi )
fx( xi ) xfi+1
( xi +1 )
i
Li
On the small interval [ xi , xi +1 ] the curve is approximated by the line segment
joining the points ( xi , f ( xi ) ) and ( xi +1 , f ( xi +1 ) ) . Let Li be the length of the segment.
Then Li = ( xi +1 − xi ) + ( f ( xi +1 )2 − f ( xi ) )
2 2
The length of a circle of radius y is 2π y . If we rotate the line segment about the
then the x− axis area of the surface of rotation will be between 2π f (ti ) Li and
2π f (si ) Li where f (ti ) and f (si ) are the minimum and maximum of f ,
respectively on the interval [ xi , xi +1 ] . This is illustrated on Fig 1.
On the other hand, by the mean value theorem we can write
f ′(ci )( xi +1 − xi ) , ci ∈ ( xi , xi +1 )
f ( xi +1 ) − f ( xi ) =
Hence Li = ( xi +1 − xi ) + f (ci )2 ( xi +1 − xi )
2 2
1 + f ′(ci )2 ( xi +1 − xi )
=
Therefore the expression 2π f (ci ) 1 + f ′(ci )2 ( xi +1 − xi )
is an approximation of the surface of revolution of the curve over the small

interval [ xi , xi +1 ] .
n −1
Now take the sum ∑ 2π f (c ) i 1 + f ′(ci )2 ( xi +1 − xi )
i =0
This is a Riemann sum, between the upper and lower sums for the integral
b
=S ∫ 2π f ( x)
a
1 + f ′( x) 2 dx
Thus it is reasonable that the surface area should be defined by this integral, as
was to be shown.
19.2.1 Area of revolution for parametric curves given in parametric form.
Suppose that
=x f (t=
), y g (t ), a ≤ t ≤ b
We take a partition a = t0 ≤ t1 ≤ t2 ≤ t3....... ≤ tn = b
Then the length of Li between ( f (ti ), g (ti ) ) and ( f (ti +1 ), g (ti +1 ) ) is given by
Li = ( f (ti +1 ) − f (ti ) ) + ( g (ti +1 ) − g (ti ) )

2 2
= f ′(ci )2 + g ′(di )2 ( ti +1 − ti )
where ci , di are numbers between ti and ti +1
(f(ti), g(ti)) (f(ti+1), g(ti+1))
Hence 2π g (ci ) f ′(ci )2 + g ′(di )2 ( ti +1 − ti ) is an approximation for the surface of
revolution of the curve in the small interval ti , ti +1  . Consequently, it is
reasonable that the surface of revolution is given by the integral
2 2
b  dx   dy 
=S ∫ 2π y   +  dt
a  dt   dt 
when t = x , this coincides with the formula found previously. It is also useful to
write this formula symbolically S = ∫ 2π yds
where symbolically, we had used
= ( dx ) + ( dy )
2 2
ds
When using this symbolic notation, we don not put limits of integration. Only
when we use explicit parameter over an interval a ≤ t ≤ b we explicitly write the
surface area as
b
ds
S = ∫ 2π y dt
a dt
Example 19.6 We wish to find the area of a sphere for radius r > 0 .
Solution: we can view the sphere as the area of revolution of a circle for radius
r , and to express the circle in parametric form,
= θ , y r sin θ ,0 ≤ θ ≤ π
x r cos=
Then the formula gives
π
S = ∫ 2π r sin θ r 2 sin θ + r 2 cos θ dθ
0
π 2
= ∫ 2π r sin θ dθ
0
π
= 2π r 2 (− cos θ ) 0
= 4π r 2
Exercises
1. Find the volume of sphere of radius r .

Find the volumes of revolution of the following:
1 π
2. y = between x = 0 and x =
cos x 4
π
3. y = sin x between x = 0 and x =
4
4. The region between y = x 2 and y = 5 x
x
5. y = xe 2 between x = 0 and x = 1
1
6. Compute the volume of revolution of the curve y = between x = 2 and
x2
x = b for any b>2. Does this volume approach a limit as b → ∞ ? If yes, what
limit ?
4 π2 π 2.54 π π π π
Ans.: 1. π r 3 , 2. π , 3. − , 4. , 5. π (e − 2) & 6. − 3 , yes:
3 8 4 3 24 3b 24
Keywords: Lengths, area, volume, surface revolution, volume of chimneys
References


CBS, India.
Suggested Readings

Lesson 20
Double Integration
20.1 Introduction
In applications of calculus we have seen with integrals of functions of a single

variable. The integral of a function y = f(x) over an interval [a, b] is the limit of
approximating sums
∫ f ( x)dx lim∑ f (ck )∆xk ------------ (20.1)

b
=
a n→∞
k =1
Where a = x0 ≤ x1 ≤ x2 ≤ ..... ≤ xn = b, ∆xk = xk +1 − xk and ck is the any point

from the interval [xk, xk+1] . The limit in (20.1) is taken as the length of the longest
subinterval approaches zero. The limit is guaranteed to exist if f is continuous
and also exists when f is bounded and has only finitely many points of
discontinuity in [a, b] . There is no loss in assuming the intervals [xk, xk+1] to have
b−a
common length ∆x = , and limit may thus obtain by letting ∆x =0 as n → ∞ .
n
b
If f(x) > 0, then ∫a
f ( x)dx from x = a and x = b, but in general the integral has
many other important interpretations (distance, volume, arc length, surface area,
moment of inertia, mass, hydrostatic pressure, work) depending on the nature and
interpretation of f .
Double Integration
In this Lesson we shall see that integrals of functions of two or more variables
which are called multiple integrals and defined I much the same way as integrals of
functions of single variable.
Double Integrals: Here we define the integral of a function f(x, y) of two variables
over a rectangular region in xy-plane. We then show how such an integral is
evaluated and generalize the definition to include bounded regions of a more
general nature.
Double Integrals over Rectangles:

y
c
a x
b
Suppose that f(x, y) is defined on a rectangular region R defined by

R : a ≤ x ≤ b, c ≤ y ≤ d
(see the figure 1.)
We imagine R to be covered by a network of lines parallel to x-axis and y-axis, as

shown in Fig 1. These lines divide R into small pieces of area
Double Integration
∆A =∆x∆y
We number these in some order
∆A1 , ∆A2 ,...., ∆An ,
Choose a point (xk, yk) in each piece of ∆Ak and from the sum
n
=Sn ∑ f ( x , y )∆A
k =1
k k k ------------------- (20.2)
If f is continuous throughout R, then we define mesh width to make both ∆x and

∆y go to zero the sums in (2) approach a limit called the double integral of f over
R that is denoted by ∫∫ f ( x, y)dA or ∫∫ f ( x, y)dx dy

R R
n
=
Thus ∫∫
f ( x, y )dA lim
∆A→0
∑ f ( x , y )∆A
1
k k k ---------- (20.3)
R
As with functions of a single variable, the sums approach this limit no matter how
the interval [a, b] and [c, d] that determine R are subdivide, along as the lengths of
the subdivisions both go to zero. The limit (20.3) is independent of the order in
which the area ∆Ak are numbered, and independent of the choice of ( xk , yk )
within each ∆Ak . The continuity of f sufficient condition or the existence of the
double integral, but not a necessary one, and limit question exists for many
discontinuous functions also.
Double Integration
20.1.1 Properties of Double Integral
Like “single” integrals, we have the following properties for double integrals of
continuous functions which are useful in computations and applications.
(i) ∫∫ k f ( x, y)dA =k ∫∫ f ( x, y)dA (any number k)

R R
(ii) ∫∫ [ f ( x, y) + g ( x, y)] dA = ∫∫ f ( x, y)dA + ∫∫ g ( x, y)dA

R R R
(iii) ∫∫ [ f ( x, y) − g ( x, y)] dA = ∫∫ f ( x, y)dA − ∫∫ g ( x, y)dA

R R R
(iv) ∫∫ f ( x, y)dA ≥ 0 if
R
f ( x, y ) ≥ 0 on R
(v) ∫∫ f ( x, y)dA ≥ ∫∫ g ( x, y)dA if

R R
f ( x, y ) ≥ g ( x, y ) on R
(vi) If R = R1  R2 , R1  R2 , R is the union of two non-overlapping rectangles R1

and R2, we have
∫∫=
f ( x, y )dA ∫∫ f ( x, y )dA + ∫∫ f ( x, y )dA
R1  R2 R1 R2
Volume: When f(x, y) > 0, we may interpret ∫∫ f ( x, y)dA as the volume of the
R
solid enclosed by R, the planes x = a, x = b, y = c, y = d , and the surface z = f(x,

y) see fig 2.
Each term f ( xk , yk ) ∆Ak in the sum
Double Integration
n
=Sn ∑ f ( x , y )∆A
k =1
k k k is the volume of a vertical rectangular prism y that
z
f ( xk , yk )
c d
• •
a• ∆Ak
b
•
approximate the volume of the portion of the solid that stands above the box -
∆Ak . The sum Sn thus approximates what we call the total volume of the solid,
and we define this volume to be
Volume = lim Sn = ∫∫ f ( x, y)dA

R
20.1.2 Fubbin’s theorem for calculating double integrals:
Theorem 20.1. (Fubbin’s theorem (1st form))
If f(x, y) is continuous on the rectangular region R : a ≤ x ≤ b, c ≤ y ≤ d , then
d b b d
∫∫ f ( x, y)dA
=
R
∫=
∫ f ( x, y)dxdy ∫ ∫ f ( x, y)dydx
c a a c
Double Integration
Fubbin’s theorem shows that double integrals over rectangles can be calculated as
iterated integrals. This means that we can evaluate a double integral by integrating
one variable at a time, using the integration techniques we already know for
function of a single variable.
Fubin’s theorem also says that we may calculate the double integral by integrating
in either order (a genuine convenience). In particular, when we calculate a volume
by slicing, we may use either planes perpendicular to the x-axis or planes
perpendicular to y-axis. We get same answer either way.
Even more important is the fact that Fubin’s theorem holds for any continuous
function f(x, y). In particular it may have negative values as well as positive values
on R, and the integrals we calculate with Fubin’s theorem may represent other
things besides volumes.
Example 20.1: Suppose we wish to calculate the volume under the plane z = 4-x-y
over the region R : 0 ≤ x ≤ 2, 0 ≤ y ≤ 1 in the xy – plane.
Solution: The volume under the plane is given by ∫∫ (4 − x − y)dA .

R
Next we have to calculate the double integral.
Now we will complete the stated example.
1 2
∫∫ f ( x, y)dA
R
= ∫ ∫ (4 − x − y )dx dy
0 0
Double Integration
2
 1
x2 
= ∫  4 x − − xy  dy
0 0
2
1
= ∫ (8 − 2 − 2 y ) dy
0
1
= ∫ ( 6 − 2 y ) dy
0
1
=6 y − y 2 =5
0
Example 20.2 Calculate ∫∫ f ( x, y)dA for

R
f ( x, y ) = 1 − 6 x 2 y and R : 0 ≤ x ≤ 2, − 1 ≤ y ≤ 1
Solution: By Fubin’s theorem
1 2
∫∫ f ( x, y=
)dA ∫ ∫ (1 − 6 x 2
y )dx dy
R −1 0
1 2
= ∫ ( x − 2x y )
3
dy
−1 −1
1
= ∫ ( 2 − 16 y ) dy
−1
1
= 2 y − 8 y2
−1
= (2 − 8) − (−2 − 8) = 4
Reversing the order of integration gives the same answer:
Double Integration
1 2 2 1
∫ ∫ (1 − 6 x ∫ y − 3x y
y )dy dx =
2 2 2
dx
−1 0 0 −1
2
= ∫ (1 − 3x ) − (−1 − 3 x 2 )  dx
2
2
= ∫ 1 − 3 y + 1 + 3 x 2  dx
2
= 2=
2
x 0 4.
20.1.2 How to determine the limits of Integration
The difficult part of evaluating a double integral can be finding the limits of
integration. But there is a procedure to follow:
If we want to evaluate over a region R, integrating first with respect to y and then
with respect to x, we take the following steps:
1. We imagine a vertical Line L cutting through in the direction of increasing y
2. We integrate from the y-value where L enters R to the y-value where L leaves R
3. We choose x-limits that include all the vertical lines that pass through R
x =1 y= 1− x 2
Example 20.3 Change the order of integral ∫ ∫

x = 0 y = 1− x
f ( x, y )dy dx
To calculate the same double integral as an iterated integral with order of

integration reversed consider (the figure), by using the above procedure, we have
Double Integration
y
Leaves where
y = 1- x2
Leaves where
1 y = 1- x
x
0 x 1
Smallest Biggest
x=0 x=1
1 =x 1− y 2
∫ ∫
0 x = 1− y
f ( x, y )dx dy
sin x
Example 20.4 Calculate ∫∫A x dA where A is the triangle in the xy-plane
bounded by the x-axis, the line y = x and the line y = 1.
 x sin x  1
Solution: ∫  ∫ dy  dx
00 
x
 sin x y = x 
1
= ∫ y  dx
 x 
0 y =0 
∫ sin x dx =
= − cos x 0 =
− cos1+  .46
1
If we reverse the order of integration and try to calculate
Double Integration
1 1
sin x sin x
∫∫
0 y
x
dx dy , we can’t evaluate it because we can’t express ∫ x
in terms of
elementary functions.
PROBLEM
Evaluate the following integrals and sketch the region over which each integration
takes place.
3 2
1. ∫∫
0 0
(4 − y 2 )dy dx
3 0
2. ∫∫
0 −2
( x 2 y − 2 xy )dy dx
π x
3. ∫ ∫
0 0
x sin y dy dx
π sin x
4. ∫ ∫
0 0
y dy dx
1 1
∫ ∫ ye xy dx dy
y
5. Find the value of the integral
10 0
2 2x
6. Sketch the region of integration of ∫ ∫ f ( x, y)dy dx and express the integral as
0 x2
an equivalent double integral with order of integration.
Ans.: 1. 16, 2. 0, 3.
( 4 + π ) , 4. π , 5. 9 − 9e & 6.
2
2 4
Keywords: Multiple Integrals, Double Integrals, Triple Integrals, Area, Volume
Double Integration
References

Jain, R. K. and Iyengar, SRK. (2010), Advanced Engineering Mathematics, 3 rd

CBS, India.
Suggested Readings

Lesson-21
Triple Integration
21.1 Introduction
If F(x, y, z) is the function defined on a bounded region D in space (a solid ball

or truncated cone, for example of something resembling a swiss cheese, or a
finite union of such objects) then the integral of F over D defined in the
following way.
We partition a rectangular region about D into rectangular cells by planes

parallel to the co-ordinate planes, as shown in Fig.
The cells have dimensions ∆x by ∆y by ∆z . We number the cells that lie inside
D in some order ∆V1 , ∆V2 ,......., ∆Vn ,
choose a point ( xk , yk , zk ) in each ∆Vk , and form the sum
n
=Sn ∑ F ( x , y , z )∆V
k =1
k k k k ------------- (21.1)
If F is continuous and the bounding surface of D is made of smooth surfaces

joined along continuous curves, then as ∆x, ∆y and ∆z all approach zero the
sum Sn will approach all limit.
lim Sn = ∫∫∫ F ( x, y, z ) dV
D
We call this limit the triple integral of F over D. The limit also exists for some
discontinuous functions.
Triple Integration
Triple integrals share many algebraic properties with double and single
integrals. Writing F by F ( x, y, z ) and G for G ( x, y, z ) , we have the following
1. ∫∫∫ k F dV =k ∫∫∫ F dV (any number k )

D V
2. ∫∫∫ ( F ± G)dV= ∫∫∫ F dV ± ∫∫∫ G dV

D D D
3. ∫∫∫ F dV ≥ 0 if F ≥ 0 in D
D
4. ∫∫∫ F dV ≥ ∫∫∫ G dV if F ≥ G on D
D D
If the domain D of a continuous function F is partitioned by smooth surface

into a finite number of cells D1 , D2 ,...., Dn , then
5. ∫∫∫ F=
dV ∫∫∫ F dV + ∫∫∫ F dV + ..... + ∫∫∫ F dV
D D1 D2 Dn
The triple integral Evaluation is hardly evaluated directly from its definition as a
limit. Instead, one applies a three-dimensional version of Fubin’s theorem to
evaluate the integral by repeated single integrations.
For example, suppose we want to integrate a continuous function F(x, y, z)

over a region D that is bounded below by a surface z = f1(x, y) above by the
surface
z = f2(x, y) , and on the side by a cylinder C parallel to the z – axis (Fig. 2). Let
R denote the vertical projection of D onto the xy-plane enclosed by C. The
integral of F over D is then evaluated as
Triple Integration
z = f2(x, y)
z = f1(x,y y)
C
x
R
Fig.12
 f2 ( x, y ) 
∫∫∫ F ( x , y , z ) dV = ∫∫  ∫
 F ( x , y , z ) dz  dy dx

D R  f1 ( x , y ) 
f2 ( x, y )
or ∫∫∫ F ( x, y, z )dV = ∫∫ ∫ F ( x, y, z )dz dy dx -------------- (21.1)
D R f1 ( x , y )
If we omit the parenthesis .The z-limits of integration indicate that for every
(x, y) in the region R, z may extend from the lower surface z = f1(x, y) to the
upper surface z = f2(x, y). The y – and x - limits of integration have not given
explicitly in Eq (21.1) but are to be determined in the usual way from the
boundaries of R.
We will find the equation of the boundary of R by eliminating z between the

two equations z = f1(x, y) and z = f2(x, y). This gives
f2(x, y) = f1(x, y),
an equation that contains no z and that defines the boundary of R in the xy -

plane.
Triple Integration
To give the z -limits of integration in any particular instance we may use a

procedure like the one for double integrals. We imagine a line L through a point
(x, y) in R and parallel to the z-axis. As z increases, the line enters D at z = f1(x,
y) and leaves D at z = f2(x, y). These give the lower and upper limits of the
integration with respect to z . The result of this integration is now a function of
x and y alone, which we integrate over R, giving limits in the familiar way.
Leaves D at z = f2(x, y)
Enters D at z = f1(x, y)
R
(x, y)
Fig. 12
Example 21.1 Find the volume enclosed between the two surfaces z = x2+3y2
and
z = 8-x2-y2.
Solution: The two surfaces intersect on the surface
x2+3y2 = 8-x2-y2
or x2+2y2 = 4
which is elliptic .
So the volume of the surface is
Triple Integration
2 (4− x 2 )/2 8− x 2 − y 2
V= ∫ ∫ ∫ dz dy dx
−2 − (4− x 2 )/2 x +3 y 2 2
2 (4− x 2 )/2
= ∫ ∫ (8 − x − y − x − 3 y ) dy dx
2 2 2 2
−2 − (4− x 2 )/2
2 (4− x 2 )/2
= ∫ ∫ (8 − 2 x − 4 y ) dy dx
2 2
−2 − (4− x 2 )/2
2 (4− x 2 )/2
= ∫ ∫ (
2 8 − 2 x 2 − 4 y 2 dy dx )
−2 0
2 (4− x 2 )/2
 4 
= ∫ 2  (8 − 2 x 2 ) y − y 3  dx
−2 
3 0

( )  dx
2 3
8
∫  2(8 − 2 x )
(4− x 2 ) (4− x 2 )
= −
2 2
2 2
−2
3
3
2
( )
2
4 2
= ∫
3 −2
4 − x 2 dx
3
2
( )
2
8 2
=
3 0∫ 4 − x 2 dx
= 8π 2
As we know, there are sometimes two different orders in which the single
integrations that evaluate a double integral may be worked (but not always). For
triple integral there are sometimes (but not always) as many as six workable
orders of integration. The next example shows an extreme case in which all six
are possible.
Example 21.2 Each of the following integrals gives the volume of the solid
shown
Triple Integration
in Fig 3.
z
y + z =1
y
Fig 3.
1 1 2 1 1− y 2
(a ) ∫∫∫ dx dy dz (b) ∫ ∫ ∫ dx dz dy
x
0 0 0 0 0 0
1 1 1− z 2 1 1− z
(c) ∫∫ ∫ dy dx dz (d ) ∫∫ ∫ dy dz dx
0 0 0 0 0 0
1 2 1− y 2 1 1− y
(e) ∫∫ ∫ dz dx dy ( f ) ∫∫ ∫ dz dy dx
0 0 0 0 0 0
EXERCISES
1. Write six different iterated triple integrals for the volume of the rectangular
solid in the first octant bounded by the co-ordinate planes and the planes x = 1,
y = 2,
z = 3. Evaluate one of the integrals.
2. Write six different intersected triple integrals of the volume in the first octant
enclosed by the cylinder x2 + z2 = 4 and the plane y = 3. Evaluate one of the
integrals.
Triple Integration
3. Write an iterated triple integrals in the order dz dy dx for the volume of the
region bounded below by the xy-plane and above by the paraboloid z = x2+ y2
and lying inside the cylinder x2 + y2 = 4.
1 1 1− y
4. Rewrite the integral ∫ ∫ ∫ dz dy dx as an equivalent integrated integral in the
−1 x 2 0
order.
a) dy dz dx b) dy dx dz c) dx dy dz d) dx dz dy e) dz dx dy
1 2 3 21 3 3 21 31 2 2 31
Ans.: 1. ∫∫∫ dz dy dx , ∫∫∫ dz dx dy , ∫∫∫ dx dy dz , ∫∫∫ dy dx dz , ∫∫∫ dx dz dy ,
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
the value of each integral is 3, 2.

3 2 4− x 2 2 3 4− x 2 2 4− x 2 3
∫∫ ∫ dz dx dy , ∫∫ ∫ dz dy dx , ∫ ∫ ∫ dy dz dx ,
0 0 0 0 0 0 0 0 0
2 1− z 2 3 3 2 4− z 2 2 3 1− x 2
∫ ∫ ∫ dy dz dx , ∫∫ ∫ dx dz dy , ∫∫ ∫ dx dy dz . Value of each integral is

0 0 0 0 0 0 0 0 0
12 π .
2 1− x 2 x2 + y 2
3. 4 ∫ ∫ ∫ dz dy dx & 4.
0 0 0
Keywords: Triple integral, Fubini’s theorem, volume
References

Triple Integration

CBS, India.
Suggested Readings

Lesson 22
Area & Volume using Double and Triple Integration
22.1 Introduction
We have seen if we take f (x, y) = 1 in the definition of the double integral over
a region in Eqn (20.2), is the partial sum reduce to
n n
Sn = ∑ f ( xk , yk )∆Ak = ∑ ∆Ak ,
=k 1=k 1
and give area of the region as n → ∞ . In that case ∆x, ∆y approach zero. In
this case we define the area on a rectangular region R to be the limit
= lim ∑ ∆A=
Area k ∫∫ dA (22.1)
R
Example 22.1 Find the area of the region R bounded by y = x and y = x2 in the
first quadrant.
Solution: The area of the region is
1 x 1
∫ ∫ dy =
dx ∫ ( x − x 2
)dx
2
0 x 0
6
x 2 x3 1
= − =
2 3 0 6
Example 22.2 Find the area of the region R enclosed by the parabola y = x2 and
the line y = x + 2.
Solution: x 2 = x + 2 ⇒ x 2 − x − 2 = 0
x 2 − 2 x=
+ x − 2 0 i.e., x( x − 2) +=
1( x − 2) 0
( x + 1)( x − 2) =
0
x = −1, 2
2 x+2
Hence the area A = ∫ ∫ dy dx
−1 x 2
2 x+2
=∫ y dx
−1 x2
2
= ∫ −1
( x + 2 − x 2 )dx
2
x2 x3
= + 2x −
2 3 −1
 8 1 1
= 2+ 4− − −2+ 
 3  2 3
8 1 1
=2+4− − − +2
3 2 3
16 + 3 + 2
= 8−
6
7 9
=8 − =
2 2
Solution:
y
For order of integration reversed, draw a horizontal lin L2. It enters at x = ,
2
leaves at x = y . To include all such lines we let y to n from y = 0 to y=
4. The integral is
4 y
∫∫
0
y
2
f ( x, y )dx dy
22.1.1 Changing to Polar Coordinates.
When we define the integral of a function f(x, y) over a region R we divide R

with rectangles, and their areas easy to compute. But when we work in polar
coordinates, however it is more natural to subdivide R into ‘polar rectangles’ we
can find the double integral in polar form as.
θ = β r = f 2 (θ )
∫∫ F (r ,θ ) dA = ∫ ∫ F (r ,θ ) r dr dθ ------------ (22.2) , give running numbers.

θ α=r f1 (θ )
=
Where the function F (r ,θ ) is defined over a region R bounded by the areas

=θ α=
, θ β and the continuous curve
= 1 (θ ) , r
r f= f 2 (θ ) .
If F (r ,θ ) ≡ 1 the constant function whose value is one, then the value over R is
the areas of R (which agrees our earlier definition). Thus
Area of R = ∫∫ r dr dθ
R
Example 22.3 Find the area enclose by the lemniscate r 2 = 2a 2 cos 2θ .
π 4
-π 4
The area of the right-hand half to be
π
2 a cos 2 2θ
π r = 2 a cos 2 2θ
4 4
r2
∫ ∫ r dr dθ = ∫ 2
dθ
− π4 0 − π4 r =0
π
4
= ∫a cos 2θ dθ
2
π
−4
π
2 4
a
= sin 2θ
2 − π4
a2
= [1 − (−1)]
2
= a2
The total area is therefore 2a2.
22.2 Volume using Triple Integral
If F ( x, y, z ) ≡ 1 is the constant function whose volume is one, then the sums in

n n
Eq (1) reduce to Sn = ∑1.∆VK = ∑ ∆VK
=k 1=k 1
As ∆ x , ∆ y , ∆ z all approaches zero, the cells ∆Vk become smaller and we need
more cells to fill up D. We therefore define the volume of D to be the triple

integral of the constant function f(x, y, z) = 1 over D.
n
Volume of D = lim ∑ ∆Vk =
∫∫∫ dV . k =1 D
The triple integral Evaluation is hardly evaluated directly from its definition as a
limit. Instead, one applies a three-dimensional version of Fubin’s theorem to
evaluate the integral by repeated single integrations.
22.3 Integrals in Cylindrical and Spherical Coordinates
Z
. (0,0,c)
. Y
(a,0,0) . (0,b,0)
X
Fig. 4
Fig. 4 shows a system of mutually orthogonal coordinates axes OX, OY, OZ.
The Cartesian coordinates of a point P(x, y, z) in the space may be read from the
coordinates axes by passing planes through P perpendicular to each axis. The
points on the x-axis have their y- and z- ordinates both zero. Points in a plane
perpendicular to the z-axis, say, all have the same z - coordinate. Thus of the
points in the plane perpendicular to the z- axis and 5 units above the xy-plane
all have coordinates of the form (x,y,5) . We can write z = 5 as an evaluation for
this plane. The three planes x = 2, y = 3, z = 5 intersect in the point P (2, 3, 5).
The points of the yz- plane are obtained setting x = 0. The three coordinates
planes x = 0, y = 0, z = 0 divide the space into eight cells, called octants. The
octant in which all three coordinates are positive is the first octant, but there is
no conventional numbering of the remaining seven octants.
Example 2. Describe the set of points P(x, y, z) whose Cartesian coordinates

satisfy the simultaneous equation x2 + y2 = 4, z = 3.
Solution: The points all horizontal plane z = 3, and in this plane they lie in this
cirle x2 + y2 = 4.Thus we may describe the set of the circle in the plane x2 + y2 =
4 in the plane z = 3.
22.3.1 Cylindrical Coordinates
It is frequently convenient to use cylindrical coordinates (r ,θ , z ) to locate a

point in space. These are just the polar coordinates (r ,θ ) used instead of (x, y)
in the plane z = 0, coupled with the z- coordinates. Cylindrical and Cartesian
coordinate are therefore related by the following equations : Equations relating
cartesian and cylindrical coordinates.
x = r cosθ =
r x2 + y 2
y
y = r sin θ tan θ =
x
z=z
22.3.2 Spherical Coordinates
Spherical coordinates are useful when there is a center of symmetry that we can
take as the origin. The spherical coordinates ( ρ ,ϕ ,θ ) are shown the first
coordinates ϕ = OP is the distance from the origin to the point. It is never

negative. The equation ϕ = constant describes the surface of the sphere of radius
ϕ with centre O. z
φ ρ P (ρ ,θ ,φ )
y
θ
x
The second spherical coordinate φ , is the angle measured down from the z-axis
to the line OP. The equation ρ = constant describes cone with vertex at O, axis
OZ and generating angle φ , provide we broaden our interpretation of the word
π
“cone” to include the xy- plane for which φ = and cones the generation
2
π
angles greater than .
2
The third spherical coordinates θ is the same as the angle θ in cylindrical

coordinates, namely, the angle from the xz-plane the plane through P and the z-
axis.
22.3.3. Coordinate Conversion Formulas
We have the following relationships between these Cartesian (x, y, z),

cylindrical (r ,θ , z ) , and spherical ( ρ ,ϕ ,θ )
Polar to Rectangular Spherical to Cylindrical Spherical to

Rectangular
x = r cosθ r = ρ sin φ x = ρ sin φ cosθ
y = r sin θ r = ρ cos φ y = ρ sin φ sin θ
z=z θ =θ z = ρ cosθ
∫∫∫ dx dy dz
Volume : = ∫∫∫
= dz rdr dθ ∫∫∫ ρ sin θ d ρ dφ dθ
2
Exercises
1. Find the area of the region R enclosed by the parabola y = x2 and the line y =
x +1
2. Find the area of the region R bounded by y = x and x = y 2 in the first

quadrant.
3. Find the volume of the solid in the first octant bounded by the paraboloid.
z = 36 – 4x2 – 9y2
4. Find the volume of the solid enclosed between the surfaces x2 + y2 = 92 and
x2 + z2 = 92 .
x y z
5. The volumes of the tetrahedron bounded by the plane + + =1 and the
a b c
coordinate planes.
6. The volume in the first octant bounded by the planes x + z = 1, y + 2z = 2.
7. The volume of the wedge cut from the cylinder x2 + y2 = 1 and the plane z =
y above and plane below.
8. The volume of the region in the first octant bounded by the coordinate
planes, above by the cylinder x2 + z = 1 and on the right by the paraboloid y =
x2 + z2
(Hint: Integrate first with respect to y)
16a 3 1 2 2 2
Ans.: 1. ,2. , 3. 27π , 4. , 5. abc , 6. , 7. & 8.
3 6 3 3 7
Keywords: Area, Volume, Double Integral, Triple Integral
References


CBS, India.
Suggested Readings

Lesson 23
Gamma Function
23.1 Introduction: We shall define a function known as the gamma function,

Γ( x) which has the property that Γ(n) =(n − 1)! for every positive integer n. It
may be regarded then a generalization of factorial n to apply to values of the
variable which are not integer. The function is defined in terms of an improper
integral. This integral cannot be evaluated in terms of the elementary functions.
It has great importance in analysis and in applications.
Definition 23.1 The Gamma Function: The gamma function is defined by the
improper integral
+∞
Γ ( λ + 1) =
∫ e t dt ------------- (23.1)
−t λ
0
which converges for all λ > −1
To deduce some of the properties of the gamma function, let us integrate Eq.
(23.1) by parts:
+∞ R
∫0
e − t t λ dt = lim
R →+∞ 0 ∫ e − t t λ dt
= lim  −e − t t λ + λ ∫ e − t t λ dt 
R R
R →+∞ 
 0 0 
 − Rλ  +∞
= lim  R + 0  + λ ∫ e − t t λ −1 dt
R →+∞
 e  0
+∞
= λ ∫ e − t t λ −1 dt
0
Gamma Function
i.e. Γ(λ + 1) =λΓ(λ ) ------------ (23.2)
If we let λ = 0 in Eq 1. these results
∞ ∞
λ ∫ e − t dt =
Γ(1) = −e − t
0
=
1
0
Using Eq 23.2, we obtain
Γ(2) =1.Γ(1) =1
Γ(3) =2.Γ(2) =2!
Γ(4) =3.Γ(3) =3! ------------ (23.3)
The equations above represent another important property of the gamma

function. 1 + λ is a positive integer.
Γ(λ + 1) =
λ ! ------------ (23.4)
It is interesting to note that Γ(λ ) is defined for all λ except

λ = 0, −1, −2,...... by the functional equation Γ(λ + 1) =λΓ(λ ) ; infact, we need
to know Γ(λ ) only for 1 ≤ λ ≤ 2 to compute Γ(λ ) for all real values of λ . Fig
1. Illustrates the graph Γ(λ )
Gamma Function
x
-4 -3 -2 -1 1 2 3 4
Fig 1. Γ(λ ) the Gamma function
Certain constants related to Γ( x) . We shall show that Γ ( 12 ) =π . In order to

do this, we compute first the so-called probability integral.
+∞
∫ e − x dx = π
2
1
Theorem 23.1. 2
0
To prove this, consider the double integral of e − x − y2

2
over two circular sectors D1
and D2 and the Square S indicated in Fig 2.
Since the integral is positive, we have
∫∫ <∫∫ < ∫∫
D1 S D2
(24.1.5)
y (R,R)
S
D1
D2
O (R,0) (R 2 ,0) x
Fig 2.
Gamma Function
Now evaluate these integrals by iterate integrals, the centre one in rectangular
coordinates, and other two in polar coordinates:
R π R R R 2 π
∫ e − r r dr ∫ dθ < ∫ e − x dx ∫ e − y dy < ∫ e − r r dr ∫ dθ
2 2 2 2 2 2
0 0 0 0 0 0
π
( ) (∫ ) π
( )
R 2
− R2 − x2
1− e < < 1 − e −2 R
2
e dx
4 0 4
Now let R → ∞ , then
(∫ ) π
+∞ 2
e − x dx =
2
0 4
+∞ π
∫ e − x dx =
2
i.e.,
0 2
1
Theorem 23.2. Γ   =π
2
1 +∞ +∞
∫ ∫
1 −1 −1
Now, =
Γ  e − t=
t 2 dt e − t t 2 dt
2 0 0
+∞
∫ e dy π set t = y2
−y
= 2=
2
+∞
∫
5
Example 23.1 Evaluate the integral x 4 e − x dx
0
Solution: Set x = t2, dx = 2tdt
+∞ +∞
∫ x 4 e − x dx = 2 ∫ t 2 e − t dt = 2Γ( 92 )
5 7
0 0
Gamma Function
From the recursive relation (2) , we obtain
7 5 3 3 105 π
Γ(=
9
2) . . Γ(=
2)
2 2 2 8 2
Finally, the volume of the integral is
+∞ 105 π 105 π
∫
5
x 4 e − x dx =×
2 × =
0 8 2 8
Example 23.2 Express the product

f (r ) = r (r + h)(r + 2h).......[r + (n − 1)h] as a quotient of gamma functions.
Solution: We have
 r  r  r  r 
f=
(r )   + 1 + 2  .....  + (n − 1)h  h n
 h  h  h  h 
Γ ( hr + 1) Γ ( hr + 2 ) Γ ( hr + n )
=h n
....
Γ ( hr ) Γ ( hr + 1) Γ ( hr + n − 1)
Γ ( hr + n ) n
= .h
Γ ( hr )
r
obtained by the recursion Eq. 2 with λ =
h
Some special cases of the result of Example 2 are interesting. For particular
case, set r = 1 and h = 2. Then
Gamma Function
2n Γ(n + 12 )
1.3.5....(2n − 1) = 1
Γ( 2 )
1 1 π
But Γ( 2 ) =
Γ( 32 ) = .
2 2
Hence
2n Γ(n + 12 )
1.3.5....(2n − 1) =
π
However,
2.4.6.....2n
=
1.3.5....(2 n − 1) 1.3.5....(2n − 1)
2.4.6.....2n
(2n)!
=
2n n !
Now combining the two equations above , we get
(2n)! π (2n)!
Γ ( n + 12=
) × = π
n
2 n! 2 n 2 n
2 n!
for n = 1, 2, ……
Other expressions for Γ ( x )
+∞
23.3. Γ ( x ) r
Theorem = x
∫e
− rt x −1
t dt , r > 0, x > 0
0
Gamma Function
This follows from the definition

+∞
=Γ ( x) ∫e
− t x −1
t=
dt , set rt y
0
+∞
Theorem 23.4. Γ ( x ) = ∫
2 e − t t 2 x−1 dt
2
Proof: Set t2 = y
Extension of definition
Definition : For n =1,2,….
Γ ( x + n)
Γ ( x) = , −n < x < −n + 1
x( x + 1)( x + 2).....( x + n − 1)
Thus we have defined Γ ( x ) for all x except x = 0,-1,-2,…. Observe that when
n =1 the right hand side of (6) depends on the values of Γ ( x ) in the interval
0<x<1. It is clear that Γ ( x ) has been defined for negative x in such a way that
equation
Γ ( x + 1) =xΓ ( x ) for x ≠ 0, −1, −2,.... ---------(23.6)
Example 4. Compute Γ ( 12 )
From equation (7), we have
Γ ( − 12 + 1) =− 12 Γ ( − 12 )
Gamma Function
i.e., Γ ( 12 ) =− 12 Γ ( − 12 )
i.e., Γ ( − 12 ) =−2 π
Exercise
Evaluate each integral

+∞
1. ∫
0
x e − x dx
+∞
∫ x 2 e − x dx
2
2.
0
+∞
∫x
−4
3. e − x dx
0
+∞
∫ (1 − x) e − x dx
3
4.
0
+∞
∫x e − x dx
3
5.
0
+∞
∫e
−t x
6. Show that the improper integral t dt converges for x > −1 and diverges
0
for x ≤ −1 .
1
dx
7. Compute ∫
0 x ln ( 1x )
∫
8. Evaluate 2−9 x dx using gamma function ( Hint : 2−9 x = e −9 x ln 2 )
2 2 2
Gamma Function
π
Ans.: 1. Γ ( 32 ) or , 2. 6, 3. ∞ , 4. -9394, 5. 2 × 7!, 6. ,7. 2π & 8.
2
1 π
6 ln 2
Keywords: Gamma Function, Convergence of Integral, Factorial Function
References

CBS, India.
Suggested Readings

Lesson 24
The Beta Function
24.1 Introduction
In this Lesson we shall introduce a useful function of two variables known as

beta function. Its usefulness is considerably overshadowed by that of gamma
function. In fact, we shall show that it can be evaluated in terms of the latter
function. As consequence, it would be unnecessary to introduce it as a new
function. Since it occurs so frequently in analysis, a special designation for it is
accepted.
Definition 2.2
For x, y positive we define the Beta function by
Using the substitution u = 1 - t it is easy to see that
Theorem 24.1.
Here we say the beta function is symmetric.
To evaluate the Beta function we usually use the Gamma function. To find their
relationship, one has to do a rather complicated calculation involving change of
variables (from rectangular into tricky polar) in a double integral.
When x and y are positive integers, it follows from the definition of the gamma
function that:
The Beta Function
Theorem 24.2. For 0 < x < ∞ ,0 < y < ∞ ,
β ( x, y ) = ∫0+ ( sin t ) ( cos t )

∞ 2 x −1 2 y −1
dt
To prove this set t = sin 2 u in the integral.
Theorem 24.3. For 0 < x < ∞ ,0 < y < ∞ ,
∞ t x −1
β ( x, y ) = ∫0+ x + y dt
( )
1 + t
Here the change of variable=

t u (1 + u ) −1 suffices.
It has many other forms, including:
Theorem. For 0 < x < ∞ , 0 < y < ∞
The Beta Function
Proof : When x and y are arbitrary positive numbers, the proof proceeds as
follows. From the double integral of the nonnegative function t 2 x −1u 2 y −1e − t −u 2
2
over the three regions D1 , D2 and S of figure 1 of Lesson 23. Now, however, t
and u are the variables, however, t and u are the variables x and y positive
constants. We have relation (23.5) of Lesson 23 as before. Again we evaluate
the central double integral by iteration in rectangular coordinates: the other two,
in polar coordinates:
π 2 x + 2 y −1 2 y −1
∫ cos 2 x −1 θ sin 2 y −1 θ dθ ∫0 e − r r dr < ∫0 t 2 x −1e − t dt ∫0 u

R R R
e − u du
2 2 2
2
π 2 x + 2 y −1
< ∫0 cos θ sin θ dθ ∫0 e r
2 2 x −1 2 y −1 R 2 −r2
dr
Now, if we let R become infinite and use Theorems 23.4 and 24.3, we obtain
1 1 Γ( x) Γ( y )
B( y, x) =Γ( x + y ) ,0 < x , 0 < y
2 2 2 2
This completes the proof of the theorem.
∫
1
Example 24.1 Evaluate 0
x 4 (1 − x)3dx
Γ(5)Γ(4) 1
∫0 x (1 − x) dx = ∫0 x (1 − x=
) 4−1dx B= =
1 4 3 5 −1 1
Solution: (5,4)
Γ(9) 280
1
∫
1
Example 24.2 Evaluate dx
0 3
x 2 (1 − x)
The Beta Function
1 −1 2
Γ( 13 )Γ( 23 )
∫0
−1
β
1
∫ − = = =Γ( 13 )Γ( 23 )
1
dx =
1 3
Solution: x (1 3
x ) dx ( ,
1 2
)
Γ(1)
3 3
0 3
x 2 (1 − x)
∫
1
Example 24.3 Evaluate 0
x .(1 − x)dx
3
Γ ( 32 ) Γ ( 2 )
∫0 x (1 − x) dx = β ( 2 ,2 ) =
−1
∫
1 1
x (1 − x)dx= 2 −1
2
3
Solution: 0
Γ ( 72 )
Γ ( 32 ) =
1
2 π
Γ ( 52 ) =
3
4 π
Γ ( 72 ) =
15
8 π
4
∫
1
Thus 0
x .(1 − x)dx =
15
Example 24.4 Given , show that
Proof: We know,
for
= Γ(q )Γ(1 − q )
Example 24.5 Evaluate
Solution: Let
The Beta Function
Exercises
1.
2.
3.
4.
5.
6.
7.
8.
9.
Keywords: Gamma Function, Beta Function, Polar Coordinate.
References

The Beta Function
CBS, India.
Suggested Readings

Lesson 25
Introduction
In this lesson we introduce basic concepts of theory of ordinary differential equations.

Formation of the differential equation from a given family of curves is explained. Differ-
ent types of solutions are defined. The given definitions are supplemented by some simple
examples.
25.1 Differential Equations
An equation involving derivatives or differentials of one or more dependent variables with

respect to one or more independent variables is called a differential equation. An ordinary
differential equation of order n is defined by the relation
F (t, x, x(1) , x(2) , . . . , x(n) ) = 0 (25.1)
where x(n) stands for the nth derivative of unknown function x(t) with respect to the
independent variable t. For example
5
d4 x d2 x dx
4
+ 2 + = et (25.2)
dt dt dt
dx
= x + sin x. (25.3)
dt
25.1.1 Order of a Differential Equation
The order of a differential equation is referred to the highest order derivative involved in
the differential equation. For example, the order of the differential Equation (25.2) is four.
25.1.2 Degree of Differential Equation
The degree of a differential equation is the degree of the highest order derivative which
occurs in it; after the differential equation has been made free from radicals and fractions
as far as derivatives are concerned, e.g. in differential Equation (25.2), the degree is one.
Introduction
25.1.3 Linear and Nonlinear Differential Equation
A differential equation is called linear if (a) every dependent variable and every derivative
involved occurs in first degree only, and (b) no product of dependent variables and/or
derivatives occur. A differential is not linear is called nonlinear. For examples, Equation
(25.2) is linear and (25.3) is nonlinear.
25.2 Solution of a Differential Equation
Any relation between the dependent and independent variables, when substituted in the
differential equation, reduces it to an identity is called a solution of differential equation.
For example, y = e2x is a solution of y ′ = 2y .
25.2.1 Example
Show that y = A/x + B is solution of

′′ 2
y + y′ = 0
x
Solution: We have the differential equation

′′ 2
y + y ′ = 0. (25.4)
x
Also given that
y = A/x + B. (25.5)
Differentiating (25.5) w.r.t. x
y ′ = −A/x2 . (25.6)
Differentiating (25.6) w.r.t. x
y ′′ = 2A/x3 . (25.7)
Substituting (25.6) and (25.7) into (25.4), we have

2A 2A
− 3 = 0.
x3 x
230 2 www.AgriMoon.Com
Introduction
25.2.2 Complete, Particular and Singular Solutions
Let
F (t, x, x(1) , x(2) , . . . , x(n) ) = 0 (25.8)
be an n-th odder differential equation.

• A solution of (25.8) containing n independent constants is called general solution.
• A solution of (25.8) obtained from a general solution by giving particular value to

one or more of the n independent arbitrary constants is called particular solution.
• A solution which cannot be obtained from any general solution by any choice of the
n independent arbitrary constants is called singular solution.
25.3 Formation of Differential Equations
An n-parameter family of curves is a set of relations of the form {(x, y) : f (x, y, c1, c2 , ..., cn ) =
0}, where f is real valued function of x, y, c1, c2 , ..., cn and each ci (i = 1, 2, ...n) ranges over
an interval of real values.
Suppose we are given a family of curves containing n arbitrary constants. Then by differ-
entiating it successively n times and eliminating all arbitrary constants from the (n + 1)
equations we obtain an nth order differential equation whose solution is the given family
of curves. We now illustrate the procedure of forming differential equations with the help
of some examples.
25.4 Example Problems
25.4.1 Problem 1
Find the differential equation of the family of curves y = emx , where m is an arbitrary
constant.
Solution: We have the family of curves
y = emx . (25.9)
Introduction
Differentiating (25.9) w.r.t x, we get
y ′ = memx . (25.10)
Now, we eliminate m from (25.9) and (25.10) and using m = loge y , we obtain the required
differential equation as
y ′ = y loge y.
25.4.2 Problem 2
Obtain the differential equation satisfied by the family of circles x2 + y 2 = a2 , where a is

an arbitrary constant.
Solution: The family of circles is given as
x2 + y 2 = a2 . (25.11)
x + yy ′ = 0,
which is the required differential equation.
25.4.3 Problem 3
Obtain the differential equation satisfied by xy = aex + be−x + x2 , where a and b are an
arbitrary constant.
Solution: Given family of curves
xy = aex + be−x + x2 . (25.12)
xy ′ + y = aex − be−x + 2x, (25.13)
Differentiating (25.14) w.r.t x and using (25.14), we get
xy ′′ + 2y ′ = (xy − x2 ) + 2, (25.14)
which is the required differential equation.
Introduction
Remark: From the above examples we observed that the number of arbitrary con-
stants in a solution of a differential equation depends upon the order of the differential
equation and is the same as its order. Hence a general solution of an nth order differential
equation will contain n arbitrary constant.
Suggested Readings
Boyce, W.E. and DiPrima, R.C. (2001). Elementary Differential Equations and Boundary
Value Problems. Seventh Edition, John Willey & Sons, Inc., New York.
Raisinghania, M.D. (2005). Ordinary & Partial Differential Equation. Eighth Edition. S.
Chand & Company Ltd., New Delhi.
Kreyszig, E. (1993). Advanced Engineering Mathematics. Seventh Edition, John Willey
& Sons, Inc., New York.
Arfken, G.B. (2001). Mathematical Methods for Physicists. Fifth Edition, Harcourt Aca-
demic Press, San Diego.
Grewal, B.S. (2007). Higher Engineering Mathematics. Fourteenth Edition. Khanna
Publishilers, New Delhi.
Edwards, C.H., Penney, D.E. (2007). Elementary Differential Equations with Boundary
Value Problems. Sixth Edition. Pearson Higher Ed, USA.
Piskunov, N. (1996). Differential and Integral Calculus (Volume - 2). First Edition. CBS
Publisher, Moscow.
Lesson 26
Differential Equation of First Order
In this lesson we present solution techniques of differential equations of first order and
first degree. We shall mainly discuss differential equation of variable separable form,
homogeneous equations and equations reducible to homogeneous form.
There are two standard forms of differential equations of first order and first degree,
namely,
dy
= f (x, y) or Mdx + Ndy = 0
dx
Here M and N are functions of x and y , or constants. We discuss here some special forms
of these equations where exact solution can easily be obtained.
26.1 Separation of Variables
If in a differential equation, it is possible to get all the functions x and dx to one side and
all the functions of y and dy to the other, the variables are said to be separable. In other
words if a differential equation can be written in the form F (x)dx + G(y)dy = 0, we say
variables are separable and its solution is obtained by integrating the equation as
Z Z
F (x)dx + G(y)dy = c,
where c is a integration constant.
26.2.1 Problem 1
dy
Solve = ex+y + x2 ey .
dx
Solution: For separating variables, we rewrite the given equation as
e−y dy = (ex + x2 )dx.
Integrating the above equation we have
−e−y = ex + x3 /3 + c,
where c is an arbitrary constant.
26.2.2 Problem 2
Solve 3ex tan ydx + (1 − ex ) sec2 ydy = 0.

Solution: Separating the variables, we get
3ex sec2 y
dx + dy = 0.
1 − ex tan y
Integration gives
−3 log(1 − ex ) + log(tan y) = log c,
26.3 Equations Reducible to Separable Form
Differential equation of the form

dy dy
= f (ax + by + c) or = f (ax + by)
dx dx
can be reduced by the substitution ax + by + c = v or ax + by = v to an equation in which
variables can be separated.
26.3.1 Example
dy
Solve = sec(x + y).
dx
Solution: Let, x + y = v so that
dy dv
= − 1. (26.1)
dx dx
Using (26.1), the given differential equation becomes
dv
= sec v + 1. (26.2)
dx
This equation is of separable form. Thus we have
1 2 cos2 (v/2) − 1
dx = dv ⇒ dx = dv
sec v + 1 1 + 2 cos2 (v/2) − 1
Further simplifications gives

1 2
dx = 1 − sec (v/2) dv
2
Integrating and substituting the value of v , we obtain y − tan 21 (x + y) = c.
26.4 Homogeneous Differential Equation
A differential equation of first order and first degree is said to be homogeneous if it can
be put in the form
dy
= f (y/x).
dx
These equations can be solved by letting y/x = v and differentiating with respect to x as
dv dv
v+x = f (v) ⇒ x = f (v) − v.
dx dx
Then, separating variables, we have
dx dv
=
x f (v) − v
Integrating the above equation we obtain

dv
Z
log x + c = ,
f (v) − v
where c is an arbitrary constant. The solution is obtained by replacing variable v by y/x.
26.4.1 Example
Solve the differential equation

dy y y
= + tan
dx x x
Solution: Since the right hand side of the given equation is function of y/x alone, the
given problem is homogeneous equation. Substituting y/x = v so that
dy dv
=v+x (26.3)
dx dx
the given equation becomes
dv dx cos v
v+x = v + tan v → = dv
dx x sin v
Integrating and substituting the value of v , we get the solution as
y
cx = sin ,
x
26.5 Equations Reducible to Homogeneous Form
Equation of the form

dy ax + by + c a b
= ′ , 6= ′ (26.4)
dx a x + b′ y + c′ a′ b
can be reduced to homogeneous form. The procedure is as follows:
Take
x = X + h and y = Y + k
where X, Y are new variables and h, k are constants to be chosen so that the resulting
equation in X, Y becomes homogeneous. From above we have dx = dX , and dy = dY , so
that dy/dx = dY /dX . Now the given differential equation in new variables becomes
dX aX + bY + (ah + bk + c)
= ′ (26.5)
dY a X + b′ Y + (a′ h + b′ k + c′ )
In order to make (26.5) homogeneous, the constant h and k must satisfy the following
algebraic equations
ah + bk + c = 0 , a′ h + b′ k + c′ = 0 (26.6)
Solving equations (26.6), we obtain

bc′ − b′ c ca′ − c′ a
h= , k = (26.7)
ab′ − a′ b ab′ − a′ b
provided ab′ − a′ b 6= 0. Knowing h and k we have
X = x − h, Y = y − k. (26.8)
The Equation (26.5) now reduces to

dY aX + b(Y /X)
= ′ (26.9)
dX a + b′ (Y /X)
which is a homogeneous equation in X and Y which can be solved by substituting Y /X =
v . After getting solution in X and y , we remove X and Y using (26.8) and obtain solution
in terms of x and y .
26.5.1 Example

dy (x + y + 4)
=
dx (x − y − 6)
Solution: Let x = X + h, y = Y + k , so that dy/dx = dY /dX and using this, the

given differential equation reduces to
dy X + Y + (h + k + 4)
= . (26.10)
dx X − Y + (h − k − 6)
Choose h and k such that h + k + 4 = 0, h − k − 6 = 0, and by solving, we get h = 1 and
k = −5. New variables becomes X = x − 1 and Y = y + 5. Using this into (26.10), we
obtain
dY 1 + Y /X
= . (26.11)
dX 1 + Y /X
Substituting
dY dv
Y = Xv and =v+X
dX dX
the Equation (26.11) becomes
dX 1−v dv vdv
= 2
dv = 2
dv − . (26.12)
X 1+v 1+v 1 + v2
Integrating the above equation, we get
log X = tan−1 v − (1/2) log(1 + v 2 ) + (1/2) log c
Further simplifications gives
2 log X + log(1 + Y 2 /X 2 ) − log c = 2 tan−1 (Y /X), as v = Y /X
Thus, we get
X 2 + Y 2 = ce2 tan (Y /X)

−1
;
Replacing X and Y as X = x − 1 and Y = y + 5 we obtain the general solution as
(x − 1)2 + (y + 5)2 = ce2 tan ((y+5)/(x−1))

−1
.
Suggested Readings
Dubey, R. (2010). Mathematics for Engineers (Volume II). Narosa Publishing House.
New Delhi.
Weir, M.D., Hass, J. and Giordano, F.R. (2005). Thomas Calculus. Eleventh Edition.
Pearson Education. New Delhi.
McQuarrie, D.A. (2009). Mathematical Methods for Scientist and Engineers. First Indian
Edition. Viva Books Pvt. Ltd. New Delhi.
Publisher, Moscow.
Lesson 27
Linear Differential Equation of First Order
In this lesson we shall learn linear differential equations of first order. Such equations
are very often used in applications. Solution strategies of solving such equations will be
discussed. Further a another special form of differential equation which can be reduced
to linear differential equation of first order will be studied.
27.1 Linear Differential Equation
A first order differential equation is called linear if it can be written in the form
dy
+ P (x)y = Q(x) (27.1)
dx
where P and Q are constants or function of x only.
A method of solving (27.1) relies on multiplying the equation by a function called inte-
grating function so that the left hand side of the differential equation can be brought under
a common derivative. Suppose R(x) is an integrating factor of the (27.1). Multiplying the
(27.1) by R(x), we obtain
dy
R(x) + P (x)R(x)y = Q(x)R(x) (27.2)
dx
Suppose, we wish that the L.H.S of (27.2) is the differential coefficient of some product.
dy
Clearly, the term R(x) dx can only be obtained by differentiating the product R(x)y(x). In
other words, we wish to have
dy d
R(x) + P (x)R(x)y(x) = (R(x)y(x)). (27.3)
dx dx
This implies
dy dy dR
R(x) + P (x)R(x)y(x) = R(x) + y(x) .
dx dx dx
On cancelling the first term on both the sides we obtain
dR dR
P (x)R(x)y(x) = y(x) ⇒ = Rdx.
dx R
R
Integrating the above equation, we get log R = P dx. Note that the constant of integration
is not important here because the integrating factor will be used to multiplying both the
sides of the differential equation and therefore it will be cancelled. Thus, an integrating
factor (I.F.) of the differential Equation (27.1) is
R
P dx
R=e (27.4)
The Equation (27.2) now reduces to

d
(Ry) = QR
dx
By integrating above equation, we have
Z
Ry = RQdx + c,
or
R Z R
P dx P dx
ye = Qe dx + c,
which is required solution of given differential equation. Here C is the constant of inte-
gration.
27.2.1 Problem 1
dy
Solve x cos x + y(x sin x + cos x) = 1, 0 < x < π/2.
dx
Solution: We rewrite the given equation as
dy 1 sec x
+ (tan x + ) = .
dx x x
An I.F. of the given differential equation is
R
(tan x+ x1 )dx
e = elog x sec x = x sec x.
Hence, the required solution is

Z
yx sec x = sec2 xdx + c,
or
yx sec x = tan x + c,
where, c is an arbitrary constant.
27.2.2 Problem 2
dy
Solve (1 + x2 ) = x(1 − y).
dx
Solution: Rewriting the given differential equation in standard form
dy x x
+ 2
y= .
dx 1 + x 1 + x2
The I.F. is R x
dx 1 2 p
I.F. = e 1+x2 = e 2 ln(1+x ) = 1 + x2 .
The solution is
x
p Z
2
y 1+x = √ +c ⇒ y = 1 + c (1 + x2 )−1/2
1 + x2
Here c is an arbitrary constant.
27.3 Equations Reducible to Linear Form
A equation of the form

dy
f ′ (y) + P f (y) = Q, (27.5)
dx
dy
can be reduced to linear form, by substituting f (y) = v so that f ′(y) dx = dv/dx. The
Equation (27.5) then becomes
dv/dx + P v = Q, (27.6)
which is linear in v and x and its solution can be obtained with the help of I.F. as before.
R
Thus, we have an I.F.= e pdx and the solution is
R Z R
pdx P dx
ve = Qe dx + c.
Finally, we replace v by f (y) to obtain the required solution.
27.4.1 Problem 1
dy
Solve cos y + 2x sin y = x.
dx
dy dv
Solution: Substitution sin y = v which implies cos y dx = dx reduces the given differential
equation to
dv
+ 2xv = x.
dx
R 2
This is a linear differential equation of first order and its I.F. is e 2xdx = ex . The solution
of the equation in v is given by
1
Z
x2 2 2
ve = xex dx + c ⇒ v = + ce−x .
2
Replacing v by sin y we get the required solution as

−1 1 −x2
y = sin + ce .
2
27.4.2 Problem 2
dy
Solve + x sin 2y = x3 cos2 y
dx
Solution: Dividing the given differential equation by cos2 y , we obtain
dy
sec2 y + 2x tan y = x3 .
dx
dy dv
Putting tan y = v so that sec2 y dx = dx . Hence the above equation becomes
dv
+ 2xv = x3 ,
dx
2
which is linear. Its I.F. is ex and its solution is given as follows
Z
x2 2
ve = ex x3 dx + c,
2 1 2
vex = (x2 − 1)ex + c.
2
Replacing v by tan y we obtain the required solution.
27.5 Bernoulli’s Equation
An equation of the form
dy/dx + P y = Qy n (27.7)
where P and Q are constants or function of x only and n is constant except 0 and 1 is
called Bernoulli differential equation. This equation can easily be solved by multiplying
both sides by y −n as
y −ndy/dx + P y 1−n = Q (27.8)
dy 1 dv
Setting y 1−n = v , so that y −n dx = (1−n) dx , the Equation (27.8) becomes
dv/dx + P (1 − n)v = Q(1 − n),

R
which is linear in v and x. Its I.F. is e P (1−n)dx and hence the required solution is
R Z R
1−n P (1−n)dx P (1−n)dx
y e = Qe dx + c,
27.5.1 Example
dy
Solve x + y = y 2 ln x .
dx
Solution: Rewrite the given equation

dy 1
y −2 + y −1 = −x−1 ln x (27.9)
dx x
dy dv
Putting y −1 = v so that −y −2 dx = dx . Then the Equation (27.9) gives
dv 1
− v = x−1 ln x. (27.10)
dx x
R 1
x dx 1
The I.F. of the differential Equation (27.10) is e− = x, and hence the solution be-
comes
1
Z
v =− x−2 log xdx + c
x
or by replacing v by y −1 we get
y −1 = 1 + ln x + cx,
Suggested Readings
New Delhi.
Publisher, Moscow.
Lesson 28
Exact Differential Equation of First Order
This lesson provides an overview of exact differential equation. A necessary condition

for a differential equation to be exact will be derived. Then different solution techniques
will be discussed. Several examples to clarify the ideas will be supplemented.
28.1 Exact Differential Equation of First Order
If M and N are functions of x and y , the equation Mdx + Ndy = 0 is called exact when
there exists a function f (x, y) such that
d(f (x, y)) = Mdx + Ndy,
or equivalently
∂f ∂f
dx + dy = Mdx + Ndy.
∂y ∂x
28.1.1 Theorem
The necessary and sufficient condition for the differential equation
Mdx + Ndy = 0 (28.1)
to be exact is
∂M ∂N
= . (28.2)
∂y ∂x
Proof: First we proof that the condition (28.2) is necessary. To prove we let the Equation
(28.1) to be exact. Then, by definition, there exists f (x, y) such that
∂f ∂f
dx + dy = Mdx + Ndy. (28.3)
∂y ∂x
Equating coefficients of dx and dy in Equation (28.3), we get

∂f
M= , (28.4)
∂y
∂f
N= . (28.5)
∂x
To eliminate the unknown f (x, y) from above equations, we assume that the 2nd order
partial derivatives of f are continuous. We now differentiate (28.4) and (28.5) w.r.t. x and
y respectively as
∂M ∂2f ∂N ∂2f
= , =
∂y ∂y∂x ∂x ∂y∂x
This implies
∂M ∂N
= .
∂y ∂x
Thus, if (28.1) is exact, M and N satisfy (28.2).
Now we show that the condition is sufficient. Suppose (28.2) holds and show that (28.1)
is exact. For this we find a function f (x, y) such that
d(f (x, y)) = Mdx + Ndy.

R ∂g
Let g(x, y) = Mdx be the partial integral of M such that = M . We first prove that
∂x
∂g
N− is function of y only. This is clear because
∂y
∂2g

∂ ∂g ∂N
N− = −
∂x ∂y ∂x ∂x∂y
∂2g ∂2g
Assuming = and using Equation (28.2) we get
∂x∂y ∂y∂x
∂2g

∂ ∂g ∂N
N− = −
∂x ∂y ∂x ∂y∂x
∂N ∂ ∂g ∂N ∂M
= − ( )= − = 0.
∂x ∂y ∂x ∂x ∂y
R ∂g
Take, f (x, y) = g(x, y) + (N − ∂y )dy . Hence taking total differentiation of this equation
gives
∂g ∂g ∂g ∂g
df = dg + (N − )dy = dx + dy + Ndy − dy,
∂y ∂x ∂y ∂y
∂g
= ( )dx + Ndy = Mdx + Ndy,
∂x
Thus, if Equation (28.2) is satisfied, Equation (28.1) is an exact equation.
28.2.1 Problem 1
Solve (x2 − 4xy − 2y 2)dx + (y 2 − 4xy − 2x2 )dy = 0 .

Solution: Comparing the given equation with Mdx + Ndy = 0, we have
M = (x2 − 4xy − 2y 2 ), N = (y 2 − 4xy − 2x2 )
Therefore
∂M ∂N
= −4x − 4y =
∂y ∂x
Hence, the given equation is exact and hence there exists a function f (x, y) such that
∂f ∂f
d(f (x, y)) = dx + dy = Mdx + Ndy
∂x ∂y
which implies
∂f ∂f
= M(x, y) and = N(x, y)
∂x ∂y
Integration of the first of above equations with respect to x gives
1
f = x3 − 2x2 y − 2y 2 x + c1 (y)
3
where c1 (y) is an arbitrary function of y only. Differentiating the above f with respect to
∂f
y and using = N(x, y) we get
∂y
∂f
= −2x2 − 4xy + c′1 (y) = +y 2 − 4xy − 2x2
∂y
This implies
y3
c′1 (y) = y 2 ⇒ c1 (y) = + c2
3
Hence the solution is given by
f (x, y) = c3 ⇒ x3 − 6xy(x + y) + y 3 = c
Here c2 , c3 and c are constants of integration.
28.2.2 Problem 2
Determine whether the differential equation (x + sin y)dx + (x cos y − 2y)dy = 0 is exact
and solve it.
Solution: For given equation we have
M(x, y) = (x + sin y) and N(x, y) = (x cos y − 2y) (28.6)
Now we check
∂M ∂N
= cos y =
∂y ∂x
Hence the given differential equation is exact. For the solution we seek a function f (x, y)
so that
∂f ∂f
= (x + sin y) and = (x cos y − 2y)
∂x ∂y
From the first relation we get
x2
f (x, y) = + x sin y + c1 (y)
2
Differentiating w.r.t. y and using the second relation of (28.6) we get
x cos y + c′1 (y) = x cos y − 2y ⇒ c′1 (y) = −2y ⇒ c1 (y) = −y 2 + c2
Therefore, we have
x2
f (x, y) = + x sin y − y 2 + c2
2
Then the solution of the given differential equation
x2
f (x, y) = c3 ⇒ + x sin y − y 2 = c.
2
28.2.3 Problem 3
Solve the differential equation (2y 2x − 2y 3)dx + (4y 3 − 6y 2 x + 2yx2 )dy

Solution: First we check the exactness of the equation by
∂M ∂N
= 4xy − 6y 2 =
∂y ∂x
So the equation is exact. Then, there exists a function f (x, y) such that
∂f ∂f
= (2y 2x − 2y 3 ) and = (4y 3 − 6y 2x + 2yx2 )
∂x ∂y
This gives
∂f
f (x, y) = (y 2 x2 − 2xy 3 ) + c1 (y) ⇒ = (2yx2 − 6xy 2) + c′1 (y)
∂y
This implies
c′1 (y) = 4y 3 ⇒ c1 (y) = y 4 + c2
Hence the solution is
f (x, y) = c3 ⇒ y 2x2 − 2xy 3 + y 4 = c.
28.2.4 Problem 4
Solve that the differential equation (3xy + y 2)dx + (x2 + xy)dy = 0. is not exact and hence
it cannot be solve by the method discussed above.
Solution: For the given differential equation we have
∂M ∂N
= 3x + 2y, and = 2x + y;
∂y ∂x
∂M ∂N
Since = , the given equation is not exact.
∂y ∂x
Now we see that it cannot be solved by the procedure described previously where we seek
a function f such that
∂f ∂f
= 3xy + y 2 and = x2 + xy (28.7)
∂x ∂y
Integration of the first relation gives

3
f (x, y) = x2 y + xy 2 + c1 (y)
2
where c1 (y) is an arbitrary function of y only. Now we differentiate the above equation
with respect to y and set the resulting expression equals to x2 +xy from the second relation
of (28.7) as
3 2
x + 2xy + c′1 (y) = x2 + xy
2
This provides
1
c′1 (y) = − x2 − xy
2
Since the right side of the above depends on x as well as on y , it is impossible to solve this
equation for c1 (y). Thus there is no f (x, y) exists and hence the given differential equation
cannot be solved in this way.
Suggested Readings
New Delhi.
Publisher, Moscow.
Lesson 29
Exact Differential Equations: Integrating Factors
In general, equations of the type M(x, y)dx + N(x, y)dy = 0 are not exact. However, it
is sometimes possible to transform the equation into an exact differential equation multi-
plying it by a suitable function I(x, y). That is, if I(x, y) is an integrating factor then the
differential equation
I(x, y)M(x, y)dx + I(x, y)N(x, y)dy = 0
becomes exact. A solution to the above equation is obtained by solving the exact differ-
ential equation as in the previous lesson. Note that the given equation may have several
integrating factors. This is exactly the procedure we have used for solving linear differ-
ential equations in earlier lesson. Here we deal with more general differential equation.
29.1 Rule I: By Inspection
There is not much theory behind finding integrating factor by inspection. This method
works based on recognition of some standard exact differentials that occur frequently in
practice. The following list of exact differentials would be quite useful in solving exact
differential equations:
(i) d(xy) = ydx + xdy

y xdy − ydx
x ydx − xdy
(ii) d = or d =
x x2 y y2
y xdy − ydx
x ydx − xdy
(iii) d ln = or d ln =
x xy y xy

y xdy − ydx x ydx − xdy
(iv) d arctan = 2 2
or d arctan =
x x +y y y 2 + x2
ydx + xdy
(v) d (ln xy) =
xy
29.1.1 Example
Solve the differential equation y(y 2 + 1)dx + x(y 2 − 1)dy .
Solution: The given equation can be rewritten as
y 2(ydx + xdy) + ydx − xdy
This is further rewritten as

ydx − xdy
(ydx + xdy) + =0
y2
Using standard differential forms given above we get

x
d(xy) + d =0
y
Integrating the above equation, the desired solution is given as
xy 2 + x = cy
Here c is an arbitrary constant.
29.2 Rule II: Mdx + N dy = 0 is homogeneous and Mx + N y 6= 0
1
If the equation Mdx+Ndy = 0 is homogeneous and Mx+Ny 6= 0, then I(x, y) =
(Mx + Ny)
is an integrating factor. In order to prove the result, we need to show that
Mdx + Ndy
= d (some function x and y)
Mx + Ny
Rewriting Mdx + Ndy as

1 dx dy dx dy
Mdx + Ndy = (Mx + Ny) + + (Mx − Ny) −
2 x y x y
Multiplying by proposed integrating factor we get

Mdx + Ndy 1 dx dy (Mx − Ny) dx dy
= + + − (29.1)
Mx + Ny 2 x y (Mx + Ny) x y
Given that M(x, y) and N(x, y) are homogeneous functions of some degree n, i.e., M(tx, ty) =
tn M(x, y) and N(x, y) = tn N(x, y). Then

x 1 1 1 x
M ,1 =M x, y = n M(x, y) ⇒ M(x, y) = y n M ,1
y y y y y
Similarly, we get
n x
N(x, y) = y N ,1
y
Now consider

x x x x x
(Mx − Ny) y nxM − y,1 y nyN y,1 yM y,1

x −N y,1
= = =f
(Mx + Ny) n x n x
y xM y , 1 + y yN y , 1 x x x y
yM y,1 + N y,1
Going back to the Equation (29.1), we have

Mdx + Ndy 1 x x
= d (ln(xy)) + f d ln
Mx + Ny 2 y y
Rewriting f (x/y) = f (exp(ln(x/y))) and defining g(x) := f (exp(x)), the above equation
becomes

Mdx + Ndy 1 x
= d (ln(xy)) + g (ln(x/y)) d ln
Mx + Ny 2 y
Hence, we have shown that

Z
Mdx + Ndy 1 1 x x
=d ln(xy) + g ln d ln
Mx + Ny 2 2 y y
1
Thus is an integrating factor of the homogenous differential equation Mdx +
Mx + Ny
Ndy = 0.
29.2.1 Example
Solve the differential equation (x2 y − 2xy 2 )dx − (x3 − 3x2 y)dy = 0
Solution: The given equation is a homogeneous differential equation. Comparing it with
Mdx + Ndy = 0, we have M = x2 y − 2xy 2 and N = −(x3 − 3x2 y). Since
Mx + Ny = (x2 y − 2xy 2 )x − y(x3 − 3x2 y) = x2 y 2 6= 0,
the integrating factor is

1 1
= 2 2
(Mx + Ny) x y
Multiply by the integrating factor, the given differential equation becomes
(1/y − 2/x)dx − (x/y 2 − 3/y)dy = 0
This is now exact and can be rewritten as

ydx − xdy 2 3 x 2 3
− dx + dy = 0 ⇒ d − dx + dy = 0
y2 x y y x y
Integrating the above equation we obtain the desired solution as
x − 2y ln x + 3y ln y = cy
29.3 Rule III: Mdx + N dy = 0 is of the form f1(xy)ydx + f2(xy)xdy = 0
1
If the equation Mdx + Ndy = 0 is of the form f1 (xy)ydx + f2(xy)xdy = 0, then
(Mx − Ny)
is an integrating factor provided Mx − Ny 6= 0. Similar to rule II we now show that
Mdx + Ndy
= d (some function x and y)
Mx − Ny
Again, rewriting Mdx + Ndy as

1 dx dy dx dy
Mdx + Ndy = (Mx + Ny) + + (Mx − Ny) −
2 x y x y
Now dividing by Mx − Ny we get

Mdx + Ndy 1 (Mx + Ny) dx dy dx dy
= + + −
Mx − Ny 2 Mx − Ny x y x y
Using M = f1 (xy)y and N = f2 (xy)x we obtain

Mdx + Ndy 1 f1 (xy) + f2 (xy) x
= d (ln xy) + d ln
Mx − Ny 2 f1 (xy) − f2 (xy) y
f1 (xy) + f2 (xy)
Let f (xy) := and g(x) := f (exp(x)), the above equation reduces to
f1 (xy) − f2 (xy)

Mdx + Ndy 1 x 1 x
= f (xy)d (ln xy) + d ln = g(ln xy)d (ln xy) + d ln
Mx − Ny 2 y 2 y
This shows that

Z
Mdx + Ndy 1 1 x
=d g(ln xy)d (ln xy) + ln
Mx − Ny 2 2 y
29.3.1 Example
Solve y(x2 y 2 + 2)dx + x(2 − 2x2 y 2)dy = 0.

Solution: Comparing with Mdx+Ndy = 0, we have M = y(x2 y 2 +2) and N = x(2−2x2 y 2 ).
The given equation is of the form
f1 (xy)ydx + f2 (xy)xdy = 0
and we have
Mx − Ny = xy(x2 y 2 + 2) − xy(2 − 2x2 y 2 ) = 3x3 y 3 6= 0
Therefore, multiplying the equation by 1/3x3 y 3 , we obtain
(1/3x + 2/(3x3 y 2))dx + (2/(3x2 y 3) − 2/3y)dy = 0
This is an exact differential equation which can be solved with the technique discussed in
previous lesson.
29.4 Rule IV: Most general approach
Now we discuss the most general approach of finding integrating function. The idea is to
multiply the given differential equation
M(x, y)dx + N(x, y)dy = 0 (29.2)
by a function I(x, y) and then try to choose I(x, y) so that the resulting equation
I(x, y)M(x, y)dx + I(x, y)N(x, y)dy = 0 (29.3)
becomes exact. The above equation is exact if and only if

∂(IM) ∂(IN)
= (29.4)
∂y ∂x
If a function I(x, y) satisfying the partial differential Equation (29.4) can be found, then
(29.3) will be exact. Unfortunately, solving Equation (29.4), is as difficult to solve as the
original Equation (29.2) by some other methods. Therefore, while in principle integrating
factors are powerful tools for solving differential equations, in practice they can be found
only in special cases. The cases we will consider are: (i) an integrating factor I that is
either as function of x only, or (ii) a function of y only.
Let us determine necessary conditions on M and N so that (29.2) has an integrating factor
I that depends on x only. Assuming that I is a function of x only, then Equation (29.4)
reduces to
dI dI IMy − INx
IMy = INx + N ⇒ = (29.5)
dx dx N
If (My − Nx )/N is a function of x only, say f(x), then there is an integrating factor I
R
that also depends only on x which can be found by solving (29.5) as I(x) = e f (x)dx . A
similar procedure can be used to determine a condition under which Equation (29.2) has
an integrating factor depending only on y . To conclude, we have:
R

1 ∂M ∂N
If − is function of x alone say f (x), then I(x) = e f (x)dx is an I.F.
N ∂y ∂x
1 ∂N ∂M R
If − is function of y alone say f (y), then I(y) = e f (y)dy is an I.F.
M ∂x ∂y
29.5.1 Problem 1
Find an integrating factor of (x2 + y 2 + x)dx + xydy = 0 Solution: Comparing with

Mdx + Ndy = 0, we have
M = (x2 + y 2 + x) and N = xy
Further, note that

1 ∂M ∂N 1
− =
N ∂y ∂x x
R
1/xdx
is a function of x alone. Hence, the integrating factor of the given problem is e = x.
29.5.2 Problem 2
Find an integrating factor of (2xy 4 ey + 2xy 3 + y)dx + (x2 y 4 ey − x2 y 2 − 3x)dy = 0

Solution: Compare with Mdx + Ndy = 0, we get
M = (2xy 4 ey + 2xy 3 + y) and N = (x2 y 4 ey − x2 y 2 − 3x)
Also, note that

1 ∂N ∂M 4
− =−
M ∂y ∂x y
R
is a function of y alone. Hence the integrating factor of the given problem is e −4/ydy
= 1/y 4.
Suggested Readings
New Delhi.
Publisher, Moscow.
Lesson 30
Linear Differential Equations of Higher Order
In this lesson we discuss linear differential equation of higher order with constant coeffi-
cients. In particular, we shall learn about the techniques of finding solutions of homoge-
nous equations. Different cases will be considered with the help of several examples.
30.1 Linear Differential Equation
In a linear differential equation, the dependent variable and its differential coefficients
occur only in the first degree and are not multiplied together. The general form of the
equation is
dn y dn−1 y dn−2 y
+ a1 (x) + a2 (x) + . . . + an (x)y = F (x), (30.1)
dxn dxn−1 dxn−2
where a1 , a2 , . . . , an and F are either constants or functions of x only. If the right hand
side, i.e. F (x), is identically zero, the equation is said to be homogeneous; otherwise it is
called nonhomogeneous. Before we discuss some particular cases of the above equation
we state two facts about the solution of a linear homogeneous differential equation. The
first says that if we know n solutions y1 , y2 , . . . , yn of the linear homogeneous equation,
then any linear combination y = c1 y1 + c2 y2 + . . . cn yn is also a solution for any constants
c1 , c2 , . . . , cn . This can easily be proved by substituting y = c1 y1 + c2 y2 + . . . cn yn into
the equation and using linearity of the equation. The second important result concerns
about the general solutions (solution containing all solutions) to the linear homogeneous
equation. This result says that any solution is some linear combination of y1 , y2 , . . . , yn for
some suitable values of constants c1 , c2 , . . . , cn . However, this is not true for any combina-
tion of solutions but is true if the solutions y1 , y2 , . . . , yn are linearly independent.
30.2 Linear Differential Equation with Constant Coefficients
An equation of the form

dn y dn−1 y dn−2 y
+ a1 + a2 + . . . + an y = F (x), (30.2)
dxn dxn−1 dxn−2
Linear Differential Equation of Higher Order
where a1 , a2 , . . . , an are constants, is called linear differential equation with constant coef-
dn
ficients. Using the symbols Dn := , the Equation (30.2) becomes
dxn
(D n + a1 D n−1 + a2 D n−2 + ... + an )y = F (x), (30.3)
Further defining f (D) := Dn + a1 Dn−1 + a2 Dn−2 + ... + an , we can rewrite the given
differential equation in a more compact form as f (D)y = F (x). Here f (D) acts as oper-
ator on y to yield F (x). The general solution of (30.2) can be written as the sum of the
general solution of the corresponding homogeneous equation, refereed as complimentary
function (C.F.), and a particular solution or sometimes called particular integral (P.I) of
nonhomogeneous equation. Thus
y = C.F. + P.I. (30.4)
Note that the C.F. involves n arbitrary constants and P.I. does not involve any arbitrary
constant. It is readily evident that y in (30.4) is the general solution of the given non-
homogeneous differential equation because it satisfies the given differential equation as
f (D) (C.F. + P.I.) = f (D) (C.F.) + f (D) (P.I.) = 0 + F (x) and it has n arbitrary constants.
30.3 C.F. of a Differential Equation
By definition, C.F. of (30.2) is the general solution of

(D n + a1 D n−1 + a2 D n−2 + ... + an )y = 0 (30.5)
To solve Equation (30.5), we seek a function which satisfies the above equation. One
intelligent guess of such a function is the exponential function emx , where m is a constant.
Differentiations of this exponential function are just constant multiples of the original
exponential. If we substitute this function into the Equation (30.5), we obtain
(mn + a1 mn−1 + a2 mn−2 + ... + an )emx = 0 (30.6)
Since the exponential function is never zero, we can divide this last equation by emx .
Thus, y = emx is a solution to Equation (30.5) if and only if m is a solution to the algebraic
equation
mn + a1 mn−1 + a2 mn−2 + ... + an = 0 (30.7)
Equation (30.7) is called the auxiliary equation (A.E.) or characteristic equation (C.E.) of
the differential Equation (30.5).
30.4 Case I: A.E. has real and distinct roots
If m1 , m2 , m3 , ..., mn be real and distinct then the solutions em1 x , emm x , . . . , emn x are lin-
early independent and the general solution of the given homogeneous differential equation
becomes
y = c1 em1 x + c2 em2 x + c3 em3 x + ... + cn emn x ,
where c1 , c2 , ..., cn are arbitrary constants.
30.4.1 Example
Find the general solution of the differential equation (D3 + 6D2 + 11D + 6)y = 0.
Solution: The A.E. is (m3 + 6m2 + 11m + 6) = 0. The roots are m = −1, −2, −3. Hence
the required solution is y = c1 e−x + c2 e−2x + c3 e−3x .
30.5 Case II: A.E. has repeated real roots
Let m1 = m2 are repeated roots of the A.E. Then, we have n − 1 linearly independent
solutions. It can be shown that a simple choice y = xem1 x is also a solution which is
independent to the rest n − 1 solutions. Thus, the general solution of the given differential
equation is given by
y = (c1 + c2 x)em1 x + c3 em3 x + ... + cn emn x
The above idea can be further extended by taking solutions xem1 x , x2 em1 x , . . . , xl−1 em1 x . . .
if the root m1 is repeating l−times.
30.5.1 Example
Find the general solution to (D4 + 2D3 − 3D2 − 4D + 4)y = 0.

Solution: The A.E. of given equation is
(m4 + 2m3 − 3m2 − 4m + 4) = 0
The roots of the A.E. are m = 1, 1, −2, −2. The required solution is y = (c1 + c2 x)ex + (c3 +
c4 x)e−2x .
30.6 Case III: A.E. has complex roots
If m1 = α + iβ and m2 = α − iβ , then the solutions em1 x , em2 x , . . . , emn x are linearly

independent and the general solution of the given homogeneous differential equation is
given by
y = c′1 em1 x + c′2 em2 x + c3 em3 x + ... + cn emn x .
The above solution can be simplified as
y = c′1 eαx (cos βx + i sin βx) + c′2 eαx (cos βx − i sin βx) + c3 em3 x + ... + cn emn x .
Defining new constants c1 = c′1 + c′2 and c2 = i(c1 − c2 ), the general solution becomes
y = eαx (c1 cos βx + c2 sin βx) + c3 em3 x + ... + cn emn x .
Similar to the case II, the solution for repeated complex roots can be found, see example
below.
30.6.1 Example
Find the general solution to the differential equation (D2 + 1)2 y = 0.

Solution: The A.E. and its roots are
(m2 + 1)2 = 0, and therefore m = ±i, ±i.
This is the case of repeated complex root, so case II and case III can be combined to give
the desired solution as y = (c1 + c2 x) cos x + (c3 + c4 x) sin x.
30.7 Miscellaneous Problems
30.7.1 Problem 1
Find the general solution of the differential equation (D3 + 3D2 + 3D + 1)y = 0.
Solution: The A.E. and its root are given by (m + 1)3 = 0 and m = −1, −1, −1. Therefore,
the required solution is y = (c1 + c2 x + c3 x2 )e−x .
30.7.2 Problem 2
Find the general solution of (D3 − 8)y = 0.

√
Solution: The A.E. of the given equation is (m3 − 8) = 0. Its root are m = 2, −1 ± i 3.
√ √
The required solution is y = c1 e2x + e−x (c2 cos 3x + c3 sin 3x).
30.7.3 Problem 3
Find the general solution of the differential equation (D2 − 2D + 5)2 y = 0.

Solution: The auxiliary equation is (m2 − 2m + 5)2 = 0. Its roots are m = 1 ± 2i, 1 ± 2i
Hence the required solution is y = ex [(c1 + c2 x) cos 2x + (c3 + c4 x) sin 2x] .
30.7.4 Problem 4
Find the general solution of (D2 + D + 1)2 (D − 2)y = 0.

Solution: The A.E. of the given equation is (m2 + m + 1)2 (m − 2) = 0. Its roots are
√ √
m = − 12 ± i 23 , − 12 ± i 23 , 2. Hence, the desired solution is
" √ √ #
− 21 x 3 3
y = c1 e2x + e (c2 + c3 x) cos x + (c4 + c5 x) sin x
2 2
30.7.5 Problem 5
Find the general solution of the differential equation (D2 + 1)3 (D2 + D + 1)2 y = 0.
Solution: The A.E. of given equation is
(D 2 + 1)3 (D 2 + D + 1)2 = 0,
√ √
1 3 1 3
The roots are m = ±i, ±i, ±i, − ± i , − ± i . Therefore, the desired solution is
2 2 2 2
y =(c1 + c2 x + c3 x2 ) cos x + (c4 + c5 x + c6 x2 ) sin x

" √ √ #
1 3 3
+ e− 2 x (c7 + c8 x) cos x + (c9 + c10 x) sin x
2 2
Suggested Readings
New Delhi.
Publisher, Moscow.
Lesson 31
In connection to the last lesson, we discuss solution methodologies of getting particular

integral of the linear differential equations of higher order. In particular, in this lesson
we present operator method which is somewhat easier than other methods for finding
particular integrals.
31.1 Determination of Particular Integral (P.I.)
As we have seen in the earlier lesson that a general nonhomogeneous linear differential
equations with constant coefficients can be written in operator form as f (D)y = F (x).
The operator, 1/f (D) is called inverse operator which gives a particular integral when
operated on both the sides of the given differential equation. Hence, a particular integral
1
of the given differential equation is given as f (D) F (x). First we give a rather general idea
of getting a particular integral with this method and then state some other useful direct
results. Note that the operator f (D) can be expressed as (D − α1 )(D − α2 ) . . . (D − αn ) and
thus a particular integral is given as
1 1 1 1
F (x) = ... F (x) (31.1)
f (D) D − α1 D − α2 D − αn
1
We give a general idea of evaluating an expression of the type F (x). This procedure
D−α
can be repeatedly applied to find a particular integral (31.1). However, applicability of this
method depends upon the form of F (x).
We give a general theorem that can be applied to any problem for finding particular inte-
gral of a differential equation.
31.1.1 Theorem 1
If F (x) is function of x and α is a constant, then

1
Z
F (x) = eαx F (x)e−αx dx.
D−α
Proof: Let us assume that

1
y= F (x)
D−α
On operating (D − α) both sides, we get
dy
(D − α)y = F (x) ⇒ − αy = F (x)
dx
The above equation is a linear differential equation of first order whose integrating factor
R
is e− αdx = e−αx . Hence, the solution is given by
Z Z
αx
ye−αx
= F (x)e
−αx
dx ⇒ y=e F (x)e−αx dx
Since our interest is finding a particular integrals, the constant of integration is dropped.
Thus,
1
Z
F (x) = eαx F (x)e−αx dx.
D−α
Now we state some useful result those will be used to find P.I. of certain special forms of
F (x).
31.1.2 Theorem 2
If α is a constant, then f (D)eαx = f (α)eαx

Proof: We know that Deαx = αeαx and similarly D2 eαx = α2 eαx . With induction we can
prove that Dn eαx = αn eαx for any natural number n. This proves the result f (D)eαx =
f (α)eαx .
31.1.3 Theorem 3
If α is a constant and g(x) is any function, then f (D) (eαx g(x)) = eαx f (D + α)g(x)
Proof: We know that D (eαx g(x)) = αeαx g(x) + eαx Dg(x) = eαx (α + D)g(x). Similar to the
proof of previous theorem we can prove with induction that Dn eαx g(x) = eαx (α + D)n g(x)
for any natural number n. This proves the result f (D) (eαx g(x)) = eαx f (D + α)g(x). This
result is known as shifting property of operator f (D).
31.1.4 Theorem 4
If α and β are arbitrary constants, then
f (D 2 ) sin(αx + β) = f (−α2 ) sin(αx + β) and f (D 2) cos(αx + β) = f (−α2 ) cos(αx + β)
Proof: It can easily be verified that D2 sin(αx + β) = −α2 sin(αx + β) and D2 cos(αx + β) =
−α2 cos(αx + β). In other words, we can replace D 2 by −α2 and this proves the desired
result.
Now we describe the method for some special form of F (x).
31.2 Rule I: F (x) is of the form eax
We know from Theorem 31.1.2 that f (D)eαx = f (α)eαx . Operating on both sides by
1/f (D) we get
1 1 αx
eαx = f (α)eαx ⇒ eαx = f (α) e
f (D) f (D)
This implies that
1 αx 1 αx
e = e , provided f (α) 6= 0
f (D) f (α)
If f (α) = 0, then (D − α) is a factor of f (D), say f (D) = (D − α)g(D). Then
1 αx 1 1 αx 1 1 αx
e = e = e provided g(α) 6= 0
f (D) (D − α) g(D) (D − α) g(α)
Now using Theorem 31.1.1, we get

1 αx 1 1 1 αx
e = eαx = e x
f (D) g(α) (D − α) g(α)
In case g(α) = 0 then , say f (D) = (D − α)2 h(D). In this case we get
1 αx 1 1 αx 1 x2 αx
e = e = e provided h(α) 6= 0
f (D) h(α) (D − α)2 g(α) 2!
Again, if h(α) = 0, the same procedure can be repeated. To conclude, we have the follow-
ing results:
1 αx 1 αx
(i) e = e , where f (α) 6= 0
f (D) f (α)
(ii) If f (α) = 0, then f (D) must posses a factor of the type (D − α)r , say f (D) =
(D − α)r g(D) where g(α) 6= 0. Then the following formula is applicable
1 αx xr αx
e = e
(D − α)r r!
.
31.3.1 Problem 1
Find the general solution of the differential equation (D2 − 3D + 2)y = e3x .
Solution: The auxiliary equation is
(m2 − 3m + 2) = 0 ⇒ (m − 1)(m − 2) = 0 ⇒ m = 1, 2.
The complimentary function is given as

C.F. = c1 ex + c2 e2x
The particular integral is

1 3x 1 3x 1 3x
P.I. = e = e = e .
D 2 − 3D + 2 32 − 3.3 + 2 2
1
The general solution is: y = c1 ex + c2 e2x + e3x .
2
31.3.2 Problem 2
Solve (4D2 − 12D + 9)y = 144e3x/2

Solution: The auxiliary equation is
(4m2 − 12m + 9) = 0 ⇒ m = 3/2, 3/2.
The complimentary function is

C.F. = (c1 + c2 x)e3x/2

144 3x/2 144 1 3x/2 x2 3x/2
P.I. = e = e = 36 e
(2D − 3)2 4 (D − 3/2)2 2!
2
The required solution is: y = (c1 + c2 x)e3x/2 + 36 x2! e3x/2 .
31.4 Rule II: F (x) is of the form cos ax or sin ax
We express f (D) as a function of D2 , say f (D) = φ(D2 ). From Theorem 31.1.4 we know
that φ(D2 ) sin(αx + β) = φ(−α2 ) sin(αx + β). Applying [φ(D2 )]−1 both sides we obtain
1
sin(αx + β) = 2
φ(−α2 ) sin(αx + β)
φ(D )
If φ(−α2 ) 6= 0, we can divide the above equation by φ(−α2 ) to get

1 1
2
sin(αx + β) = sin(αx + β)
φ(D ) φ(−α2 )
Similarly,
1 1
2
cos(αx + β) = cos(αx + β), provided φ(−α2 ) 6= 0
φ(D ) φ(−α2 )
In case, φ(−α2 ) = 0, we can rewrite sin(αx + β) = Im(ei(αx+β) ) and cos(αx + β) =

Re(ei(αx+β) ). Now case I can be applied as

1 1 i(αx+β) 1 i(αx+β)
sin(αx + β) = Im e = Im e provided f (iα) 6= 0
f (D) f (D) f (iα)
Similarly,

1 1 i(αx+β)
cos(αx + β) = Re e provided f (iα) 6= 0
f (D) f (iα)
31.5.1 Problem 1
Solve the differential equation (D2 + 1)y = cos 2x.

Solution: The characteristic equation of the corresponding homogeneous equation is
(m2 + 1) = 0 ⇒ m = ±i
Hence, C.F. = (c1 cos x + c2 sin x). The particular integral is given by
1 1 1
P.I. = cos 2x = cos 2x = cos 2x.
D2 +1 2
(−2 + 1) −3
1
The required solution is: y = (c1 cos x + c2 sin x) − cos 2x.
3
31.5.2 Problem 2
Solve the differential equation (D2 − 4D + 3)y = sin x.

Solution: The roots of the characteristic equations are 1 and 3. The complementary
function is C.F. = c1 ex + c2 e3x . The particular integral is
1
P.I. = sin x
D2 − 4D + 3
Replacing D2 by −1, we get
1 1 1 1 1 + 2D
P.I. = sin x = sin x = sin x
2 − 4D 2 1 − 2D 2 1 − 4D 2
Again, replacing D2 by −1, we obtain
1 1
P.I. = (1 + 2D) sin x = (sin x + 2 cos x)
10 10
Hence the complete solution is
1
y = c1 ex + c2 e3x + (sin x + 2 cos x),
10
where c1 and c2 are arbitrary constants.
Suggested Readings
Lesson 32
Linear Differential Equation of Higher Order (Cont.)
Here we continue discussion for solving linear equation of the form f (D)y = F (x). In the
last lesson, we have found particular integral for two different types of functions F (x).
In this lesson we shall continue discussing various other situations for finding particular
integral.
32.1 Rule III: F (x) is a polynomial of degree l
Take out the lowest degree term from f (D), so as to reduce it in the form [1 ± f (D)]n . Take
it to numerator, i.e., [1 ± f (D)]−n and expand it in ascending powers of D with the help of
Binomial series:
α(α − 1) 2 α(α − 1)(α − 2) 3

(1 + x)α = 1 + αx + x + x + ...
2! 3!
Note that in the expansion we do not need to consider terms with power more than l, since
l + 1th and higher order derivatives of the polynomial of degree l will be zero.
32.1.1 Example
Solve the differential equation (D2 + D)y = x2 + 2x + 4

Solution: The characteristic equation of the corresponding homogeneous equation is
(m2 + m) = 0 ⇒ m = 0, −1.
The complementary function is c1 + c2 e−x . The particular integral is

1 1 1
P.I. = x2 + 2x + 4 = x2 + 2x + 4
D2 +D D (1 + D)
Taking 1 + D into numerator and expending this into an infinite series we get
1 1
P.I. = (1 − D + D 2 − D 3 + ...)(x2 + 2x + 4) = (x2 + 2x + 4 − 2x − 2 + 2)
D D
1
Operating 1/D on each term, we obtain P.I. = (x2 + 4) = (x3 /3 + 4x). The desired gen-
D
eral solution is 3
x
y = c1 + c2 e−x + + 4x .
3
32.2 Rule IV: F (x) is of the form eαx V , where V is any function of x
Using shift property of the operator discussed in the last lesson we can easily prove that
1 αx 1
e V = eαx V.
f (D) f (D + α)
32.2.1 Example
Solve (D2 − 2D + 1)y = x2 ex .

Solution: The characteristic equation and its roots are
m2 − 2m + 1 = 0, and m = 1, 1.
Thus, the complimentary function is
C.F. = (c1 + c2 x)ex

1 1
P.I. = x2 ex = x2 ex
D2 − 2D + 1 (D − 1)2
Using shift property we get

1
P.I. = ex 2
x2 ,
(D − 1 + 1)

x 1 2 x 1 1 2
= e 2x = e x
D D D
3
x 1 x x4
=e = ex .
D 3 12
x4
The required solution is y = (c1 + c2 x)ex + ex .
12
32.3 Rule V: F (x) is of the form xV , where V is any function of x
Here, we prove the following result

1 1 d 1
(xV ) = x V + V
f (D) f (D) dD f (D)
where V is a function of x. We start with the fact that for a given function g(x) we have
D (xg(x)) = xD (g(x)) + g(x)
Which can be rewritten as

d
D (xg(x)) = xD (g(x)) + D (g(x))
dD
Operating D once more and after simplifications we obtain

2 2 d 2
D (xg(x)) = xD (g(x)) + D (g(x))
dD
In general, by the method of induction for any natural number n we can show that

n n d n
D (xg(x)) = xD (g(x)) + D g(x)
dD
Direct implication of the above result leads

d
f (D) (xg(x)) = xf (D) (g(x)) + f (D) g(x) (32.1)
dD
Let us assume that f (D)g(x) = V (x) so that we have

1
g(x) = V (x)
f (D)
Substituting g(x) in Equation (32.1) we get

1 1 d 1
f (D) x V (x) = xf (D) V (x) + f (D) V (x)
or

1 d 1
f (D) x V (x) = xV (x) + f (D) V (x)
f (D) dD f (D)
This implies

1 d 1
xV (x) = f (D) x V (x) − f (D) V (x)
f (D) dD f (D)
Operating the above equation by 1/f (D) we get

1 1 d 1
(xV (x)) = x V (x) − f (D) V (x)
f (D) f (D) dD f (D)2
Equivalently, we have the final result

1 1 d 1
(xV (x)) = x V (x) + V (x)
32.3.1 Example
Solve (D2 + 9)y = x sin x

Solution: The roots of the characteristic equations are ±3i. Hence, the complimentary
function is given by
C.F. = (c1 cos 3x + c2 sin 3x)

1
P.I. = x sin x
D2 +9
Using Rule V, we get

1 d 1
P.I. = x 2 sin x + 2
sin x
D +9 dD D +9
This can be now evaluated as

1 2D 1 2D 1 1
P.I. =x sin x − sin x = x sin x − sin x = x sin x − cos x
8 (D 2 + 9)2 8 64 8 32
The required general solution is

1 1
y = (c1 cos 3x + c2 sin 3x) + x sin x − cos x
8 32
32.4 Rule VI: F (x) is of the form xm sin αx or xm cos αx
In this case Rule IV or Rule V can be applied. For the application of rule IV we should
note that

1 m 1 m iαx
1. x sin αx = Im x e
f (D) f (D)

1 m 1 m iαx
2. x cos αx = Re x e .
f (D) f (D)
32.4.1 Example
Find a particular integral of (D2 + 1)y = x2 sin 2x

Solution The particular integral is
1 2 1
P.I. = 2
x sin 2x = Im 2
x2 e2ix
D +1 D +1
Applying Rule IV, we get the particular integral as

2ix 1 2 2ix 1 2
P.I. = Im e x = Im e x
(D + 2i)2 + 1 D 2 + 4Di − 3
−1 !
e2ix 4iD D 2

P.I. = Im 1− + x2
−3 3 3
Using the Binomial expansion, we get

2 !
e2ix 4iD D 2 4iD D 2

= Im 1+ + + + ... x2
−3 3 3 3 3
2ix
4iD D 2 16D 2

e 2
= Im 1+ + − + ... x
−3 3 3 9

1 2 8ix 26
= Im − (cos 2x + i sin 2x) x − −
3 3 9
Collecting the imaginary part we have

1 2 26 8
P.I. = − x − sin 2x + x cos 2x .
3 9 3
Suggested Readings
New Delhi.
Publisher, Moscow.
Lesson 33
Method of Undetermined Coefficients
In the last lesson we have discussed operator method of finding particular integral. In
this lesson we lean method of undetermined coefficients for finding particular integral of
non-homogeneous differential equations. This method is relatively easier to apply once a
possible form of a particular integral is known. This method is mainly applicable to linear
differential equations with constant coefficients.
33.1 Method of Undetermined Coefficients
The method of undetermined coefficients requires that we make an initial assumption

about the form of a particular solution of the differential equation, but with the coeffi-
cients left unspecified. We then substitute the assumed expression into the given differ-
ential equation and attempt to determine the coefficients so as to satisfy that differential
equation. If we are successful, then we have found a particular solution of the differential
equation. If we cannot determine the coefficients, then this means that there is no solution
of the form that we assumed. In this case we may modify the initial assumption and try
again.
The main advantage of the method of undetermined coefficients is that it is straightfor-

ward to execute once the assumption is made as to the form of the particular solution. Its
major limitation is that it is useful primarily for equations for which we can easily write
down the correct form of the particular solution in advance. This method is usually used
only for problems in which the homogeneous equation has constant coefficients and the
nonhomogeneous term is restricted to a relatively small class of functions. In particular,
we consider only nonhomogeneous terms that consist of polynomials, exponential func-
tions, sines, and cosines. Despite this limitation, the method of undetermined coefficients
is useful for solving many problems that have important applications.
We shall demonstrate the method by taking a couple of different examples.
33.2.1 Problem 1
Solve the following initial value problem
y ′′ + 5y ′ + 6y = 2x + 1 (33.1)
with the initial conditions y(0) = 0 and y ′(0) = 31 .

Solution: First we solve the corresponding homogeneous equation. The characteristic
equation is
m2 + 5m + 6 ⇒ m = −2, −3.
Hence the complementary function is
C.F. = C1 e−2x + C2 e−3x .
To find particular integral, the trick is to somehow to guess one particular solution to
Equation (33.1). Note that 2x + 1 is a polynomial, and the left hand side of the equation
will be a polynomial if we let y be a polynomial of the same degree. Let us try
yp = Ax + B
We plug in to the differential equation to obtain
yp′′ + 5yp′ + 6yp =(Ax + B)′′ + 5(Ax + B)′ + 6(Ax + B)

=0 + 5A + 6Ax + 6B = 6Ax + (5A + 6B).
So 6Ax + (5A + 6B) = 2x + 1. Therefore,

1 −1
A= and B =
3 9
That means
1 1 3x − 1
yp = x− =
3 9 9
Hence the general solution to (33.1) is
3x − 1
y = C1 e−2x + C2 e−3x + .
9
The general solution must satisfy the given initial conditions. First find
1
y ′ = −2C1 e−2x − 3C2 e−3x +
3
Then
1 1 1
0 = y(0) = C1 + C2 − , = y ′ (0) = −2C1 − 3C2 + .
9 3 3
We solve to get C1 = 1/3 and C2 = −2/9. The particular solution we want is
1 2 3x − 1 3e−2x − 2e−3x + 3x − 1
y(x) = e−2x − e−3x + = .
3 9 9 9
33.2.2 Problem 2
Find a particular solution of the differential equation
y ′′ + 2y ′ + 2y = cos(2x).
Solution: We start by guessing the solution that includes some multiple of cos(2x). We
may have to also add a multiple of sin(2x) to our guess since derivatives of cosine are
sines. We try
yp = A cos(2x) + B sin(2x).
We plug yp into the equation and we get
−4A cos(2x) − 4B sin(2x) − 4A sin(2x) + 4B cos(2x) + 2A cos(2x) + 2B sin(2x) = cos(2x).
The left hand side must equal to right hand side. We group terms and get −4A+4B +2A =
1 and −4B − 4A + 2B = 0. So −2A + 4B = 1 and 2A + B = 0 and hence A = −1 10 and
1
B = 5 . Hence a particular solution is
− cos(2x) + 2 sin(2x)
yp = A cos(2x) + B sin(2x) = .
10
Remark 1: If the right hand side contains exponentials we try exponentials. For
example, for
Ly = e3x ,
we will try y = Ae3x as our guess and try to solve for A.
Remark 2: If the right hand side is a multiple of sines, cosines, exponentials, and
polynomials, we can use the product rule for differentiation to come up with a guess. We
need to guess a form for yp such that Lyp is of the same form, and has all the terms needed
to for the right hand side. For example,
Ly = (1 + 3x2 ) e−x cos(πx).
For this equation, we will guess
yp = (A + Bx + Cx2 ) e−x cos(πx) + (D + Ex + F x2 ) e−x sin(πx).
We will plug in and then hopefully get equations that we can solve for A, B, C, D, E, and
F.
Remark 3: If the right hand side has several terms, such as

Ly = e2x + cos x.
In this case we find u that solves Lu = e2x and v that solves Lv = cos x (that is, do each
term separately). Then note that if y = u + v , then Ly = e2x + cos x. This is because L is
linear; we have Ly = L(u + v) = Lu + Lv = e2x + cos x.
33.2.3 Problem 3
Find a particular solution of
y ′′ − 3y ′ − 4y = 3e2t + 2 sin t − 8et cos 2t.
Solution: By splitting up the right side of the given differential equation, we obtain the
three differential equations
y ′′ − 3y ′ − 4y = 3e2t , y ′′ − 3y ′ − 4y = 2 sin t, y ′′ − 3y ′ − 4y = 8et cos 2t
Solutions of these three equations can be found with appropriate guess of the particular
integral discussed above. Finally, a particular solution is their sum, namely,
1 3 5 10 2
Y (t) = e2t + cos t sin t + et cos2t + et sin 2t.
2 17 17 13 13
The procedure illustrated in these examples enables us to solve a large class of problems
in a reasonably efficient manner. However, there is one difficulty that sometimes occurs.
It could be that our guess actually solves the associated homogeneous equation. The next
example illustrates how it arises.
33.2.4 Problem 4
Solve the following differential equation
y ′′ − 9y = e3x
Solution: In order to find a particular integral an intelligent guess would be y = Ae3x , but
if we plug this into the left hand side of the equation we get
y ′′ − 9y = 9Ae3x − 9Ae3x = 0 6= e3x .
There is no way we can choose A to make the left hand side be e3x because our guess sat-
isfies homogeneous equation. Note that the general solution of the homogeneous equation
is
C.F. = C1 e−3x + C2 e3x
Thus our assumed particular solution is actually a solution of the corresponding homo-
geneous equation; consequently, it cannot possibly be a solution of the nonhomogeneous
equation. To find a particular solution we must therefore consider functions of a some-
what different form. We modify our guess to y = Axe3x and notice there is no difficulty
anymore. Note that y ′ = Ae3x + 3Axe3x and y ′′ = 6Ae3x + 9Axe3x . So
y ′′ − 9y = 6Ae3x + 9Axe3x − 9Axe3x = 6Ae3x .
Thus 6Ae3x is supposed to equal e3x . Hence, 6A = 1 and so A = 61 . We can now write the
general solution as
1 3x
y = yc + yp = C1 e−3x + C2 e3x + xe .
6
33.2.5 Problem 5
Find a particular solution of

y ′′ + 4y = 3 cos 2t
Solution: First we write its complimentary function

C.F. = c1 cos 2t + c2 sin 2t
As in earlier example, we guess

yp = At cos 2t + Bt sin 2t
Then, upon calculating yp′ and Yp′′ , substituting them into the given differential equation,
we find that
4A sin 2t + 4B cos 2t = 3cos2t
Therefore A = 0 and B = 3/4, so a particular solution of the given differential equation is
3
yp (t) = t sin 2t
4
Remark 4: It is also possible that multiplying by x does not get rid of the problem we
had faced in last two examples. For example,
y ′′ − 6y ′ + 9y = e3x .
The complementary solution is yc = C1 e3x + C2 xe3x . Guessing y = Ae3x or y = Axe3x

would not get us anywhere. In this case we will guess yp = Ax2 e3x .
Suggested Readings
New Delhi.
Lesson 34
Method of Variation of Parameters
In the last lesson we have discussed method of undetermined coefficients for finding par-
ticular integral. In this lesson we lean another rather general method, called method of
variation of parameters, of finding particular integral of non-homogeneous differential
equation. In contrast to the method of undetermined coefficients, this method is also ap-
plicable for solving linear equations with variable coefficients. For the sake of simplicity
we restrict ourselves for second order linear differential equations. However the method
is also applicable for higher order linear differential equations.
34.1 Method of Variation of Parameters
Consider a second order differential equation of the form
y ′′ + P y ′ + Qy = R (34.1)
where P, Q, R are functions of x or constants. If u and v are two linearly independent

solutions of the corresponding homogeneous differential equation
y ′′ + P y ′ + Qy = 0 (34.2)
Then, the complimentary function is
y = au + bv (34.3)
where a, b are two arbitrary constants and u, v are functions of x. Since u and v are solu-
tions of (34.2), we have
u′′ + P u′ + Qu = 0, v ′′ + P v ′ + Qv = 0. (34.4)
The method of variation of parameters relies on finding a particular integral of nonhomo-

geneous equation by replacing constants a and b with functions of x. The aim is to find
functions A(x) and B(x) such that
yp = Au + Bv (34.5)
is a particular integral of (34.1). To determine A(x) and B(x) we need to have two equa-
tions. These are obtained as follows. First we compute
yp′ = Au′ + Bv ′ + A′ u + B ′ v (34.6)
In order to avoid second order derivatives of A and B and to simplify the above expression
we take
A′ u + B ′ v = 0, (34.7)
Now, the Equation (34.6) reduces to
yp′ = Au′ + Bv ′ (34.8)
Differentiating (34.8), we obtain
yp′′ = A′ u′ + Au′′ + B ′ v ′ + Bv ′′ (34.9)
Using the values of y, y ′ and y ′′ given by (34.5), (34.8) and (34.9) into the Equation (34.1),
we get
A′ u′ + Au′′ + B ′ v ′ + Bv ′′ + P (Au′ + Bv ′ ) + Q(Au + Bv) = R
Further simplifications lead to
A(u′′ + P u′ + Qu) + B(v ′′ + P v ′ + Qv) + A′ u′ + B ′ v ′ = R
Using Equation (34.4) we get
A′ u′ + B ′ v ′ = R (34.10)
Solving (34.7) and (34.10) for A′ and B ′ , we get

−vR uR
A′ = , and B ′ = (34.11)
W W
where W is Wronskian of u and v , and given by W = uv ′ − u′ v 6= 0. Note that the
Wronskian is nonzero because u and v are two linearly independent solutions. Integrating
(34.11), we get
A = f (x), B = g(x), (34.12)
where Z Z
vR uR
f (x) = − dx, g(x) = dx
W W
Using (34.12) into (34.5), we have
yp = uf (x) + vg(x).
Hence, the general solution of the given differential equation is
yp = au + bv + uf (x) + vg(x).
34.2.1 Problem 1
Solve the differential equation y ′′ + n2 y = sec nx.

Solution: Comparing the given equation with the standard equation y ′′ + P y ′ + Qy = R,
we get P = 0, Q = n2 and R = sec nx. The characteristic equation of the corresponding
homogeneous equation is
(m2 + n2 )y = 0, so that m = ±in
C.F. = (c1 cos nx + c2 sin nx)
In order to find a particular integral, we have u = cos nx and v = sin nx and R = sec nx.
The Wronskian is given as

cos nx sin nx
W = = n 6= 0.

−n sin nx n cos nx
Then, the particular integral of the given equation is
P.I. = uf (x) + vg(x)
where
sin nx sec nx 1
Z Z
vR
f (x) = − dx = − dx = 2 ln(cos nx)
W n n
and
cos nx sec nx
Z Z
uR x
g(x) = dx = − dx =
W n n
Hence, the required solution is
1 x
y = (c1 cos nx + c2 sin nx) + cos nx 2
ln(cos nx) + sin nx
n n
34.2.2 Problem 2
Find the general solution of the differential equation y ′′ + n2 y = tan nx.

Solution: We compare the given equation with y ′′ + P y ′ + Qy = R to have P = 0, Q = n2
and R = tan nx. Similar to the previous example, we have the complimentary function as
To find particular integral we have u = cos nx, v = sin nx, R = sec nx. The Wronskian is
given by
cos nx sin nx
W = = n 6= 0.


P.I. = uf (x) + vg(x)
where
sin nx tan nx 1
Z Z
vR
f (x) = − dx = − dx = 2 [sin nx − ln(sec nx + tan nx)]
W n n
and
cos nx tan nx 1
Z Z
uR
g(x) = dx = − dx = − 2 cos nx
W n n
The desired general solution is
cos nx 1
y = (c1 cos nx + c2 sin nx) + 2
[sin nx − ln(sec nx + tan nx)] − 2 sin nx cos nx
n n
34.2.3 Problem 3
d2 y
Solve the differential equation 2
+ n2 y = cot nx.
dx
Solution: Similar to the previous example, the complimentary function is given by
In this case, we have u = cos nx, v = sin nx and R = cot nx. The Wronskian is given by

cos nx sin nx
W = = n 6= 0

Then, the particular integral is
P.I. = uf (x) + vg(x)
where
sin nx cot nx 1
Z Z
vR
f (x) = − dx = − dx = − 2 sin nx
W n n
and
cos nx cot nx 1 h
Z Z
uR nx i
g(x) = dx = − dx = 2 cos nx + ln tan
W n n 2
The required solution is
1 1 h nx i
y = (c1 cos nx + c2 sin nx) − cos nx sin nx + cos nx + ln tan sin nx
n2 n2 2
34.2.4 Problem 4
Using the method of variation of parameters, find the general solution of the differential
equation
d2 y
+ y = sec2 x
dx2
Solution: The complimentary function is given by
C.F. = (c1 cos x + c2 sin x)
Also we have u = cos x and v = sin x and R = sec2 x and

cos x sin x
W = = 1 6= 0.

− sin x cos x
Then, the particular integral is
P.I. = uf (x) + vg(x)
where
Z Z Z
vR 2
f (x) = − dx = − sin x sec xdx = − sec x tan xdx = − sec x
W
and Z Z Z
uR 2
g(x) = dx = cos x sec xdx = sec xdx = ln[sec x + tan x]
W
The required solution is
y = (c1 cos x + c2 sin x) − cos x sec x + sin x ln[sec x + tan x].
Suggested Readings
New Delhi.
Lesson 35
Equations Reducible to Linear Differential Equations with Constant

Coefficients
In this lesson we shall study two special forms of linear equations with variable coeffi-
cients which can be reduced to linear differential equations with constant coefficients by
a suitable substitution. Those special forms which we study here are called Cauchy-Euler
homogeneous linear differential equations and Legendre’s homogeneous linear differen-
tial equations.
35.1 Cauchy-Euler Homogeneous Linear Differential Equation
A linear differential equation of the form

ny n−1 y n−2 y
nd n−1 d n−2 d
a0 x + a1 x + a2 x + . . . + an y = F (x), (35.1)
dxn dxn−1 dxn−2
where a1 , a2 , . . . , an are constants and F is either a constant or a function of x only, is
called Cauchy-Euler homogeneous linear differential equation. Note that the index of x
and order of derivative is same in each term of such equations.
Using the symbols D(= d/dx), D2(= d2 /dx2 ), . . . , Dn(= dn /dxn ), the Equation (35.1) be-
comes
(a0 xn D n + a1 xn−1 D n−1 + a2 xn−2 D n−2 + ... + an )y = F (x) (35.2)
The above equation can be reduced to linear differential equation with constant coeffi-
cients by substituting
dz 1
x = ez , or ln x = z, so that = (35.3)
dx x
Using chain rule for differentiation we obtain
dy dy dz 1 dy
= =
dx dz dx x dz
Equations Reducible to Linear Differential Equations with Constant Coefficients
d
Defining =: D1 , we have
dz
dy dy
x = ⇔ xDy = D1 y
dx dz
Similarly, for the second order derivative
d2 y

ddy d 1 dy 1 dy 1 d dy
2
= = =− 2 +
dx dxdx dx x dz x dz x dx dz
1 d2 y

1 dy 1 d dy dz 1 dy
=− 2 + =− 2 + 2 2
x dz x dz dz dx x dz x dz
Thus, we have
d2 y d2 y dy
x2 = − ⇒ x2 D 2 y = D1 (D1 − 1)y.
dx2 dz 2 dz
Similarly, x3 D3 y = D1 (D1 − 1)(D1 − 2)y and so on. In general, we have the relationship
xn D n = D1 (D1 − 1)(D1 − 2) . . . (D1 − n + 1)y
Substituting the above values of x, xD, x2 D2 , . . . , xn Dn in the Equation (35.1), we get
[a0 D1 (D1 − 1) . . . (D1 − n + 1) + . . . + an−2 D1 (D1 − 1) + an−1 D1 + an ] y = F (ez ) (35.4)
The Equation (35.4) is a linear differential equation with constant coefficients which can
solved with the methods discussed in previous lessons. Finally, by replacing z by ln x we
obtain the desired solution of the given differential equation.
35.2.1 Problem 1
Solve the differential equation (x2 D2 + xD − 4)y = 0.

Solution: Substituting x = ez ⇒ ln x = z ⇒ xD = D1 , x2 D2 = D1 (D1 − 1), the given
equation reduces to
[D1 (D1 − 1) + D1 − 4] y = 0 ⇒ (D12 − 4)y = 0
The roots of the corresponding characteristic equation are m = 2, −2. The required solu-
tion of the transformed equation is
y = c1 e2z + c2 e−2z
Putting log x = z , we have the desired solution as
y = c1 x2 + c2 x−2 .
Here c1 and c2 are arbitrary constants.
35.2.2 Problem 2
Find the general solution of the differential equation (x2 D2 + y)y = 3x2 .
Solution: Substituting x = ez , the given equation reduces to
(D1 (D1 − 1) + 1)y = 3e2z ⇒ (D12 − D1 + 1)y = 3e2z
The characteristic equation of this differential equation is

√
(m2 − m + 1) = 0 ⇒ m = (1 ± i 3)/2

h √ √ i
C.F. = ez/2 c1 cos z 3/2 + c1 sin z 3/2
Substituting z = ln x, we get
√ h √ √ i
C.F. = x c1 cos ln x 3/2 + c1 sin ln x 3/2
The particular integral of the transformed equation is

1 1
P.I. = 3e2z = 2 3e2z = e2z
D12 − D1 + 1 2 −2+1
Hence, the desired solution of the given differential equation is

√ h √ √ i
y= x c1 cos ln x 3/2 + c1 sin ln x 3/2 + x2
35.3 Legendre’s Homogeneous Linear Differential Equations
A linear differential equation of the form is
[(a + bx)n a0 D n + a1 (a + bx)n−1 D n−1 + a2 (a + bx)n−2 D n−2 + ... + an ]y = F (x), (35.5)
where a, b, a1 , a2 , ..., an are constants, and F is either a constant or a function of x only,

is called a Legendre’s homogeneous linear differential equation. Note that the index of
(a + bx) and the order of derivative is same in each term of such equation. To solve the
Equation (35.5), we introduce a new independent variable z such that
a + bx = ez , or ln(a + bx) = z, so that b/(a + bx) = dz/dx. (35.6)
Now, for the first order derivative we have

dy dy dz b dy
= =
dx dz dx (a + bx) dz
This implies
dy dy
(a + bx) =b ⇔ (a + bx)Dy = bD1 y
dx dz
Similarly for the second order derivative we get
d2 y

d dy d b dy
2
= =
dx dx dx dx (a + bx) dz
This can be further simplified to get
d2 y b2

dy b d dy
=− +
dx2 (a + bx)2 dz (a + bx) dx dz
b2

dy b d dy dz
=− +
(a + bx)2 dz (a + bx) dz dz dx
Substituting dz/dx from Equation (35.6), we obtain
d2 y b2 dy b2 d2 y
= − +
dx2 (a + bx)2 dz (a + bx)2 dz 2
This gives us
2y
d2 y dy

2d 2
(a + bx) =b − ⇔ (a + bx)2 D 2 y = b2 D1 (D1 − 1)y
dx2 dz 2 dz
In general, we have
(a + bx)n D n = bn D1 (D1 − 1)(D1 − 2) . . . (D1 − n + 1)y
Substituting the above values of (a + bx), (a + bx)D, (a + bx)2 D2 , . . . , (a + bx)n Dn in the

Equation (35.5), we get the following linear differential equation with constant coeffi-
cients
ez − a

n 2

a0 b D1 (D1 − 1)...(D1 − n + 1) + ... + an−2 b D1 (D1 − 1) + an−1 bD1 + an y = F
b
The methods of solving this transformed equation are same as discussed in previous sec-
tion.
35.3.1 Example

d3 y 2
3d y dy 1
(1 + x)4 3
+ 2(1 + x) 2
− (1 + x)2 + (1 + x)y =
dx dx dx (1 + x)
d
Solution: Using D = and dividing both sides by (x+1), the given differential equation
dx
can be rewritten as
(1 + x)3 D 3 + 2(1 + x)2 D 2 − (1 + x)D + 1 y = (1 + x)−2 .

This is the Legendre’s homogeneous linear equation which can be solved by substituting
(1 + x) = ez ⇔ ln(1 + x) = z
This substitution readily implies
(1 + x)D = D1 , (1 + x)2 D 2 = D1 (D1 − 1), (1 + x)3 D 3 = D1 (D1 − 1)(D1 − 2)
The given differential equation reduces to
[D1 (D1 − 1)(D1 − 2) + 2D1 (D1 − 1) − D1 + 1] y = e−2z
or
D13 − D12 − D1 + 1 y = e−2z

The characteristic equation of the corresponding homogeneous equation is
m3 − m2 − m + 1 y = 0

The roots of the characteristics equations are m = 1, 1, −1. Hence the complimentary
function of the transformed differential equation is
C.F. = (c1 + c2 z)ez + c3 e−z
The particular integral of the transformed differential equation can be found as

1
P.I. = e−2z
(D13 − D12− D1 + 1)
1
= e−2z
−23 2
−2 +2+1
1
= − e−2z
9
Hence the general solution of the transformed differential equation is
1
y = (c1 + c2 z)ez + c3 e−z − e−2z
9
Replacing z by ln(1 + x) we obtain the desired solution of the given differential equation
c3 1 1
y = [c1 + c2 ln(1 + x)] (1 + x) + − .
(1 + x) 9 (1 + x)
Suggested Readings
New Delhi.
Lesson 36
Methods for Solving Simultaneous Ordinary Differential Equations
In this lesson we shall consider systems of simultaneous linear differential equations

which contain a single independent variable and two or more dependent variables. We
will consider two different techniques, mainly the method of elimination and the method
of differentiation, for solving system of differential equations.
36.1 Simultaneous Ordinary Linear Differential Equations
Let x and y be the dependent and t be the independent variable. Thus, in such equations
there occur differential coefficients of x, y with respect to t. Let D = d/dt, then such
equations can be put into the form
f1 (D)x + f2 (D)y = T1 (36.1)

g1 (D)x + g2 (D)y = T2 (36.2)
where T1 and T2 are functions of the independent variable t and f1 (D), f2 (D), g1 (D), and
g2 (D) are all rational integral functions of D with constant coefficients. In general, the
number of equations will be equal to the number of dependent variables, i.e., if there are
n dependent variables there will be n equations.
36.2 Method of Elimination
In order to eliminate y between equations (36.1) and (36.2), operating on both sides of
(36.1) by g2 (D) and on both sides of (36.2) by f2 (D) and subtracting, we get
(f1 (D)g2 (D) − g1 (D)f2 (D)) x = g2 (D)T1 − f2 (D)T2 (36.3)
This is a linear differential equation with constant coefficients in x and t and can be solved
to give the value of x in terms of t. Substituting this value of x in either (36.1) or (36.2),
we get the value of y in terms of t.
Remark 1: The above Equations (36.1) and (36.2) can be also solved by first elimi-
nating x between them and solving the resulting equation to get y in terms of t. Substitut-
ing this value of y in either (36.1) or (36.2), we get the value of x in terms of t.
Remark 2: In the general solutions of (36.1) and (36.2) the number of arbitrary
constants will be equal to the sum of the orders of the equations (36.1) and (36.2).
36.3.1 Problem 1
Solve the simultaneous equations

dx
− 7x + y = 0 (36.4)
dt
dy
− 2x − 5y = 0 (36.5)
dt
Solution: Writing D for d/dt, the given equations can be rewritten in the following
symbolic form as
(D − 7)x + y = 0 (36.6)
−2x + (D − 5)y = 0 (36.7)
Now, we eliminate x by multiplying Equation (36.6) by 2 and operating (36.7) by (D − 7)

as follows
2(D − 7)x + 2y = 0 (36.8)

−2(D − 7)x + (D − 7)(D − 5)y = 0 (36.9)
Adding (36.8) and (36.9), we get
[(D − 7)(D − 5) + 2]y = 0
or
(D 2 − 12D + 37)y = 0
This is a linear equation with constants coefficients. Its auxiliary equation is
(m2 − 12m + 37) = 0
The roots of the auxiliary equation are m = 6 ± i. Therefore, we get the general solution
for the variable y as
y = e6t (c1 cos t + c2 sin t), (36.10)
where c1 and c2 being arbitrary constants. We now find x by using Equation (36.7). Now
from (36.10), differentiating w.r.t. t, we get
Dy = 6e6t [(c1 cos t + c2 sin t) + e6t (−c1 sin t + c2 cos t)],
or on simplifications we obtain
Dy = 6e6t [(6c1 + c2 ) cos t + (−c1 + 6c2 ) sin t] (36.11)
Now, substituting y and Dy in the Equation (36.7), we get
x = (1/2) × e6t [(c1 + c2 ) cos t + (−c1 + c2 ) sin t] (36.12)
Thus, equations (36.10) and (36.12) give the desired general solution.
36.3.2 Problem 2
Solve the linear system of differential equations
D 2 y − y + 5Dv = x (36.13)
2Dy − D 2 v + 4v = 2 (36.14)
Solution: Multiplying (36.13) by 2D and (36.14) by (D2 − 1) and then subtracting (36.14)
from the Equation (36.13) we obtain
[10D 2 + (D 2 − 1)(D 2 − 4)]v = 2Dx − (D 2 − 1)2
or
(D 4 + 5D 2 + 4)v = 4 (36.15)
This is a linear differential equations with constant coefficients whose solution can easily
be found. The characteristic equation of the corresponding homogeneous equation is
m4 + 5m2 + 4 = 0 ⇒ (m2 + 1)(m2 + 4) = 0 ⇒ m = ±i, ±2i
C.F. = c1 cos x + +c2 sin x + c3 cos 2x + c4 sin 2x

1
P.I. = 4e0x = 1
D4 + 5D 2 + 4
We write the general solution for v as
v = 1 + c1 cos x + c2 sin x + c3 cos 2x + c4 sin 2x (36.16)
Now we find an equation giving y in terms of v . This can be done by eliminating from the
equations (36.13) and (36.14) those terms which involve derivatives of y. So multiplying
Equation (36.13) by 2 and Equation (36.14) by D we get
(2D 2 − 2)y + 10Dv = 2x (36.17)

2D 2 y − (D 3 − 4D)v = 0 (36.18)
Subtracting (36.17) from (36.18) we get
2y − D 3 v − 6Dv = −2x (36.19)
or
1
y = −x + D 3 v + 3Dv (36.20)
2
Substitute v from (36.16) into the Equation (36.21) to obtain the expression for y as
5 5
y = −x − c1 cos x + c2 cos x + 2c4 cos 4x − 2c3 sin 2x, (36.21)
2 2
where c1 , c2 , c3 and c4 are arbitrary constants.
36.4 Method of Differentiation
Sometimes, x and y can be eliminated if we differentiate (36.1) or (36.2). For example,

assume that the given equations (36.1) and (36.2) relates four quantities x, y, dx/dt and
dy/dt. Differentiating (36.1) and (36.2) with respect to t, we obtain four equations con-
taining x, dx/dt, d2 x/dt2 , y, dy/dt and d2 y/dt2 . Eliminating three quantities y, dy/dt and
d2 y/dt2 from these four equations, y is eliminated and we get an equation of the second
order with x as the dependent and t as the independent variable. Solving this equation we
get value of x in terms of t. Substituting this value of x in either (36.1) or (36.2), we get
value of y in terms of t. The technique will be illustrated by the following example.
36.4.1 Example
Determine the general solutions for x and y for

dx
−y =t
dt
dy
+x=1
dt
Solution: Writing D for d/dt, the given equations become
Dx − y = t (36.22)
x + Dy = 1 (36.23)
Differentiating the equation Equation (36.22) w.r.t. t we get
D 2 x − Dy = 1 (36.24)
Now we can eliminate y by adding equations (36.24) and (36.23) to get
(D 2 + 1)x = 2 (36.25)
The auxiliary equation of the above differential equation is m2 + 1 = 0 and therefore the
general solution of the homogeneous equation is
C.F. = c1 cos t + c2 sin t
where c1 and c2 are arbitrary constants. The particular integral is

1
P.I = 2
2 = (1 + D 2 )−1 2 = (1 − D 2 + ...)2 = 2
D +1
Hence, the general solution of (36.25) is
x = c1 cos t + c2 sin t + 2 (36.26)
From Equation (36.22), we get
y = c2 cos t − c1 sin t − t (36.27)
Thus, the required solution is given by (36.26) and (36.27).
Suggested Readings
Waltman, P. (2004). A Second Course in Elementary Differential Equations. Dover Pub-

lications, Inc. New York.
New Delhi.
Publisher, Moscow.
Lesson 37
Series Solutions about an Ordinary Point
37.1 Introduction
If we can’t find a solution to a differential equations in a form of nice functions, we can

still look for a series representation of the solution. Series solutions are very useful be-
cause if we know that the series converges, we can approximate the solution as closely as
we want. In this lesson we describe series solutions of solving second order linear homo-
geneous differential equations with variables coefficients. Series solution can be used in
conjunction with variation of parameters to solve linear nonhomogeneous equations. For
simplicity, we shall be dealing mainly with polynomial coefficients. Here we consider the
second order homogeneous equation of the form
P (x)y ′′ + Q(x)y ′ + R(x)y = 0 (37.1)
where P, Q and R are polynomials or analytic functions in general. Many problems in

mathematical physics leads to equations of the form (37.1) having polynomial coeffi-
cients; for example, the Bessel equation
x2 y ′′ + xy ′ + (x2 a2 )y = 0,
where a is a constant, and the Legendre equation
(1 − x)2 y ′′ − 2xy ′ + c(c + 1)y = 0
where c is a constant.
37.2 Useful Definitions
Here we provide some definitions which will be very useful for finding series solution of
the differential equations.
37.2.1 Analytic Function
A function f (x) defined on an interval containing the point x = x0 is called analytic at x0

if its Taylor series,
∞
X f (n) (x0 )
(x − x0 ) (37.2)
n!
n=0
exists and converges to f (x) for all x in the interval of convergence of (37.2).
37.2.2 Ordinary Points
A point x = x0 is called an ordinary point of the Equation (37.1) if P , Q, and R are

polynomials that do not have any common factors, then a point x0 is called an ordinary
point if P (x0 ) 6= 0. A point x1 where P (x1 ) = 0 is called a singular point. If any of P , Q,
or R is not a polynomial, then we call x0 an ordinary point if Q(x)/P (x) and R(x)/P (x)
are analytic about x0 .
It is often useful to rewrite Equation (37.1) as
y ′′ + p(x)y ′ + q(x)y = 0 (37.3)
where p(x) = Q(x)/P (x) and q(x) = R(x)/P (x). The Equation (37.3) is called equivalent
normalized form of the Equation (37.1).
37.2.3 Singular Points
If the point x = x0 is not an ordinary point of the differential Equation (37.1) or (37.3),
then it is called a singular point of the differential equation of (37.3). There are two types
of singular points: (i) regular singular points, and (ii) irregular singular points. A singular
point x = x0 of the differential Equation (37.3) is called a regular singular point of the
differential Equation (37.3) if both
(x − x0 )p(x) and (x − x0 )2 q(x)
are analytic at x = x0 . A singular point, which is not regular is called an irregular singular
point.
37.3.1 Problem 1
Show that x = 0 is an ordinary point of (x2 − 1)y ′′ + xy ′ − y = 0, but x = 1 is a regular

singular point.
Solution: Writing the given equation in normalized form
d2 y x dy 1
2
+ − y = 0. (37.4)
dx (x − 1)(x + 1) dx (x − 1)(x + 1)
Comparing (37.4) with the standard equation y ′′ + p(x)y ′ + q(x)y = 0, we have
p(x) = x/(x − 1)(x + 1) and q(x) = −1/(x − 1)(x + 1).
Since both p(x) and q(x) are analytic at x = 0, the point x = 0 is an ordinary point of the
given Equation (37.4). Further note that both p(x) and q(x) are not analytic at x = 1, thus
x = 1 is not an ordinary point and so x = 1 is a singular point. Also
(x − 1)P (x) = x/(x + 1) and (x − 1)2 Q(x) = −(x − 1)/(x + 1)
show that both (x − 1)P (x) and (x − 1)2 Q(x) are analytic at x = 1. Therefore x = 1 is a
regular singular point.
37.3.2 Problem 2
Determine whether the point x = 0 is an ordinary point or regular point of the differential
equation
xy ′′ + sin(x)y + x2 y = 0
Solution: Comparing with the normalized equation we get

sin x
p(x) = and q(x) = x
x
Since p(x) and q(x) both are analytic at x = 0, the point x = 0 is an ordinary point. This
example shows that singular point does not always occur where P (x) = 0.
37.3.3 Problem 3
Discuss the singular points of the differential equation
x2 (x − 2)2 y ′′ + (x − 2)y ′ + 3x2 y = 0.
Solution: Clearly the function

1
p(x) =
(x2 (x − 2))
is not analytic at x = 0 and x = 2. Also the function

3
q(x) =
((x − 2)2 )
is not analytic at x = 2. Hence both x = 0 and x = 2 are singular point of the differential
equations. At x = 0 we have
1 3x2
xp(x) = and x2 q(x) =
(x(x − 2)) (x − 2)2
Note that x2 q(x) is non-singular at x = 0 but xp(x) is not analytic at this point. Hence
x = 0 is an irregular singular point. At x = 2 we have
1
(x − 2)p(x) = and (x − 2)2 q(x) = 2
x2
Both functions are analytic at x = 2 and hence x = 2 is a regular singular point.
37.4 Brief Overview of Power Series
A power series about a point x0 is a series of the form

∞
X
cn (x − x0 )n
n=0
where x is a variable and cn are constants, called coefficients of the series. There are
three possibilities about the convergence of a power series. The series may converge only
at x = 0 or it may converge for all values of x. If this is not the case then a definite
positive number R exists such that the given series converges for every |x − x0 | < R and
diverges for every |x − x0 | > R. Such a number is known as the radius of convergence and
]x0 − R, x0 + R[, the interval of convergence, of the given series.
Among several formulas for determining convergence of the power series, ratio test is
most common and simple to use. Given a power series ∞
P
n=0 cn (x − x0 ) we compute
n

1 cn+1
= lim ,
R n→∞ cn
then the series is convergence for |x − x0 | < R and divergent |x − x0 | > R.
37.4.1 Example
Determine the radius of convergence of the power series

∞
X (x + 1)n
n2n
n=1
Solution: Ratio test gives

n2 n 1
lim =
n→∞ (n + 1)2n+1 2
Hence the radius of convergence of the power series is R = 2 and the interval of conver-
gence is −3 < x < 1. The convergence at the end points x = −3 and x = 1 needs to be
checked separately.
37.5 Power Series Solution near Ordinary Point
Let the given equation be
y ′′ + p(x)y ′ + q(x)y = 0 (37.5)
If x = x0 is an ordinary point of (37.5), then (37.5) has two non-trivial linearly indepen-
dent power series solutions of the form
∞
X
Cn (x − x0 )n (37.6)
n=0
and these power series converge in some interval of convergence |x − x0 | < R, (where R
is the radius of convergence of (37.6)) about x0 .
To find series solutions we suppose that we have a series representation,

∞
X
y= Cn (x − x0 )n (37.7)
n=0
and then to find out coefficients Cn we need to differentiate (37.7) and plug in the deriva-
tives into the Equation (37.6). Once we have the appropriate coefficients, we call (37.7)
the series solution to (37.5) near x = x0 . More precisely, differentiating twice, the Equa-
tion (37.7) yields
∞
X ∞
X
′ ′′
y = nCn (x − x0 ) n−1
and y = n(n − 1)Cn (x − x0 )n−2 (37.8)
n=0 n=0
Substituting the above values of y, y ′ and y ′′ in (37.5), we obtain
A0 + A1 (x − x0 ) + A2 (x − x0 )2 + . . . + An (x − x0 )n + . . . = 0, (37.9)
where the coefficients A0 , A1 , A2 . . . etc. are now some functions of the coefficients C0 , C1 , C2 , . . .
etc. Since the Equation (37.9) is an identity, all the coefficients A0 , A1 , A2 . . . of (37.9)
must be zero, i.e.,
A0 = 0, A1 = 0, A2 = 0, . . . , An = 0 (37.10)
Solving Equation (37.10), we obtain the coefficients of (37.7) in terms of C0 and C1 .

Substituting these coefficients in (37.7), we obtain the required series solution of (37.5)
in power of (x − x0 ).
Suggested Readings
Raisinghania, M.D. (2009). Advanced Differential Equations. Twelfth Edition. S. Chand
& Company Ltd., New Delhi.
Lesson 38
Series Solution about an Ordinary Point (Cont.)
In the last lesson we have discussed series solution of the homogeneous differential equa-
tions. In this lesson we demonstrate the method by using a couple of basic examples. For
demonstration we take first example of a differential equation with constant coefficients
and then some more involved examples will be discussed.
38.1.1 Problem 1
Determine a series solution to y ′′ − y = 0.

Solution: Suppose that the series solution is of the form
∞
X
y(x) = cn xn
n=0
Differentiating y , we have
∞
X ∞
X
′ ′′
y (x) = ncn xn−1 and y (x) = n(n − 1)cn xn−2
n=1 n=2
Substituting these into the differential equation, we have

∞
X ∞
X
n(n − 1)cn xn−2 − cn xn = 0
n=2 n=0
Re-indexing the first sum

∞
X ∞
X
(n + 2)(n + 1)cn+2 xn − cn xn = 0
n=0 n=0
This implies
∞
X
[(n + 2)(n + 1)cn+2 − cn xn ] xn = 0
n=0
Series Solutions about an Ordinary Point (Cont.)
Since the series is always equal to 0 then each coefficient must be zero. Thus we have
(n + 2)(n + 1)cn+2 − cn = 0 (38.1)
This can be rewritten in the form of recurrence relation as

cn
cn+2 = (38.2)
(n + 2)(n + 1)
Putting n = 0, 1, 2 . . ., we get
c0 c1 c0 c1
c2 = , c3 = , c4 = , c5 = ,...
2! 3! 4! 5!
In general, we have
c0 c1
c2k = , c2k+1 = . . . for k = 1, 2, . . . .
(2k)! (2k + 1)!
Putting these values into the series and collecting the c0 and c1 terms we get
x2 x2k x3 x2k+1

y(x) = c0 1+ + ...+ + . . . + c1 x + + ...+ + ...
2! (2k)! 3! (2k + 1)!
This can be further rewritten in summation form as

∞ ∞
X x2k X x2k+1
y(x) = c0 + c1
(2k)! (2k + 1)!
k=0 k=0
This is the desired series solution. It should be noted that this series solution can be
rewritten into the form of well known solution y(x) = c1 ex + c2 e−x of the given differential
equation as
x2 x2

x −x
c1 e + c2 e = c1 1+x+ + . . . + c2 1 − x + + ...
2! 2!
This can be rewritten as

x2 x3

x −x
c1 e + c2 e = (c1 + c2 ) 1 + + . . . + (c1 − c2 ) x + + ...
2! 3!
Denoting (c1 + c2 ) =: c0 and (c1 − c2 ) =: c1 we get

∞ ∞
x −x
X x2k X x2k+1
c1 e + c2 e = c0 + c1
(2k)! (2k + 1)!
k=0 k=0
This proves that both representations are equivalent.
38.1.2 Problem 2
Find the series solution, about x = 0, of the equation (1 − x)2 y ′′ − 2y = 0 in powers of x.

Solution: Since x = 0 is an ordinary point and we can therefore get two linearly indepen-
dent solution by substituting
∞
X
y= cn xn .
n=0
After substitution we get
∞
X ∞
X
2 n−2
(1 − 2x + x ) n(n − 1)cn x −2 cn xn = 0,
n=2 n=0
which leads to
∞
X ∞
X ∞
X ∞
X
n−2 n−1 n
n(n − 1)cn x −2 n(n − 1)cn x + n(n − 1)cn x − 2 cn xn = 0
n=2 n=2 n=2 n=0
In order to write the series in terms the coefficients of xn we shift the summation index as
∞
X ∞
X ∞
X ∞
X
n n n
(n + 2)(n + 1)cn+2 x − 2 n(n + 1)cn+1 x + n(n − 1)cn x − 2 cn xn = 0
n=0 n=1 n=2 n=0
The sum in second and third series can also start from 0 without changing the series. This
leads to
∞
X
[(n + 2)(n + 1)cn+2 − 2n(n + 1)cn+1 + n(n − 1)cn − 2cn ] xn = 0
n=0
This can be further simplified as

∞
X
(n + 1) [(n + 2)cn+2 − 2ncn+1 + (n − 2)cn ] xn = 0
n=0
Equating the coefficients we obtain the recurrence relation
(n + 2)cn+2 − 2ncn+1 + (n − 2)cn = 0.
Putting n = 0, 1, 2, . . . we get
1
c2 = c0 , c3 = (2c0 + c1 ) =: c, c4 = c, c5 = c . . .
3
Hence the series solution becomes
∞
X
2
y = c0 + c1 x + c0 x + c xn .
n=3
38.1.3 Problem 3
Find the power series solution of the equation (x2 + 1)y ′′ + xy ′ − xy = 0 in powers of x
(i.e. about x = 0).
Solution: Clearly x = 0 is an ordinary point of the given differential equation. Therefore,
to find the series solution, we take power series
∞
X
2 3
y = c0 + c1 x + c2 x + c3 x + . . . = cn xn . (38.3)
n=0
Differentiating twice in succession, (38.3) gives

∞
X ∞
X
n−1
′
y = ncn x and y = ′′
n(n − 1)cn xn−2 (38.4)
n=1 n=1
Putting the above value of y, y ′ and y ′′ in the given differential equation, we obtain
∞
X ∞
X ∞
X
(x2 + 1) n(n − 1)cn xn−2 + x ncn xn−1 − x ncn xn = 0
n=2 n=1 n=0
∞
X ∞
X ∞
X ∞
X
n n−2 n
⇒ n(n − 1)cn x + n(n − 1)cn x − ncn x − cn xn+1 = 0
n=2 n=2 n=1 n=0
This leads to
∞
X ∞
X ∞
X ∞
X
n n n
n(n − 1)cn x + (n + 2)(n + 1)cn+2 x + ncn x − cn−1 xn = 0
n=2 n=0 n=1 n=1
Finally we have the identity

∞
X
2c2 + (6c3 + c1 − c0 )x + [n(n − 1)cn + (n + 2)(n + 1)cn+2 + ncn − cn−1 ]xn = 0.
n=2
Equating the constant term and the coefficients of various powers of x, we get
c2 = 0, 6c3 + C1 − c0 = 0 so that c3 = (c0 − c1 )/6
and the recurrence relation

cn−1 − n2 cn
cn+2 = , for all n ≥ 2. (38.5)
(n + 1)(n + 2)
Putting n = 2 in (38.5), c4 = (1/12)c1, as c2 = 0.

9c3 3
Putting n = 3 in (38.5), c5 = − (20) = − 40 (c0 − c1 )
Putting the above values of c2 , c3 , c4 , c5 , . . . ets. in (38.3), we have
y = c0 + c1 x + c2 x2 + c3 x3 + c4 x4 + c5 x5 + . . . ∞
⇒ y = c0 + c1 x + (1/6)(c0 − c1 )x3 + (1/12)c1 x4 − (3/40)(c0 − c1 )x5 + . . . ∞
This can be rewritten as

1 3 1 1 3
y = c0 1 + x3 − x5 + . . . + c1 x − x3 + x4 + x5 − . . . ,
6 40 6 12 40
which is the required solution near x = 0, where c0 and c1 are arbitrary constants.
38.1.4 Problem 4
Find the power series solution of the initial value problem xy ′′ + y ′ + 2y = 0, y(1) = 1,
y ′ (1) = 2 in powers of (x − 1).
Solution: Since x = 1 is an ordinary point of the given differential equation, we find

series solution
∞
X ∞
X ∞
X
n n−1
y= cn (x − 1) ⇒ y = ′
ncn (x − 1) and y =′′
n(n − 1)cn (x − 1)n−2 (38.6)
n=0 n=1 n=2
Substituting y and y ′ in the given differential equation we obtain

∞
X ∞
X ∞
X
n−2 n−1
[(x − 1) + 1] n(n − 1)cn (x − 1) + ncn (x − 1) +2 cn (x − 1)n = 0
n=2 n=1 n=0
This leads to
∞
X ∞
X ∞
X ∞
X
n−1 n−2 n−1
n(n − 1)cn (x − 1) + n(n − 1)cn (x − 1) + ncn (x − 1) +2 cn (x − 1)n = 0
n=2 n=2 n=1 n=0
Shifting summation index of the first three terms we get

∞
X ∞
X
n
n(n + 1)cn+1 (x − 1) + [(n + 1)(n + 2)cn+2 + (n + 1)cn+1 + 2cn ](x − 1)n = 0
n=1 n=0
Equating the coefficients to zero we get

c1 + c0
2c2 + c1 + c0 = 0 ⇒ c2 = −
2
(n + 1)2 cn+1 + 2cn

cn+2 =− , for all n ≥ 1
(n + 1)(n + 2)
Using initial conditions in Equation (38.6) we get c0 = 1 and c1 = 2. Using these values
we obtain
2 1 1
c2 = −2, c3 = , c4 = − , c5 = ,...
3 6 15
Putting these constants in series we get the desired solution as
y = 1 + 2(x − 1) − 2(x − 1)2 + (2/3)(x − 1)3 − (1/6)(x − 1)4 + (1/15)(x − 1)5 + . . .
Suggested Readings
Waltman, P. (2004). A Second Course in Elementary Differential Equations. Dover Pub-

lications, Inc. New York.
New Delhi.
Lesson 39
Series Solutions about a Regular Singular Point
39.1 Introduction
In this lesson we discuss series solution about a singular point. In particular, the power
series method discussed in last lessons will be generalized. The generalized power series
method is also known as Frobenius method.
Let us consider a simple first order differential equation 2xy ′ − y = 0 and try to apply the
power series method discussed in the last lessons. Note that x = 0 is a singular point. If
we plug in
∞
X
y= ak xk ,
k=0
into the given differential equation, we obtain

∞
! ∞
!
X X
0 = 2xy ′ − y = 2x kak xk−1 − ak xk
k=1 k=0
X∞
= a0 + (2kak − ak ) xk .
k=1
First, a0 = 0. Next, the only way to solve 0 = 2kak − ak = (2k − 1) ak for k = 1, 2, 3, . . . is

for ak = 0 for all k . Therefore we only get the trivial solution y = 0. We need a nonzero
solution to get the general solution.
39.2 Frobenius Method
Consider the differential equation of the form y ′′ + p(x)y ′ + q(x)y = 0. Note that xp(x) and
x2 q(x) are analytic at x = 0. We try a series solution of the from
∞
X
r
y=x cn xn = xr (c0 + c1 x + c2 x2 + . . .), where c0 6= 0
n=0
The derivative of y with respect to x are given by

∞
X
′
y = (n + r)cn xn+r−1
n=0
∞
X
′′
y = (n + r)(n + r − 1)cn xn+r−2
n=0
Also, we can write power series corresponding to xp(x) and x2 q(x) as

∞
X ∞
X
n
xp(x) = an x 2
and x q(x) = bn xn
n=0 n=0
The given differential equation can be rewritten as
xp(x) ′ x2 q(x)
y ′′ + y + y=0
x x2
Substituting all values of y , y ′, y ′′, xp(x) and x2 q(x) series into the above differential
equation we get
∞
X ∞
X ∞
X ∞
X ∞
X
(n+r)(n+r−1)cn xn+r−2 + an xn−1 × (n+r)cn xn+r−1 + bn xn−2 × cn xn+r = 0
n=0 n=0 n=0 n=0 n=0
Multiplying by x2 we get
∞
X ∞
X ∞
X ∞
X ∞
X
n+r n n+r n
(n + r)(n + r − 1)cn x + an x × (n + r)cn x + bn x × cn xn+r = 0
n=0 n=0 n=0 n=0 n=0
We can now equate coefficients of various powers of x to zero to form a system of equa-
tions involving unknown coefficients cn . Equating the coefficient of xr we obtain
[r(r − 1) + a0 r + b0 ]c0 = 0
Since c0 6= 0, we obtain
r 2 + (a0 − 1)r + b0 = 0 (39.1)
The above quadratic equation is known as the indicial equation of the given differential
equation. The general solution of the given differential equation depends on the roots of
the indicial equation. There are three possible general cases:
39.2.1 Case I: The indicial equation has two real roots which do not differ by an
integer
Let r1 and r2 are the roots of the indicial equation. Then the two linearly independent
solution will follow from
∞
X ∞
X
r1 n r2
y1 (x) = x cn z y2 (x) = x cn z n
n=0 n=0
where c0 , c1 , . . . are coefficients corresponding to r = r1 and c0 , c1 , . . . are coefficients

corresponding to r = r2 . The general solution will be of the form y = ay1 + by2 , where a
and b are arbitrary coefficients.
39.2.2 Case II: The indicial equation has a doubled root
If the indicial equation has a doubled root r, then we find one solution
∞
X
y1 = xr ak xk ,
k=0
and then obtain another solution by plugging

∞
X
r
y2 = x bk xk + (ln x)y1 ,
k=0
into the given equation and solving for the constants bk .
39.2.3 Case III: The indicial equation has two real roots which differ by an integer
If the indicial equation has two real roots such that r1 − r2 is an integer, then one solution
is ∞ X
y1 = xr1 ak xk ,
k=0
and the second linearly independent solution is of the form

∞
X
y2 = xr2 bk xk + C(ln x)y1 ,
k=0
where we plug y2 into the given equation and solve for the constants bk and C .
Remark 1: Note that the case-I also includes complex numbers because in that case
r1 − r2 will be a complex number which cannot be equal to a real integer.
Remark 2: Note that the mai idea is to find at least one Frobenius-type solution. If
we are lucky and find two, we are done. If we only get one, we either use the ideas above
or the method of variation of parameters to obtain a second solution.
39.3 Working Rules
Now we summarize the working steps of the Frobenius method:

∞
X
1. We seek a Frobenius-type solution of the form y = ak xk+r .
k=0
2. We plug this y into the given differential equation.

3. The obtained series must be zero. Setting the first coefficient (usually the coefficient
of xr ) in the series to zero we obtain the indicial equation, which is a quadratic
polynomial in r.
4. If the indicial equation has two real roots r1 and r2 such that r1 − r2 is not an integer,
then find two linearly independent solutions according to Case-I.
5. If the indicial equation has a doubled root r, or the indicial equation has two real
roots such that r1 − r2 is an integer then follow Case-II or Case-III accordingly.
39.3.1 Example
Find the power series solutions about x = 0 of
4xy ′′ + 2y ′ + y = 0
Solution: Clearly, x = 0 is a regular singular point. Comparing with y ′′ +p(x)y ′ +q(x)y = 0

we have xp(x) = 1/2 and x2 q(x) = x/4. We substitute Frobenius series
∞
X
r
y=x cn xn (39.2)
n=0
into the differential equation to get

∞ ∞ ∞
X 1 X 1 X
(n + r)(n + r − 1)cn xn+r−2 + (n + r)cn xn+r−1 + cn xn+r = 0
2x 4x
n=0 n=0 n=0
Multiplying by x2 we obtain
∞ ∞ ∞
X
n+r 1X 1X
(n + r)(n + r − 1)cn x + (n + r)cn xn+r + cn xn+r+1 = 0 (39.3)
2 4
n=0 n=0 n=0
Equating coefficients of xr to zero and noting c0 6= 0 we obtain indicial equation

1
r(r − 1) + r = 0
2
which has roots r = 1/2, 0. These roots are unequal and do not differ by an integer. To
obtain the recurrence relation, we equate to zero the coefficient of xn+r in Equation (39.3)
and obtain
1 1
(n + r)(n + r − 1)cn + (n + r)cn + cn−1 = 0
2 4
Corresponding to r = 1/2 we get
cn−1 (−1)n
(4n2 + 2n)cn + cn−1 = 0 ⇒ cn = − ⇒ cn = −c0
2n(2n + 1) (2n + 1)!
Substituting these values in (39.2), we get one solution as

∞ √ √
√ X (−1)n n √ ( z)3 ( z)5 √

y 1 = c0 x x = c0 z− + + . . . = sin z
(2n + 1)! 3! 5!
n=0
To obtain the second solution we use r = 0 to get

cn−1 (−1)n
(4n2 − 2n)cn + cn−1 = 0 ⇒ cn = − ⇒ cn =
2n(2n − 1) (2n)!
Hence the second solution is

∞
X √
y 2 = c0 xn = cos( z)
n=0
The general solution is given as

√ √
y = b cos( z) + b cos( z)
where a and b are arbitrary constants.
Suggested Readings
New Delhi.
Lesson 40
Series Solutions about a Regular Singular Point (Cont...)
In this lesson we continue series solution about a singular point. We shall demonstrate the
method with some useful differential equations.
40.1.1 Problem 1
Find one series solution of the differential equation
4x2 y ′′ − 4x2 y ′ + (1 − 2x)y = 0,
Solution: Note that x = 0 is a singular point. Let us try

∞
X ∞
X
y = xr ak xk = ak xk+r ,
k=0 k=0
where r is a real number, not necessarily an integer. Again if such a solution exists, it may
only exist for positive x. First let us find the derivatives
∞
X
′
y = (k + r) ak xk+r−1 ,
k=0
X∞
y ′′ = (k + r) (k + r − 1) ak xk+r−2 .
k=0
Plugging into our equation we obtain

∞
X ∞
X ∞
X
4 (k + r) (k + r − 1) ak xk+r − 4 (k + r) ak xk+r+1 + (1 − 2x) ak xk+r = 0
k=0 k=0 k=0
Splitting the last series into two series we get

∞
X ∞
X ∞
X ∞
X
k+r k+r+1 k+r
4(k + r) (k + r − 1) ak x − 4(k + r) ak x + ak x −2 ak xk+r+1 = 0
k=0 k=0 k=0 k=0
Re-indexing leads to
∞
X ∞
X ∞
X ∞
X
k+r k+r k+r
4(k + r) (k + r − 1) ak x − 4(k + r − 1) ak−1 x + ak x − 2ak−1 xk+r = 0
k=0 k=1 k=0 k=1
Combining different series into one series

∞
X
4(k + r) (k + r − 1) + 1 ak − 4(k + r − 1) + 2 ak−1 xk+r .

4r(r − 1) + 1 a0 +
k=1
The indicial equation is given by
4r(r − 1) + 1 = 0
It has a double root at r = 21 . All other coefficients of xk+r also have to be zero so

4(k + r) (k + r − 1) + 1 ak − 4(k + r − 1) + 2 ak−1 = 0.
1
If we plug in r = 2 and solve for ak , we get
4(k + 12 − 1) + 2 1
ak = 1 1 ak−1 = ak−1 .
4(k + 2 ) (k + 2 − 1) + 1 k
Let us set a0 = 1. Then

1 1 1
a1 = a0 = 1, a2 = a1 = ,
1 2 2
1 1 1 1
a3 = a2 = , a4 = a3 = ,...
3 3·2 4 4·3·2
In general, we notice that
1 1
ak = = .
k(k − 1)(k − 2) · · · 3 · 2 k!
In other words,
∞ ∞ ∞
X
k+r
X 1 k+1/2 √ X 1 k √ x
y= ak x = x = x x = xe .
k! k!
k=0 k=0 k=0
So we have one solution of the given differential equation. Here we have written the series
in terms of elementary functions. However this is not always possible.
40.1.2 Problem 2
Solve the Bessel’s equation of order p.
x2 y ′′ + xy ′ + x2 − p2 y = 0.

(40.1)
where 2p is not an integer.

Solution: We take the following generalized power series
∞
X
y= cm xk+m , c0 6= 0. (40.2)
m=0
which implies
∞
X ∞
X
y′ = cm (k + m)xk+m−1 , y ′′ = cm (k + m)(k + m − 1)xk+m−2
m=0 m=0
Substitution for y, y ′, y ′′ in (40.2) gives

∞
X ∞
X ∞
X
k+m−2 k+m−1
x2
cm (k + m)(k + m − 1)x +x cm (k + m)x 2
+ (x − n )2
cm xk+m = 0
m=0 m=0 m=0
Combining the first two series we obatin

∞
X ∞
X
cm (k + m)(k + m − 1) + (k + m) − p 2
xk+m + cm xk+m+2 = 0
m=0 m=0
Further simplifications leads to

∞
X ∞
X
cm (k + m + p)(k + m − p)xk+m + cm xk+m+2 = 0 (40.3)
m=0 m=0
Equating the smallest power of x to zero, we get the indicial equation as
c0 (k + p)(k − p) = 0, i.e, (k + p)(k − p) = 0, as c0 6= 0.
So the roots of indicial equation are k = p, −p. Next equating to zero the coefficient of
xk+1 in (40.3) gives
c1 (k + 1 + p)(k + 1 − p) = 0, so that c1 = 0 for k = p and − p.
Finally equating to zero the coefficient of xk+m in (40.3) gives
cm (k + m + p)(k + m − p) + cm−2 = 0
1
⇒ cm = cm−2 .
(k + m + p)(p − k − m)
1
⇒ cm = cm−2 . (40.4)
(k + m + p)(p − k − m)
Putting m = 3, 5, 7, . . . in (40.4) and using c1 = 0, we find
c1 = c3 = c5 = c7 = . . . = 0.
Putting m = 2, 4, 6, . . . in (40.4), we find

1
c2 = c0
(k + 2 + p)(p − k − 2)
1 1
c4 = c2 = c0
(k + 4 + p)(p − k − 4) (k + 4 + p)(p − k − 4)(k + 2 + p)(p − k − 2)
and so on. Putting these values in (40.2) and also replacing c0 by 1, we get
x2 x4

y = 1+ + + ...
(k + 2 + p)(p − k − 2) (k + 4 + p)(p − k − 4)(k + 2 + p)(p − k − 2)
Replacing k by p and −p in the above equation gives
(−1)k x2k
∞
x2
X
p
y1 = x 1− + . . . = xp
4(1 + p) 22k k!(k + p)(k − 1 + p) · · · (2 + p)(1 + p)
k=0
(−1)k x2k
∞
x2
X
−p
y2 = x 1− + . . . = x−p
4(1 − p) 22k k!(k − p)(k − 1 − p) · · · (2 − p)(1 − p)
k=0
Therefore when 2p is not an integer, we have the general solution to Bessel’s equation of
order p
y = c1 y1 (x) + c2 y2 (x),
where c1 and c2 are arbitrary constants.
Remark: We define the Bessel functions of the first kind Bessel function of the first
kind of order p and −p as
(−1)k
∞
1 X x 2k+p
Jp (x) = p y1 = ,
2 Γ(1 + p) k!Γ(k + p + 1) 2
k=0
(−1)k
∞
1 X x 2k−p
J−p (x) = −p y2 = .
2 Γ(1 − p) k!Γ(k − p + 1) 2
k=0
As these are constant multiples of the solutions we found above, these are both solutions to
Bessel’s equation of order p. When p is not an integer, Jp and J−p are linearly independent.
When 2p is an integer we obtain
(−1)k x 2k+p
∞
X
Jp (x) = .
k!(k + p)! 2
k=0
In this case it turns out that

Jp (x) = (−1)n J−p (x),
and so in that case we do not obtain a second linearly independent solution.
40.1.3 Problem 3
Find one series solution of xy ′′ + y ′ + y = 0.

Solution: The indicial equation is
r(r − 1) + r = r 2 = 0.
This equation has only one root r = 0. The recursion equation is
(n + r)2 an = −an−1 , n ≥ 1.
The solution with a0 = 1 is

1
an (r) = (−1)n
(r + 1)2 (r + 2)2 · · · (r + n)2
Setting r = 0 gives the solution

∞
X xn
y1 = (−1)n .
(n!)2
n=0
Suggested Readings
New Delhi.
Module-IV: Vector Calculus
Lesson 41
Introduction
41.1 Introduction to Vector Calculus

We first introduce scalar and vector functions and some basic notation and terminology related to
these.
41.1.1 Scalar Function
A scalar function 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) is a function defined at each point in a certain domain 𝐷𝐷 in space. It
takes real values. It depends on the specific point 𝑃𝑃(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) in space, but not on any particular
coordinate system which may be used. For every point (𝑥𝑥, 𝑦𝑦, 𝑧𝑧) ∈ 𝐷𝐷, 𝑓𝑓 takes a real value. We
say the a scalar field 𝑓𝑓 is defined in 𝐷𝐷. For example, The distance function in the three
dimensional space taken as the Euclidean distance between the points 𝑃𝑃(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) and
𝑃𝑃0 (𝑥𝑥0 , 𝑦𝑦0, 𝑧𝑧0 )
𝑓𝑓(𝑃𝑃) = 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = �(𝑥𝑥 − 𝑥𝑥0 )2 + (𝑦𝑦 − 𝑦𝑦0 )2 + (𝑧𝑧 − 𝑧𝑧0 )2
defines a scalar field.
41.1.2 Vector function
A vector function is defined at each point 𝑃𝑃 ∈ 𝐷𝐷 in three dimensional space by
V =V ( P ) =v1i + v2 j + v3 k
and we say that a vector field is defined in 𝐷𝐷. In Cartesian system of coordinates, it can be
written as
𝑉𝑉 = 𝑣𝑣1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑖𝑖 + 𝑣𝑣2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑗𝑗 + 𝑣𝑣3 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑘𝑘.
An example of a vector field is the velocity field 𝑉𝑉(𝑃𝑃) defined at any point 𝑃𝑃 on a rotating body.
41.1.3 Level surface
Let 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) be a single valued continuous scalar function defined at every point 𝑃𝑃 ∈ 𝐷𝐷. Then
an equation of a surface is defined by 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑐𝑐, a constant. It is called a level surface of the
function.
41.1.4 Example : We determine the level surface of the scalar field in space, defined by the
following function f ( x, y, z ) = x + y + z.
Introduction
We find that 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑐𝑐 gives x + y + z =c which is equation of a plane. For different c they
define parallel planes. Therefore, the level surfaces are parallel planes.
41.1.5 Example: Determine the level surface of the scalar field in space, defined by the
function f ( x, y, z ) =x 2 + 9 y 2 + 16 z 2 .
Note that f ( x, y, z ) = c gives x 2 + 9 y 2 + 16 z 2 =

c which defines ellipsoids. So the level surfaces
are ellipsoids.
41.2 Parametric Representation of Vector Functions
In this section we introduce the parametric representation of vector functions.
41.2.1 Parametric representation of curves
The parametric representation of a curve 𝐶𝐶 in the two dimensional Cartesian plane is given by
=x x ( t=
) , y y (t ) , a ≤ t ≤ b . Using this the position vector of a point 𝑃𝑃 on the curve 𝐶𝐶 can be
r ( t ) x ( t ) i + y (t ) j .
written as =
Therefore, the position vector of a point on a curve defines a vector function. Similarly a three
dimensional curve or a space curve or a space curve 𝐶𝐶 can be parameterized as
𝑟𝑟(𝑡𝑡) = 𝑥𝑥(𝑡𝑡)𝑖𝑖 + 𝑦𝑦(𝑡𝑡)𝑗𝑗 + 𝑧𝑧(𝑡𝑡)𝑘𝑘, 𝑎𝑎 ≤ 𝑡𝑡 ≤ 𝑏𝑏
41.2.2 Parametric Form of a Straight Line
The parametric form of a line passing through a point with position vector a and with the
direction of vector b is given by
𝑟𝑟(𝑡𝑡) = 𝑎𝑎 + 𝑡𝑡𝑡𝑡 = (𝑎𝑎1 + 𝑡𝑡𝑏𝑏1 )𝑖𝑖 + (𝑎𝑎2 + 𝑡𝑡𝑏𝑏2 )𝑗𝑗 + (𝑎𝑎3 + 𝑡𝑡𝑡𝑡3 )𝑘𝑘
41.2.3 Parametric Form of a Circle
The parametric form of the circle 𝑥𝑥 2 + 𝑦𝑦 2 = 𝑎𝑎2 , is defined by
𝑟𝑟(𝑡𝑡) = 𝑎𝑎 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑖𝑖 + 𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑗𝑗
41.2.4 Parametric Form of an Ellipse
The parametric form of the ellipse
2
Introduction
x2 y 2
+ =
1
a 2 b2
is given by
𝑟𝑟(𝑡𝑡) = 𝑎𝑎 cos 𝑡𝑡 𝑖𝑖 + 𝑏𝑏 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑗𝑗
41.2.5 Parametric Form of a Parabola:
Let us consider the parabola 𝑦𝑦 2 = 4𝑎𝑎𝑥𝑥. Now take 𝑦𝑦 = 𝑡𝑡 as one parameter and then we can write
the parametric form of the parabola as
𝑡𝑡 2
𝑟𝑟(𝑡𝑡) = �4𝑎𝑎 � 𝑖𝑖 + 𝑡𝑡𝑡𝑡
41.2.6 Parametric Representation of Surfaces
We can give parametric representation of surfaces can be done using two parameters. Let
𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑐𝑐 or 𝑔𝑔(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 0 be the equation of a surface. Let an explicit representation of the
surface be written as 𝑧𝑧 = ℎ(𝑥𝑥, 𝑦𝑦). Then, if we substitute 𝑢𝑢 = 𝑥𝑥, 𝑦𝑦 = 𝑣𝑣, the parametric form of
the surface can be reduced to
𝑟𝑟(𝑢𝑢, 𝑣𝑣) = 𝑢𝑢𝑢𝑢 + 𝑣𝑣𝑣𝑣 + ℎ(𝑢𝑢, 𝑣𝑣)𝑘𝑘.
41.2.7 Example
The parametric representation of the cylinder 𝑥𝑥 2 + 𝑦𝑦 2 = 𝑎𝑎2 is
𝑟𝑟(𝑢𝑢, 𝑣𝑣) = 𝑎𝑎 cos 𝑢𝑢 𝑖𝑖 + 𝑎𝑎 sin 𝑢𝑢 𝑗𝑗 + 𝑣𝑣 𝑘𝑘.
41.2.8 Example
The parametric representation of the sphere 𝑥𝑥 2 + 𝑦𝑦 2 + 𝑧𝑧 2 = 𝑎𝑎2 is given by
𝑟𝑟(𝑢𝑢, 𝑣𝑣) = 𝑎𝑎 cos 𝑢𝑢 cos 𝑣𝑣 𝑖𝑖 + 𝑎𝑎 sin 𝑢𝑢 cos 𝑣𝑣 𝑗𝑗 + 𝑎𝑎 sin 𝑣𝑣 𝑘𝑘, 0 ≤ 𝑢𝑢 ≤ 2𝜋𝜋, −𝜋𝜋/2 ≤ 𝑣𝑣 ≤ 𝜋𝜋/2
41.2.9 Example
The parametric representation of the ellipsoid
𝑥𝑥 2 𝑦𝑦 2 𝑧𝑧 2
+ 𝑏𝑏 2 + 𝑐𝑐 2 = 1
𝑎𝑎 2
is given by
𝑟𝑟(𝑢𝑢, 𝑣𝑣) = 𝑎𝑎 cos 𝑢𝑢 cos 𝑣𝑣 𝑖𝑖 + 𝑏𝑏 sin 𝑢𝑢 cos 𝑣𝑣 𝑗𝑗 + 𝑐𝑐 sin 𝑣𝑣 𝑘𝑘
3
Introduction
41.3 Limit, Continuity and Differentiability of Vector Function

In this section the analytical concepts of the limit, continuity and differentiability of vector
function are introduced.
41.3.1 Limit of Vector Function
The vector function 𝑣𝑣(𝑡𝑡) is said to have the limit p as 𝑡𝑡 → 𝑙𝑙 if 𝑣𝑣(𝑡𝑡) is defined in some
neighbourhood of l , except possibly at 𝑡𝑡 = 𝑙𝑙, and
lim𝑡𝑡→𝑙𝑙 |𝑣𝑣(𝑡𝑡) − 𝑝𝑝| = 0
We write 𝑝𝑝. In the Cartesian system, this implies that limits of the component functions
𝑣𝑣1 (𝑡𝑡), 𝑣𝑣2 (𝑡𝑡) and 𝑣𝑣3 (𝑡𝑡) exist as 𝑡𝑡 → 𝑙𝑙 and
lim𝑡𝑡→𝑙𝑙 𝑣𝑣1 (𝑡𝑡) = 𝑝𝑝1 , lim𝑡𝑡→𝑙𝑙 𝑣𝑣2 (𝑡𝑡) = 𝑝𝑝2 , lim𝑡𝑡→𝑙𝑙 𝑣𝑣3 (𝑡𝑡) = 𝑝𝑝3
where 𝑝𝑝 = 𝑝𝑝1 𝑖𝑖 + 𝑝𝑝 𝑗𝑗 + 𝑝𝑝3 𝑘𝑘.
41.3.2 Continuity
A vector function 𝑣𝑣(𝑡𝑡) is defined to be continuous at 𝑡𝑡 = 𝑙𝑙, if
(𝑖𝑖) 𝑣𝑣(𝑡𝑡) is defined in some neighbourhood of l , (ii)lim𝑡𝑡→𝑙𝑙 𝑣𝑣(𝑡𝑡) exists, and (iii)lim𝑡𝑡→𝑙𝑙 𝑣𝑣(𝑡𝑡) =
𝑣𝑣(𝑙𝑙).
In Cartesian system, this implies that 𝑣𝑣(𝑡𝑡)is continuous at 𝑡𝑡 = 𝑙𝑙, if and only if the component
functions 𝑣𝑣1 (𝑡𝑡), 𝑣𝑣2 (𝑡𝑡) and 𝑣𝑣3 (𝑡𝑡) are continuous at 𝑡𝑡 = 𝑙𝑙.
41.3.3 Differentiability
A vector function 𝑣𝑣(𝑡𝑡) is said to be differentiable at a point, if the limit

𝑣𝑣(𝑡𝑡+∆𝑡𝑡)−𝑣𝑣(𝑡𝑡)
lim ∆𝑡𝑡
∆𝑡𝑡→0
𝑑𝑑𝑑𝑑
exists. If the limit exists, then we write it as 𝑣𝑣 ′ (𝑡𝑡) or as 𝑑𝑑𝑑𝑑
.
In Cartesian system, this implies that the component functions 𝑣𝑣1 (𝑡𝑡), 𝑣𝑣2 (𝑡𝑡) and 𝑣𝑣3 (𝑡𝑡) are
differentiable at a point 𝑡𝑡, and the limits
𝑣𝑣𝑖𝑖 (𝑡𝑡+∆𝑡𝑡)−𝑣𝑣𝑖𝑖 (𝑡𝑡)
lim ∆𝑡𝑡
, 𝑖𝑖 = 1,2,3 exist.
∆𝑡𝑡→0
Therefore, 𝑣𝑣 ′ (𝑡𝑡) = 𝑣𝑣1 ′ (𝑡𝑡)𝑖𝑖 + 𝑣𝑣2 ′ (𝑡𝑡)𝑗𝑗 + 𝑣𝑣3 ′ (𝑡𝑡)𝑘𝑘
Let 𝑣𝑣(𝑡𝑡) = 𝑟𝑟(𝑡𝑡) = 𝑥𝑥(𝑡𝑡)𝑖𝑖 + 𝑦𝑦(𝑡𝑡)𝑗𝑗 + 𝑧𝑧(𝑡𝑡)𝑘𝑘 be the parametric representation of a curve 𝐶𝐶.
4
Introduction
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 (𝑡𝑡) 𝑑𝑑𝑑𝑑 (𝑡𝑡) 𝑑𝑑𝑑𝑑 (𝑡𝑡)

Then 𝑑𝑑𝑑𝑑
= 𝑟𝑟 ′ (𝑡𝑡) = 𝑑𝑑𝑑𝑑
𝑖𝑖 + 𝑗𝑗 + 𝑘𝑘.
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
41.3.4 Example
Let us consider the represent of the parabola 𝑦𝑦 = 1 − 2𝑥𝑥 2 , −1 ≤ 𝑥𝑥 ≤ 1 in parametric form.

𝜋𝜋
Using this we will find 𝑟𝑟 ′ (0) and 𝑟𝑟 ′ ( 4 ).
𝜋𝜋 𝜋𝜋
Assume x = sin t . Then 𝑦𝑦 = 1 − 2𝑠𝑠𝑠𝑠𝑠𝑠2 𝑡𝑡 = 𝑐𝑐𝑐𝑐𝑐𝑐2𝑡𝑡 , The range of t is − 2 ≤ 𝑡𝑡 ≤ 2 . So
𝜋𝜋 𝜋𝜋
𝑟𝑟(𝑡𝑡) = 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑖𝑖 + 𝑐𝑐𝑐𝑐𝑐𝑐2𝑡𝑡 𝑗𝑗, − 2 ≤ 𝑡𝑡 ≤ 2
Therefore 𝑟𝑟 ′ (𝑡𝑡) = 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑖𝑖 − 2𝑠𝑠𝑠𝑠𝑠𝑠2𝑡𝑡 𝑗𝑗 ,

𝜋𝜋 𝑖𝑖
Further, 𝑟𝑟 ′ (0) = 𝑖𝑖, 𝑟𝑟 ′ �4 � = � � − 2𝑗𝑗. The tangent at 𝑡𝑡 = 0 is parallel to 𝑥𝑥-axis.
√2
It may be noted that 𝑡𝑡 = 0 gives 𝑥𝑥 = 0, 𝑦𝑦 = 1 which is the vertex of the parabola.
41.3.5 Example
We find the tangent vector to the curve with parametric representation given by
𝑡𝑡+1
𝑥𝑥 = 𝑡𝑡 3 , 𝑦𝑦 = 𝑡𝑡
, 𝑧𝑧 = 𝑡𝑡 2 + 1, at the point 𝑡𝑡 = 2.
We will also find the parametric representation of the tangent vector.
First note that the position vector of a point on the given curve is
1
𝑟𝑟(𝑡𝑡) = 𝑡𝑡 3 𝑖𝑖 + �1 + 𝑡𝑡 � 𝑗𝑗 + (𝑡𝑡 2 + 1)𝑘𝑘, 𝑡𝑡 ≠ 0.
Therefore the tangent vector is

1
𝑟𝑟 ′ (𝑡𝑡) = 3𝑡𝑡 2 𝑖𝑖 − 𝑡𝑡 2 𝑗𝑗 + 2𝑡𝑡 𝑘𝑘
1
and 𝑟𝑟 ′ (2) = 12𝑖𝑖 − 4 𝑗𝑗 + 4𝑘𝑘 .
3
The position vector of the point at which 𝑟𝑟 ′ (2) is the tangent is 𝑟𝑟(2) = 8𝑖𝑖 + 2 𝑗𝑗 + 5𝑘𝑘.
Therefore we require the position vector of a point on the line passing through the point whose
position vector is 𝑟𝑟(2) and has the direction of 𝑟𝑟 ′ (2). Hence, parametric form of the line is given
by
3 𝑡𝑡 3 1
𝑥𝑥 = 8 + 12𝑡𝑡, 𝑦𝑦 = 2 − 4 , 𝑧𝑧 = 5 + 4𝑡𝑡 or 𝑟𝑟 ′ (𝑡𝑡) = �8, 2 , 5� + 𝑡𝑡(12, − 4 , 4).
5
Introduction
41.3.6 Higher Order Derivatives and Rules of Differentiation
Assuming that the existence of derivatives, we have the following results
𝑣𝑣 ′′ (𝑡𝑡) = 𝑣𝑣1 ′′ (𝑡𝑡)𝑖𝑖 + 𝑣𝑣2 ′′ (𝑡𝑡)𝑗𝑗 + 𝑣𝑣3 ′′ (𝑡𝑡)𝑘𝑘
(𝑢𝑢 + 𝑣𝑣)′ = 𝑢𝑢′ + 𝑣𝑣 ′
(𝑓𝑓(𝑡𝑡)𝑢𝑢(𝑡𝑡))′ = 𝑓𝑓 ′ (𝑡𝑡)𝑢𝑢(𝑡𝑡) + 𝑓𝑓(𝑡𝑡)𝑢𝑢′ (𝑡𝑡)
where 𝑓𝑓(𝑡𝑡) is any real valued scalar function.
(𝑢𝑢(𝑡𝑡). 𝑣𝑣(𝑡𝑡))′ = 𝑢𝑢(𝑡𝑡). 𝑣𝑣 ′ (𝑡𝑡) + 𝑢𝑢′ (𝑡𝑡). 𝑣𝑣(𝑡𝑡)
(𝑢𝑢(𝑡𝑡) × 𝑣𝑣(𝑡𝑡))′ = 𝑢𝑢(𝑡𝑡) × 𝑣𝑣 ′ (𝑡𝑡) + 𝑢𝑢′ (𝑡𝑡) × 𝑣𝑣(𝑡𝑡)
where . and × represent the dot and cross products, respectively. It must be mentioned that the
cross product of two vectors is not commutative.
41.3.7 Example
Find 𝑣𝑣 ′ (𝑡𝑡) in each of the following cases.
(𝑖𝑖) 𝑣𝑣(𝑡𝑡) = (cos 𝑡𝑡 + 𝑡𝑡 2 )(𝑡𝑡𝑡𝑡 + 𝑗𝑗 + 2𝑘𝑘) (𝑖𝑖𝑖𝑖) 𝑣𝑣(𝑡𝑡) = (3𝑡𝑡𝑡𝑡 + 5𝑡𝑡 2 𝑗𝑗 + 6𝑘𝑘). (𝑡𝑡 2 𝑖𝑖 − 2𝑡𝑡𝑡𝑡 + 𝑡𝑡𝑡𝑡)
Solution
(𝒊𝒊) 𝑣𝑣 ′ (𝑡𝑡) = (cos 𝑡𝑡 + 𝑡𝑡 2 )′ (𝑡𝑡𝑡𝑡 + 𝑗𝑗 + 2𝑘𝑘) + (cos 𝑡𝑡 + 𝑡𝑡 2 )(𝑡𝑡𝑡𝑡 + 𝑗𝑗 + 2𝑘𝑘)′
= (− sin 𝑡𝑡 + 2𝑡𝑡)(𝑡𝑡𝑡𝑡 + 𝑗𝑗 + 2𝑘𝑘) + (cos 𝑡𝑡 + 𝑡𝑡 2 )(𝑖𝑖)
= (3𝑡𝑡 2 + 𝑡𝑡 sin 𝑡𝑡 + cos 𝑡𝑡 )𝑖𝑖 + (2𝑡𝑡 − sin 𝑡𝑡)(𝑗𝑗 + 2𝑘𝑘)
(𝒊𝒊𝒊𝒊) 𝑣𝑣 ′ (𝑡𝑡) = (3𝑡𝑡𝑡𝑡 + 5𝑡𝑡 2 𝑗𝑗 + 6𝑘𝑘)′ . (𝑡𝑡 2 𝑖𝑖 − 2𝑡𝑡𝑡𝑡 + 𝑡𝑡𝑡𝑡) + (3𝑡𝑡𝑡𝑡 + 5𝑡𝑡 2 𝑗𝑗 + 6𝑘𝑘). (𝑡𝑡 2 𝑖𝑖 −
2𝑡𝑡𝑡𝑡 + 𝑡𝑡𝑡𝑡)′
= (3𝑖𝑖 + 10𝑡𝑡 𝑗𝑗)(𝑡𝑡 2 𝑖𝑖 + 2𝑡𝑡𝑡𝑡 + 𝑡𝑡𝑡𝑡) + (3𝑡𝑡𝑡𝑡 + 5𝑡𝑡 2 𝑗𝑗 + 6𝑘𝑘). (2𝑡𝑡𝑡𝑡 − 2𝑗𝑗 + 𝑘𝑘)
= 6 − 21𝑡𝑡 2
41.3.8 Length of a Space Curve
Let the curve 𝐶𝐶 represented in parametric form as 𝑟𝑟 = 𝑟𝑟(𝑡𝑡), 𝑎𝑎 ≤ 𝑡𝑡 ≤ 𝑏𝑏. In Cartesian system, we
have 𝑟𝑟(𝑡𝑡) = 𝑥𝑥(𝑡𝑡)𝑖𝑖 + 𝑦𝑦(𝑡𝑡)𝑗𝑗 + 𝑧𝑧(𝑡𝑡)𝑘𝑘. Then, the length of the curve is given by
6
Introduction
𝑏𝑏 2 2 2 𝑏𝑏
′ (𝑡𝑡)� ′ (𝑡𝑡)� ′ (𝑡𝑡)�
𝑙𝑙 = � [�𝑥𝑥 + �𝑦𝑦 + �𝑧𝑧 1/2
] 𝑑𝑑𝑑𝑑 = � [𝑟𝑟 ′ (𝑡𝑡). 𝑟𝑟 ′ (𝑡𝑡)]1/2
𝑎𝑎 𝑎𝑎
We observe that the integrand is the norm of 𝑟𝑟 ′ (𝑡𝑡), that is

2 2 2
||𝑟𝑟 ′ (𝑡𝑡)|| = [�𝑥𝑥 ′ (𝑡𝑡)� + �𝑦𝑦 ′ (𝑡𝑡)� + �𝑧𝑧 ′ (𝑡𝑡)� ]1/2
Then, we can write

𝑏𝑏
𝑙𝑙 = ∫𝑎𝑎 ||𝑟𝑟 ′ (𝑡𝑡)|| 𝑑𝑑𝑑𝑑
Sometimes the notation |𝑟𝑟 ′ (𝑡𝑡)| is also used instead of ||𝑟𝑟 ′ (𝑡𝑡)||.
Now, define the real valued function 𝑠𝑠(𝑡𝑡) as
𝑡𝑡 2 2 2 𝑡𝑡
𝑠𝑠(𝑡𝑡) = ∫𝑎𝑎 [�𝑥𝑥 ′ (𝜉𝜉)� + �𝑦𝑦 ′ (𝜉𝜉)� + �𝑧𝑧 ′ (𝜉𝜉)� ]1/2 𝑑𝑑𝑑𝑑 = ∫𝑎𝑎 ||𝑟𝑟 ′ (𝜉𝜉)|| 𝑑𝑑𝑑𝑑 (41.3.1)
Then, 𝑠𝑠(𝑡𝑡) is the arc length of the curve from its initial point (𝑥𝑥(𝑎𝑎), 𝑦𝑦(𝑎𝑎), 𝑧𝑧(𝑎𝑎)) to an arbitrary
point (𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) on the curve 𝐶𝐶. Therefore, 𝑠𝑠(𝑡𝑡) is the length function. Using relation
(41.3.1), it is possible to solve for 𝑡𝑡 as a function of 𝑠𝑠 , that is 𝑡𝑡 = 𝑠𝑠(𝑡𝑡). Then the curve 𝐶𝐶 can be
parameterised in terms of the arc length 𝑠𝑠 as
𝑟𝑟(𝑠𝑠) = 𝑟𝑟(𝑡𝑡(𝑠𝑠)) = 𝑥𝑥�𝑡𝑡(𝑠𝑠)�𝑖𝑖 + 𝑦𝑦�𝑡𝑡(𝑠𝑠)�𝑗𝑗 + 𝑧𝑧(𝑡𝑡(𝑠𝑠))𝑘𝑘
41.3.9 Example
We try to find the length of the Helix which is given by
𝑟𝑟(𝑡𝑡) = 𝑎𝑎 cos 𝑡𝑡 𝑖𝑖 + 𝑎𝑎 sin 𝑡𝑡 𝑗𝑗 + 𝑐𝑐𝑐𝑐 𝑘𝑘, 𝑎𝑎 > 0,0 ≤ 𝑡𝑡 ≤ 2𝜋𝜋.
First note that we can write
=x (t ) a=
cos t , y ( t ) a=
sin t , z ( t ) ct .
Hence 𝑥𝑥 ′ (𝑡𝑡) = − asin 𝑡𝑡 , 𝑦𝑦 ′ (𝑡𝑡) = acos 𝑡𝑡, 𝑧𝑧 = 𝑐𝑐.
Therefore , we have
2𝜋𝜋
𝑠𝑠 = arc length = ∫0 [𝑎𝑎2 𝑠𝑠𝑠𝑠𝑠𝑠2 𝑡𝑡 + 𝑎𝑎2 𝑐𝑐𝑐𝑐𝑐𝑐 2 𝑡𝑡 + 𝑐𝑐 2 ]1/2 𝑑𝑑𝑑𝑑 = (2𝜋𝜋)(𝑎𝑎2 + 𝑐𝑐 2 )1/2
7
Introduction
Suggested Readings
Courant, R. and John, F. (1989), Introduction to Calculus and Analysis, Vol. II, Springer-Verlag,
New York.
Jain, R.K. and Iyengar, S.R.K. (2002) Advanced Engineering Mathematics, Narosa Publishing
House, New Delhi.
Jordan, D.W. and Smith, P. (2002) Mathematical Techniques, Oxford University Press, Oxford.
Kreyszig, E. (1999) Advanced Engineering Mathematics, John Wiley, New York.
Piskunov, N. (1974) Differentail and Integral Calculus, Vol. II, MIR Publishers, Moscow.
Wylie, C. R. and Barrett, L.C. (2003) Advanced Engineering Mathematics, Tata McGraw-Hill,
New Delhi.
8
Lesson 42
Gradient and Directional Derivative
42.1 Gradient of a Scalar Field

Let 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) be a real valued function defining a scalar field. To define the gradient of a scalar
field, we first introduce a vector operator called del operator denoted by ∇. We define the vector
differential operator in two and three dimensions as
𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕
∇= 𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 and ∇= 𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕
The gradient of a scalar field 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧), denoted by ∇𝑓𝑓 or grad (𝑓𝑓) is defined as
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
∇f = 𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕
Note that the del operator ∇ operates on a scalar field and produces a vector field.
42.1. 1 Example
Find the gradient of the following scalar fields
(𝑖𝑖) 𝑓𝑓(𝑥𝑥, 𝑦𝑦) = 𝑦𝑦 2 − 4𝑥𝑥𝑥𝑥 at (1,2),
Solution
𝜕𝜕 𝜕𝜕
∇f(x, y) = �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 � (𝑦𝑦 2 − 4𝑥𝑥𝑥𝑥) = −4𝑦𝑦 𝑖𝑖 + (2𝑦𝑦 − 4𝑥𝑥)𝑗𝑗
42.1. 2 Example
1
𝒓𝒓 = 𝑥𝑥𝑥𝑥 + 𝑦𝑦𝑦𝑦 + 𝑧𝑧𝑧𝑧, |𝒓𝒓| = 𝑟𝑟 and 𝑟𝑟̂ = 𝒓𝒓/𝑟𝑟, then show that grad�𝑟𝑟 � = −𝑟𝑟̂ /𝑟𝑟 2 .
Solution
1 𝜕𝜕 𝜕𝜕 𝜕𝜕 1 1 𝜕𝜕𝜕𝜕 1 𝜕𝜕𝜕𝜕 1 𝜕𝜕𝜕𝜕 1 𝑥𝑥 𝑦𝑦
Grad�𝑟𝑟 � = �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � �𝑟𝑟 � = 𝑖𝑖 �− 𝑟𝑟 2 𝜕𝜕𝜕𝜕 � + 𝑗𝑗 �− 𝑟𝑟 2 𝜕𝜕𝜕𝜕 � + 𝑘𝑘 �− 𝑟𝑟 2 𝜕𝜕𝜕𝜕 � = − 𝑟𝑟 2 �𝑟𝑟 𝑖𝑖 + 𝑟𝑟 𝑗𝑗 +
𝑧𝑧
𝑟𝑟
𝑘𝑘�
1 𝒓𝒓 𝑟𝑟̂
= − 𝑟𝑟 2 �𝑟𝑟 � = − 𝑟𝑟 2
where 𝑟𝑟̂ = (𝑥𝑥𝑥𝑥 + 𝑦𝑦𝑦𝑦 + 𝑧𝑧𝑧𝑧)/𝑟𝑟
42.1. 3 Geometrical Representation of the Gradient
Let 𝑓𝑓(𝑃𝑃) = 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) be a differentiable scalar field. Let 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑘𝑘 be a level surface and
𝑃𝑃0 (𝑥𝑥0 , 𝑦𝑦0 , 𝑧𝑧0 ) be a point on it. There are infinite number of smooth curves on the surface passing
through the point 𝑃𝑃0 . Each of these curves has a tangent at 𝑃𝑃0 . The totality of these tangent lines
form a tangent plane to the surface at a point 𝑃𝑃0 . A vector normal to this plane at 𝑃𝑃0 is called the
normal vector to the surface at this point.
Consider now a smooth curve 𝐶𝐶 on the surface passing through a point 𝑃𝑃 on the surface. Let
𝑥𝑥 = 𝑥𝑥(𝑡𝑡), 𝑦𝑦 = 𝑦𝑦(𝑡𝑡), 𝑧𝑧 = 𝑧𝑧(𝑡𝑡) be the parametric representation of the curve 𝐶𝐶. Any point 𝑃𝑃 on 𝐶𝐶
has the position vector 𝑟𝑟(𝑡𝑡) = 𝑥𝑥(𝑡𝑡)𝑖𝑖 + 𝑦𝑦(𝑡𝑡)𝑗𝑗 + 𝑧𝑧(𝑡𝑡)𝑘𝑘. Since the curve lies on the surface, we
have
𝑓𝑓(𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) = 𝑘𝑘

𝑑𝑑
Then, 𝑑𝑑𝑑𝑑 𝑓𝑓(𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) = 0
𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑

By chain rule, we have 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
+ 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
+ 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
=0
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑

or �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . �𝑖𝑖 𝑑𝑑𝑑𝑑 + 𝑗𝑗 𝑑𝑑𝑑𝑑 + 𝑘𝑘 𝑑𝑑𝑑𝑑 � = 0
or ∇𝑓𝑓. 𝑟𝑟 ′ (𝑡𝑡) = 0
Let ∇𝑓𝑓(𝑃𝑃) ≠ 0 and 𝑟𝑟 ′ (𝑡𝑡) ≠ 0. Now 𝑟𝑟 ′ (𝑡𝑡) is a tangent to 𝐶𝐶 at the point 𝑃𝑃 and lies in the tangent
plane to the surface at . Hence ∇𝑓𝑓(𝑃𝑃) is orthogonal to every tangent vector at 𝑃𝑃. Therefore,
∇𝑓𝑓(𝑃𝑃) is the vector normal to the surface 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑘𝑘 at the point 𝑃𝑃.
42.1. 4 Example
We will find a unit normal vector to the surface 𝑥𝑥𝑦𝑦 2 + 2𝑦𝑦𝑦𝑦 = 8 at the point (3, −2,1).
Let 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑥𝑥𝑦𝑦 2 + 2𝑦𝑦𝑦𝑦 = 8 then

𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕𝜕𝜕
= 𝑦𝑦 2 , 𝜕𝜕𝜕𝜕 = 2𝑥𝑥𝑥𝑥 + 2𝑧𝑧 and 𝜕𝜕𝜕𝜕
= 2𝑦𝑦
Therefore
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
∇f = 𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 = 𝑦𝑦 2 𝑖𝑖 + (2𝑥𝑥𝑥𝑥 + 2𝑧𝑧)𝑗𝑗 + 2𝑦𝑦𝑦𝑦
At (3, −2,1), we obtain the normal vector as ∇f(3, −2,1) = 4i − 10j − 4k. The unit normal
vector at (3, −2,1) is given by
4𝑖𝑖−10𝑗𝑗 −4𝑘𝑘 2𝑖𝑖−5𝑗𝑗 −2𝑘𝑘
= .
√16+100+16 √33
2
42.1. 5 Example
Here we will find the angle between the two surfaces 𝑥𝑥𝑥𝑥𝑥𝑥𝑥𝑥 𝑧𝑧 = 𝑦𝑦 2 − 1 and 𝑥𝑥 2 𝑦𝑦 = 2 − 𝑧𝑧 at the
given point (1,1,1).
First note that the angle between two surfaces at a common point is the angle between their
normals at that point. Now we have
𝑓𝑓1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑥𝑥𝑥𝑥𝑥𝑥𝑥𝑥 𝑧𝑧 − 𝑦𝑦 2 + 1 = 0, ∆𝑓𝑓1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = (log 𝑧𝑧)𝑖𝑖 − 2𝑦𝑦𝑦𝑦 + (𝑥𝑥/𝑧𝑧)𝑘𝑘
∆𝑓𝑓1 (1,1,1) = −2𝑗𝑗 + 𝑘𝑘 = 𝑛𝑛1
𝑓𝑓2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑥𝑥 2 𝑦𝑦 − 2 + 𝑧𝑧 = 0, ∆𝑓𝑓2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 2𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑥𝑥 2 𝑗𝑗 + 𝑘𝑘
∆𝑓𝑓2 (1,1,1) = 2𝑖𝑖 + 𝑗𝑗 + 𝑘𝑘 = 𝑛𝑛2

𝑛𝑛 .𝑛𝑛 1 1
Therefore cos 𝜃𝜃 = �|𝑛𝑛 1||𝑛𝑛2 |� = or 𝜃𝜃 = 𝑐𝑐𝑐𝑐𝑐𝑐 −1 � �.
1 2 √30 √30
42.1.6 Properties of Gradient
Let 𝑓𝑓 and 𝑔𝑔 be any two differentiable scalar fields. The gradient satidfies the following algebraic
properties,
∆(𝑓𝑓 + 𝑔𝑔) = ∆𝑓𝑓 + ∆𝑔𝑔
∆(𝑐𝑐1 𝑓𝑓 + 𝑐𝑐2 𝑔𝑔) = 𝑐𝑐1 ∆𝑓𝑓 + 𝑐𝑐2 ∆𝑔𝑔, where 𝑐𝑐1 , 𝑐𝑐2 are arbitrary constants
∆(𝑓𝑓𝑓𝑓) = 𝑓𝑓∆𝑔𝑔 + 𝑔𝑔∆𝑓𝑓

𝑓𝑓 𝑔𝑔∆𝑓𝑓−𝑓𝑓∆𝑔𝑔
∆ �𝑔𝑔 � = 𝑔𝑔 2
42.2 Directional Derivative

Let 𝑓𝑓(𝑃𝑃) = 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) be a differentiable scalar field.
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
Then 𝜕𝜕𝜕𝜕 , 𝜕𝜕𝜕𝜕 , 𝜕𝜕𝜕𝜕 denotes the rates of change of 𝑓𝑓 in the direction of 𝑥𝑥, 𝑦𝑦 and 𝑧𝑧 axis, respectively.
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

If 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑘𝑘 is the level surface and 𝑃𝑃0 is any point, then 𝜕𝜕𝜕𝜕
, 𝜕𝜕𝜕𝜕
, 𝜕𝜕𝜕𝜕 at 𝑃𝑃0 (𝑥𝑥0 , 𝑦𝑦0 , 𝑧𝑧0 ) denote
the slopes of the tangent lines in the directions of 𝑖𝑖, 𝑗𝑗, 𝑘𝑘 respectively. It is natural to give the
definition of derivative in any direction which we call as the directional derivative.
Let 𝑏𝑏� = 𝑏𝑏1 𝑖𝑖 + 𝑏𝑏2 𝑗𝑗 + 𝑏𝑏3 𝑘𝑘 be any unit vector. Let 𝑃𝑃0 be any point 𝑃𝑃0 : 𝑎𝑎 = 𝑎𝑎1 𝑖𝑖 + 𝑎𝑎2 𝑗𝑗 + 𝑎𝑎3 𝑘𝑘.
Then, the position vector of any point 𝑄𝑄 on the line passing through 𝑃𝑃0 and in the direction of 𝑏𝑏�
is given by
3
𝑟𝑟 = 𝑎𝑎 + 𝑡𝑡𝑏𝑏� = (𝑎𝑎1 + 𝑡𝑡𝑏𝑏1 )𝑖𝑖 + (𝑎𝑎2 + 𝑡𝑡𝑏𝑏2 )𝑗𝑗 + (𝑎𝑎3 + 𝑡𝑡𝑡𝑡3 )𝑘𝑘 = 𝑥𝑥(𝑡𝑡)𝑖𝑖 + 𝑦𝑦(𝑡𝑡)𝑗𝑗 + 𝑧𝑧(𝑡𝑡)𝑘𝑘
This is, the point 𝑄𝑄(𝑎𝑎1 + 𝑡𝑡𝑏𝑏1 , 𝑎𝑎2 + 𝑡𝑡𝑏𝑏2 , 𝑎𝑎3 + 𝑡𝑡𝑡𝑡3 ) is on this line. Now, the vector formthe point
𝑃𝑃0 to 𝑄𝑄 is given by 𝑡𝑡𝑏𝑏�. Since | 𝑏𝑏�|=1, the distance from 𝑃𝑃0 to 𝑄𝑄 is 𝑡𝑡. Then
𝜕𝜕𝜕𝜕 𝑓𝑓(𝑄𝑄)−𝑓𝑓(𝑃𝑃)
𝜕𝜕𝜕𝜕
= lim𝑡𝑡→0 𝑡𝑡
if it exists, is called the directional derivative of 𝑓𝑓 at the point 𝑃𝑃0 in the direction to 𝑏𝑏� .
𝜕𝜕
Therefore 𝜕𝜕𝜕𝜕 𝑓𝑓(𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) is rate of change of 𝑓𝑓 with respect to the distance 𝑡𝑡.
We have
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
𝜕𝜕𝜕𝜕
= 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
+ 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
+ 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑
where
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
,
, are evaluated at 𝑡𝑡 = 0 .
We write
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝜕𝜕𝜕𝜕
= �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . �𝑖𝑖 𝑑𝑑𝑑𝑑 + 𝑗𝑗 𝑑𝑑𝑑𝑑 + 𝑘𝑘 𝑑𝑑𝑑𝑑 � = ∇𝑓𝑓. 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
But 𝑑𝑑𝑑𝑑
= 𝑏𝑏�(a unit vector). Therefore, the directional derivative of 𝑓𝑓 in the direction of 𝑏𝑏� in given
by
Directional derivative =∇𝑓𝑓. 𝑏𝑏� = grad(𝑓𝑓). 𝑏𝑏�,
which is denoted by 𝐷𝐷𝑏𝑏 (𝑓𝑓). Note that 𝑏𝑏� is a unit vector. If the direction is specified by a vector
𝑢𝑢, then 𝑏𝑏� = 𝑢𝑢/|𝑢𝑢|.
42.2.1 Example
We will determine the directional derivative of 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑥𝑥𝑦𝑦 2 + 4𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑧𝑧 2 at the point
(1,2,3) in the direction of 3𝑖𝑖 + 4𝑗𝑗 − 5𝑘𝑘.
Consider
∇𝑓𝑓 = (𝑦𝑦 2 + 4𝑦𝑦𝑦𝑦)𝑖𝑖 + (2𝑥𝑥𝑥𝑥 + 4𝑥𝑥𝑥𝑥)𝑗𝑗 + (4𝑥𝑥𝑥𝑥 + 2𝑧𝑧)𝑘𝑘.
At the point (1,2,3), we have ∇𝑓𝑓 = 28𝑖𝑖 + 16𝑗𝑗 + 14𝑘𝑘. The unit vector in the given direction is
𝑏𝑏� = (3𝑖𝑖 + 4𝑗𝑗 − 5𝑘𝑘)/5√2.
Therefore
4
1 78
𝐷𝐷𝑏𝑏 (1,2,3) = 5√2 (28𝑖𝑖 + 16𝑗𝑗 + 14𝑘𝑘). (3𝑖𝑖 + 4𝑗𝑗 − 5𝑘𝑘) = 5√2
Suggested Readings
Courant, R. and John, F. (1989) Introduction to Calculus and Analysis, Vol. II, Springer-Verlag,
New York.
House, New Delhi.
New Delhi.
5
Lesson 43
Divergence and Curl
43.1 Divergence of a Vector Field

Let 𝑣𝑣 = 𝑣𝑣1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑖𝑖 + 𝑣𝑣2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑗𝑗 + 𝑣𝑣3 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑘𝑘 define a vector field.
We define the divergence of vector field as below:
Divergence of 𝑣𝑣, denoted bi div 𝑣𝑣 , is defined as the scalar

𝜕𝜕𝑣𝑣1 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣3
div 𝑣𝑣 = + +
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕𝑣𝑣1 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣3

Also div 𝑣𝑣 = ∇. 𝑣𝑣 = �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . (𝑣𝑣1 𝑖𝑖 + 𝑣𝑣2 𝑗𝑗 + 𝑣𝑣3 𝑘𝑘) = + +
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
43.1.1 Example
Here we will find the divergence of the vector field 𝑣𝑣 = (𝑥𝑥 2 𝑦𝑦 2 − 𝑧𝑧 3 )𝑖𝑖 + 2𝑥𝑥𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥 𝑘𝑘.
Note that we have

𝜕𝜕 𝜕𝜕 𝜕𝜕
div 𝑣𝑣 = 𝜕𝜕𝜕𝜕 (𝑥𝑥 2 𝑦𝑦 2 − 𝑧𝑧 3 ) + 𝜕𝜕𝜕𝜕 (2𝑥𝑥𝑥𝑥𝑥𝑥) + 𝜕𝜕𝜕𝜕 (𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥 )
= 2𝑥𝑥𝑦𝑦 2 + 2𝑥𝑥𝑥𝑥 + 𝑥𝑥𝑥𝑥𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥
43.2 Curl of a Vector Field 𝒗𝒗

Curl of a vector field 𝑣𝑣, denoted by curl 𝑣𝑣, is defined as the vector field
𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣3 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣1
Curl 𝑣𝑣 = � 𝜕𝜕𝜕𝜕3 − � 𝑖𝑖 + � 𝜕𝜕𝜕𝜕1 − � 𝑗𝑗 + � 𝜕𝜕𝜕𝜕2 − � 𝑘𝑘
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
Curl 𝑣𝑣 can also be written in terms of the gradient operator as
𝑖𝑖 𝑗𝑗 𝑘𝑘
𝜕𝜕 𝜕𝜕 𝜕𝜕
Curl 𝑣𝑣 = ∇ × 𝑣𝑣 = � �.
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝑣𝑣1 𝑣𝑣2 𝑣𝑣3
43.2.1 Example
Find the curl of the vector field 𝑣𝑣 = (𝑥𝑥 2 𝑦𝑦 2 − 𝑧𝑧 3 )𝑖𝑖 + 2𝑥𝑥𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥 𝑘𝑘
Solution
Divergence and Curl
𝜕𝜕 𝜕𝜕 𝜕𝜕
Curl 𝑣𝑣 = � 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
�
𝑥𝑥 𝑦𝑦 − 𝑧𝑧 3
2 2
2𝑥𝑥𝑥𝑥𝑥𝑥 𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥
= 𝑖𝑖(𝑥𝑥𝑥𝑥𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥 − 2𝑥𝑥𝑥𝑥) − 𝑗𝑗(𝑦𝑦𝑦𝑦𝑒𝑒 𝑥𝑥𝑥𝑥𝑥𝑥 − 3𝑧𝑧 2 ) + 𝑘𝑘(2𝑦𝑦𝑦𝑦 − 2𝑥𝑥 2 𝑦𝑦)
43.2.2 Curl of Gradient
Let 𝑓𝑓 be a differentiable vector field. Then
Curl(grad𝑓𝑓)=0 or ∇ × ∇𝑓𝑓 = 0
Proof : From the definition, we have
𝜕𝜕 𝜕𝜕 𝜕𝜕
∇ × ∇𝑓𝑓 = ��𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 ��
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓
= 𝑖𝑖 � − � + 𝑗𝑗 � − � + 𝑘𝑘 � − �=0
𝜕𝜕𝜕𝜕𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝜕𝜕𝜕𝜕
43.2.3 Divergence of Curl
Let 𝑣𝑣 be a differentiable vector field. Then
div(curl 𝑣𝑣)=0 or ∇. (∇ × 𝑣𝑣) = 0
proof. Form the definition, we have for 𝑣𝑣 = 𝑣𝑣1 𝑖𝑖 + 𝑣𝑣2 𝑗𝑗 + 𝑣𝑣3 𝑘𝑘

𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣3 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣1
∇. (∇ × 𝑓𝑓) = �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . �� 𝜕𝜕𝜕𝜕3 − � 𝑖𝑖 + � 𝜕𝜕𝜕𝜕1 − � 𝑗𝑗 + � 𝜕𝜕𝜕𝜕2 − � 𝑘𝑘�
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕 𝜕𝜕𝑣𝑣3 𝜕𝜕𝑣𝑣2 𝜕𝜕 𝜕𝜕𝑣𝑣1 𝜕𝜕𝑣𝑣3 𝜕𝜕 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣1

� − �+ � − �+ � − �=0
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
43.2.4 Example
Prove that div(𝑓𝑓𝑓𝑓)= 𝑓𝑓 (div 𝑣𝑣)+grad(𝑓𝑓). 𝑣𝑣, where 𝑓𝑓 is scalar function
Solution
𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕
∇. (𝑓𝑓𝑓𝑓) = �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . (𝑓𝑓𝑓𝑓1 𝑖𝑖 + 𝑓𝑓𝑣𝑣2 𝑗𝑗 + 𝑓𝑓𝑣𝑣3 𝑘𝑘) = 𝜕𝜕𝜕𝜕 (𝑓𝑓𝑓𝑓1 ) + 𝜕𝜕𝜕𝜕 (𝑓𝑓𝑣𝑣2 ) + 𝜕𝜕𝜕𝜕 (𝑓𝑓𝑣𝑣3 )
2
Divergence and Curl
𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣3 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

= 𝑓𝑓 � 𝜕𝜕𝜕𝜕1 + + � + (𝑣𝑣1 𝑖𝑖 + 𝑣𝑣2 𝑗𝑗 + 𝑣𝑣3 𝑘𝑘). �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 �
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
= 𝑓𝑓(∇. 𝑣𝑣) + 𝑣𝑣. (∇𝑓𝑓) = 𝑓𝑓(∇. 𝑣𝑣) + ∇𝑓𝑓. 𝑣𝑣
43.2.5 Example
If 𝒓𝒓 = 𝑥𝑥𝑥𝑥 + 𝑦𝑦𝑦𝑦 + 𝑧𝑧𝑧𝑧, |𝒓𝒓| = 𝑟𝑟, show that div (𝒓𝒓/𝑟𝑟 3 )=0
Solution
𝜕𝜕 𝑥𝑥
𝒓𝒓 𝜕𝜕 𝜕𝜕 𝜕𝜕 𝑥𝑥 𝑦𝑦 𝑧𝑧 � �
∆. �𝑟𝑟 3 � = �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . �𝑖𝑖 𝑟𝑟 3 + 𝑗𝑗 𝑟𝑟 3 + 𝑘𝑘 𝑟𝑟 3 � = ∑ 𝜕𝜕𝜕𝜕 𝑟𝑟 3
3 3 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

= − �𝒙𝒙 + 𝒚𝒚 + 𝒛𝒛 �
𝑟𝑟 3 𝑟𝑟 4 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
Since 𝑟𝑟 2 = 𝑥𝑥 2 +𝑦𝑦 2 + 𝑧𝑧 2
𝒓𝒓 3 3
Therefore, ∆. �𝑟𝑟 3 � = 𝑟𝑟 3 − 𝑟𝑟 3 = 𝟎𝟎
43.2.6 Example
Prove the following identities
(i) curl (𝑓𝑓𝑓𝑓) =(grad𝑓𝑓)× 𝑣𝑣 + 𝑓𝑓curl 𝑣𝑣

𝜕𝜕 2 𝜕𝜕 2 𝜕𝜕 2
(ii) div(grad 𝑓𝑓)=∇2 𝑓𝑓 where ∇2 = 𝜕𝜕𝑥𝑥 2 + 𝜕𝜕𝑦𝑦 2 + 𝜕𝜕𝑧𝑧 2 is the Laplacian operator
(iii) curl(curl𝑣𝑣)= ∇(∇. 𝑣𝑣) − ∇2 𝑣𝑣 or grad(div 𝑣𝑣) = ∇ × (∇ × 𝑣𝑣) + ∇2 𝑣𝑣.
where 𝑓𝑓 is a scalar function.
Solution
𝜕𝜕 𝜕𝜕
(i) curl (𝑓𝑓𝑓𝑓) = ∇ × (𝑓𝑓𝑓𝑓) = ∇ × (𝑓𝑓𝑓𝑓1 𝑖𝑖 + 𝑓𝑓𝑣𝑣2 𝑗𝑗 + 𝑓𝑓𝑣𝑣3 𝑘𝑘) = ∑[𝜕𝜕𝜕𝜕 (𝑓𝑓𝑣𝑣3 ) − 𝜕𝜕𝜕𝜕 (𝑓𝑓𝑣𝑣2 )]
𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣3 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣1 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

= 𝑓𝑓 �� 𝜕𝜕𝜕𝜕3 − � 𝑖𝑖 + � 𝜕𝜕𝜕𝜕1 − � 𝑗𝑗 + � 𝜕𝜕𝜕𝜕2 − � 𝑘𝑘� + [�𝑣𝑣3 𝜕𝜕𝜕𝜕 − 𝑣𝑣2 𝜕𝜕𝜕𝜕 � 𝑖𝑖 +
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
�𝑣𝑣1 𝜕𝜕𝜕𝜕 − 𝑣𝑣3 𝜕𝜕𝜕𝜕 � 𝑗𝑗 + �𝑣𝑣2 𝜕𝜕𝜕𝜕 − 𝑣𝑣1 𝜕𝜕𝜕𝜕 � 𝑘𝑘]
𝜕𝜕 𝜕𝜕 𝜕𝜕
= 𝑓𝑓(curl 𝑣𝑣) + �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � × (𝑣𝑣1 𝑖𝑖 + 𝑣𝑣2 𝑗𝑗 + 𝑣𝑣3 𝑘𝑘)
= 𝑓𝑓(curl 𝑣𝑣) + (grad 𝑓𝑓)× 𝑣𝑣
3
Divergence and Curl
𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓 𝜕𝜕 2 𝑓𝑓

(ii) div(grad 𝑓𝑓)= �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � . �𝑖𝑖 𝜕𝜕𝜕𝜕 + 𝑗𝑗 𝜕𝜕𝜕𝜕 + 𝑘𝑘 𝜕𝜕𝜕𝜕 � = �𝜕𝜕𝑥𝑥 2 + 𝜕𝜕𝑦𝑦 2 + 𝜕𝜕𝑧𝑧 2 � = ∇2 𝑓𝑓
𝜕𝜕 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣2
(iii) grad(div 𝑣𝑣)= ∇ × (∇ × 𝑣𝑣) = �∑ 𝑖𝑖 𝜕𝜕𝜕𝜕 � × �∑ 𝑖𝑖 � 𝜕𝜕𝜕𝜕3 − ��
𝜕𝜕𝜕𝜕
𝜕𝜕 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣1 𝜕𝜕 𝜕𝜕𝑣𝑣1 𝜕𝜕𝑣𝑣3

= � 𝑖𝑖 � � − �− � − ��
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝜕𝜕 𝜕𝜕𝑣𝑣 𝜕𝜕𝑣𝑣2 𝜕𝜕𝑣𝑣3 𝜕𝜕 2 𝑣𝑣 𝜕𝜕 2 𝑣𝑣1 𝜕𝜕 2 𝑣𝑣1

= ∑ 𝑖𝑖 �𝜕𝜕𝜕𝜕 � 𝜕𝜕𝜕𝜕1 + + � − � 𝜕𝜕𝑥𝑥 21 + + ��
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝑦𝑦 2 𝜕𝜕𝑧𝑧 2
𝜕𝜕 𝜕𝜕 2 𝜕𝜕 2 𝜕𝜕 2
= �∑ 𝑖𝑖 𝜕𝜕𝜕𝜕 � (∇. 𝑣𝑣) − �𝜕𝜕𝑥𝑥 2 + 𝜕𝜕𝑦𝑦 2 + 𝜕𝜕𝑧𝑧 2 � (∑ 𝑖𝑖𝑣𝑣1 )
= ∇(∇. 𝑣𝑣) − ∇2 𝑣𝑣
Suggested Readings
New York.
House, New Delhi.
New Delhi.
4
Lesson 44
Line Integral
44.1 Introduction
Let 𝐶𝐶 be a simple curve. Let the parametric representation of 𝐶𝐶 be written as
𝑥𝑥 = 𝑥𝑥(𝑡𝑡), 𝑦𝑦 = 𝑦𝑦(𝑡𝑡), 𝑧𝑧 = 𝑧𝑧(𝑡𝑡), 𝑎𝑎 ≤ 𝑡𝑡 ≤ 𝑏𝑏 (44.1.1)
Therefore, the position vector of appoint on the curve 𝐶𝐶 can be written as
𝑟𝑟(𝑡𝑡) = 𝑥𝑥(𝑡𝑡)𝑖𝑖 + 𝑦𝑦(𝑡𝑡)𝑗𝑗 + 𝑧𝑧(𝑡𝑡)𝑘𝑘, 𝑎𝑎 ≤ 𝑡𝑡 ≤ 𝑏𝑏 (44.1.2)
44.2 Line Integral with Respect to Arc Length

Let 𝐶𝐶 be a simple smooth curve whose parametric representation is given as Eqs.(1) and (2).
Let 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) be continuous on 𝐶𝐶. Then, we define the line integral 𝑓𝑓 of over 𝐶𝐶 with respect
to the arc length 𝑠𝑠 by
𝑏𝑏
∫𝐶𝐶 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) 𝑑𝑑𝑑𝑑 = ∫𝑎𝑎 𝑓𝑓(𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡))�𝑥𝑥 ′ (𝑡𝑡)2 + 𝑦𝑦 ′ (𝑡𝑡)2 + 𝑧𝑧 ′ (𝑡𝑡)2 𝑑𝑑𝑑𝑑
since
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 2 𝑑𝑑𝑑𝑑 2 𝑑𝑑𝑑𝑑 2

𝑑𝑑𝑑𝑑 = 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 = �� 𝑑𝑑𝑑𝑑 � + � 𝑑𝑑𝑑𝑑 � + �𝑑𝑑𝑑𝑑 � 𝑑𝑑𝑑𝑑
44.2.1 Example
1
Evaluate ∫𝐶𝐶 (𝑥𝑥 2 + 𝑦𝑦𝑦𝑦)𝑑𝑑𝑑𝑑, where 𝐶𝐶 is the curve defined by 𝑥𝑥 = 4𝑦𝑦, 𝑧𝑧 = 3 form (2, 2 , 3) to
(4,1,3).
Solution
Let 𝑥𝑥 = 𝑡𝑡. Then, 𝑦𝑦 = 𝑡𝑡/4 and 𝑧𝑧 = 3. Therefore, the curve 𝐶𝐶 represented by

𝑡𝑡
𝑥𝑥 = 𝑡𝑡, 𝑦𝑦 = 4 , 𝑧𝑧 = 3, 2 ≤ 𝑡𝑡 ≤ 3.
We have 𝑑𝑑𝑑𝑑 = √17/4.
√17 4 3 139√17
Hence ∫𝐶𝐶 (𝑥𝑥 2 + 𝑦𝑦𝑦𝑦)𝑑𝑑𝑑𝑑 = ∫ �𝑡𝑡 2 + 4 𝑡𝑡� 𝑑𝑑𝑑𝑑 = .
4 2 24
44.2.2 Line Integral of Vector Fields
Let 𝐶𝐶 be a smooth curve whose parametric representation is given in Eqs. (44.1.1) and
(44.1.2). Let
Line Integral
𝑣𝑣(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑣𝑣1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑖𝑖 + 𝑣𝑣2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑗𝑗 + 𝑣𝑣3 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑘𝑘
be a vector field that is continuous on 𝐶𝐶. Then, the line integral of 𝑣𝑣 over 𝐶𝐶 is defined by
∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣1 𝑑𝑑𝑑𝑑 + 𝑣𝑣2 𝑑𝑑𝑑𝑑 + 𝑣𝑣3 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
= ∫𝐶𝐶 𝑣𝑣�𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)� . 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 (44.2.1)
If 𝑣𝑣 = 𝑣𝑣1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑖𝑖, then Eq.( 44.2.1) reduces to

𝑑𝑑𝑑𝑑
∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣1 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣1 (𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
Similarly, if 𝑣𝑣 = 𝑣𝑣2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑗𝑗 or 𝑣𝑣 = 𝑣𝑣3 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑘𝑘, we respectively obtained
𝑑𝑑𝑑𝑑
∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣2 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣2 (𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
and ∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣3 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣3 (𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)) 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑.
44.2.2 Example
Evaluate the line integral of 𝑣𝑣 = 𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑦𝑦 2 𝑗𝑗 + 𝑒𝑒 𝑧𝑧 𝑘𝑘 over the curve 𝐶𝐶 whose parametric
representation is given by 𝑥𝑥 = 𝑡𝑡 2 , 𝑦𝑦 = 2𝑡𝑡, 0 ≤ 𝑡𝑡 ≤ 1.
Solution:
The position vector of any point on 𝐶𝐶 is given by 𝑟𝑟 = 𝑡𝑡 2 𝑖𝑖 + 2𝑡𝑡𝑡𝑡 + 𝑡𝑡𝑡𝑡. We have

𝑑𝑑𝑑𝑑 1
∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 = ∫0 (2𝑡𝑡 3 𝑖𝑖 + 4𝑡𝑡 2 𝑗𝑗 + 𝑒𝑒 𝑡𝑡 𝑘𝑘). (2𝑡𝑡𝑡𝑡 + 2𝑗𝑗 + 𝑘𝑘)𝑑𝑑𝑑𝑑
1 37
= ∫0 (4𝑡𝑡 4 + 8𝑡𝑡 2 + 𝑒𝑒 𝑡𝑡 )𝑑𝑑𝑑𝑑 = 15 + 𝑒𝑒
44.2.3 Example
Evaluate the integral ∫𝑐𝑐 (𝑥𝑥 2 + 𝑦𝑦𝑦𝑦)𝑑𝑑𝑑𝑑, where 𝐶𝐶 is given by 𝑥𝑥 = 𝑡𝑡, 𝑦𝑦 = 𝑡𝑡 2 , 𝑧𝑧 = 3𝑡𝑡, 1 ≤ 𝑡𝑡 ≤ 2.
Solution:
2 163
We have ∫𝑐𝑐 (𝑥𝑥 2 + 𝑦𝑦𝑦𝑦)𝑑𝑑𝑑𝑑 = 2 ∫1 (𝑡𝑡 2 + 3𝑡𝑡 3 ) 𝑑𝑑𝑑𝑑 = 4
44.3 Line Integral of Scalar Fields

Let 𝐶𝐶 be a smooth curve whose parametric representation is as given in Eqs. (44.1.1) and
(44.1.2). Let 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧), 𝑔𝑔(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) and ℎ(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)be scalar fields which are continuous at
point over 𝐶𝐶. Then, we define a line integral as
∫𝐶𝐶 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑 + 𝑔𝑔(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑 + ℎ(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑
2
Line Integral

= ∫𝐶𝐶 �𝑓𝑓�𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)� 𝑑𝑑𝑑𝑑 + 𝑔𝑔�𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)� 𝑑𝑑𝑑𝑑 + ℎ�𝑥𝑥(𝑡𝑡), 𝑦𝑦(𝑡𝑡), 𝑧𝑧(𝑡𝑡)� 𝑑𝑑𝑑𝑑 � 𝑑𝑑𝑑𝑑
If 𝐶𝐶 is closed curve, then we usually write
∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∮𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑
44.3.1 Example
Evaluate ∫𝐶𝐶 (𝑥𝑥 + 𝑦𝑦)𝑑𝑑𝑑𝑑 − 𝑥𝑥 2 𝑑𝑑𝑑𝑑 + (𝑦𝑦 + 𝑧𝑧)𝑑𝑑𝑑𝑑 , where 𝐶𝐶 is 𝑥𝑥 2 = 4𝑦𝑦, 𝑧𝑧 = 𝑥𝑥, 0 ≤ 𝑡𝑡 ≤ 2.
Solution
𝑡𝑡 2
First we consider parametric form of 𝐶𝐶 as 𝑥𝑥 = 𝑡𝑡, 𝑦𝑦 = , 𝑧𝑧 = 2, 0 ≤ 𝑡𝑡 ≤ 2.
4
Therefore,
2 𝑡𝑡 2 𝑡𝑡 𝑡𝑡 2 10
∫𝐶𝐶 (𝑥𝑥 + 𝑦𝑦)𝑑𝑑𝑑𝑑 − 𝑥𝑥 2 𝑑𝑑𝑑𝑑 + (𝑦𝑦 + 𝑧𝑧)𝑑𝑑𝑑𝑑 = ∫0 ��𝑡𝑡 + 4
� − 𝑡𝑡 2 �2� + � 4 + 𝑡𝑡�� 𝑑𝑑𝑑𝑑 = 3
44.4 Application of Line Integrals

In this section, we consider some physical applications of the concept of line integral.
44.4.1 Work Done By A Force
Let 𝑣𝑣(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) = 𝑣𝑣1 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑖𝑖 + 𝑣𝑣2 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑗𝑗 + 𝑣𝑣3 (𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑘𝑘 be a vector function defined and
continuous at every point on 𝐶𝐶. Then the line integral of tangential component of 𝑣𝑣 along the
curve 𝐶𝐶 from a point 𝑃𝑃 to the point 𝑄𝑄 is given by
𝑄𝑄
∫𝑃𝑃 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∫𝑐𝑐 𝑣𝑣1 𝑑𝑑𝑑𝑑 + 𝑣𝑣2 𝑑𝑑𝑑𝑑 + 𝑣𝑣3 𝑑𝑑𝑑𝑑
Let now 𝑣𝑣 = 𝐹𝐹, a variable force acting on a particle which moves along a curve 𝐶𝐶. Then, the
work 𝑊𝑊 done by the force 𝐹𝐹 in displacing the particle from the point 𝑃𝑃 to the point 𝑃𝑃 along
the curve 𝐶𝐶 is given by
𝑄𝑄
𝑊𝑊 = ∫𝑃𝑃 𝐹𝐹. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 ∗ 𝐹𝐹. 𝑑𝑑𝑑𝑑
where 𝐶𝐶 ∗ is the part of𝐶𝐶 , whose initial and terminal point are 𝑃𝑃 and 𝑄𝑄.
Suppose that 𝐹𝐹is a conservative vector field . Then 𝐹𝐹 can be written as 𝐹𝐹 = grad(𝑓𝑓), where 𝑓𝑓
is a scalar potential(field). Then, the work done
𝑊𝑊 = ∫𝐶𝐶 ∗ 𝐹𝐹. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 ∗ 𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔(𝑓𝑓). 𝑑𝑑𝑑𝑑
3
Line Integral
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑄𝑄 𝑄𝑄

= ∫𝐶𝐶 ∗ ( 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 + 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 + 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑) = ∫𝑃𝑃 𝑑𝑑𝑑𝑑 = [𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)]𝑃𝑃
44.4.1 Example
Find the work done by the force 𝐹𝐹 = −𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑦𝑦 2 𝑗𝑗 + 𝑧𝑧𝑧𝑧 in moving a particle over the circular
path𝑥𝑥 2 + 𝑦𝑦 2 = 4, 𝑧𝑧 = 0 form (2,0,0) to (0,2,0).
Solution
The parametric representation of the given curve is 𝑥𝑥 = 2 cot 𝑡𝑡, 𝑦𝑦 = 2 sin 𝑡𝑡, 𝑧𝑧 = 0, 0 ≤ 𝑡𝑡 ≤
𝜋𝜋2 . Therefore, work done 𝑊𝑊 is given by
𝑊𝑊 = ∫𝐶𝐶 𝐹𝐹. 𝑑𝑑𝑑𝑑 = ∫𝐶𝐶 −𝑥𝑥𝑥𝑥𝑥𝑥𝑥𝑥 + 𝑦𝑦 2 𝑑𝑑𝑑𝑑 + 𝑧𝑧𝑧𝑧𝑧𝑧
𝜋𝜋/2
16
� [−4 sin 𝑡𝑡 cos 𝑡𝑡 (−2 sin 𝑡𝑡 ) + 4 𝑠𝑠𝑠𝑠𝑠𝑠2 𝑡𝑡(2𝑐𝑐𝑐𝑐𝑐𝑐) ] 𝑑𝑑𝑑𝑑 =
0 13
44.4.2 Circulation
A line integral of a vector field 𝑣𝑣 around a simple closed curve 𝐶𝐶 is defined as the
circulation of 𝑣𝑣 around 𝐶𝐶.
𝑑𝑑𝑑𝑑
Circulation = ∮𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 = ∮𝐶𝐶 𝑣𝑣. 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 = ∮𝑐𝑐 𝑣𝑣. 𝑇𝑇𝑇𝑇𝑇𝑇,
where 𝑇𝑇 is the tangent vector to 𝐶𝐶. For example, in fluid mechanics, let 𝑣𝑣 represents the
velocity field of a fluid and 𝐶𝐶 be a closed curve in its domain. Then, circulation gives the
amount by which the fluid tends to turn the curve rotating or circulating around 𝐶𝐶. If
∮𝑐𝑐 𝑣𝑣. 𝑇𝑇𝑇𝑇𝑇𝑇 > 0 then the fluid tends to rotate 𝐶𝐶in the anti-clockwise direction, while if
∮𝑐𝑐 𝑣𝑣. 𝑇𝑇𝑇𝑇𝑇𝑇 < 0 , then the fluid tends to rotate 𝐶𝐶 in the clockwise direction perpendicular to𝑇𝑇
at every point on 𝐶𝐶, then ∮𝑐𝑐 𝑣𝑣. 𝑇𝑇𝑇𝑇𝑇𝑇 = 0, that is the curve does not move at all.
44.5 Line Integral Independent of the Path

Let 𝜙𝜙(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) be a differentiable scalar function. The differential of 𝜙𝜙(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) is defined as
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
𝑑𝑑𝑑𝑑 = 𝜕𝜕𝜕𝜕
𝑑𝑑𝑑𝑑 + 𝜕𝜕𝜕𝜕 𝑑𝑑𝑑𝑑 + 𝜕𝜕𝜕𝜕
𝑑𝑑𝑑𝑑 =grad 𝜙𝜙. 𝑑𝑑𝑑𝑑
Therefore, a differential expression expre 𝑑𝑑𝑑𝑑 = 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑥𝑥 + 𝑔𝑔(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑 + ℎ(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑 is
an exact differential, if there exists a scalar function 𝜙𝜙(𝑥𝑥, 𝑦𝑦, 𝑧𝑧) such that
𝑑𝑑𝑑𝑑 = 𝑓𝑓(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑 + 𝑔𝑔(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑 + ℎ(𝑥𝑥, 𝑦𝑦, 𝑧𝑧)𝑑𝑑𝑑𝑑.
We now present the result on the independence of the path of a line integral
4
Line Integral
44.5.1 Theorem
Let 𝐶𝐶 be a curve in simply connected domain 𝐷𝐷 in space. Let 𝑓𝑓, 𝑔𝑔 and ℎ be continuous
function having continuous first partial derivatives in 𝐷𝐷. Then ∫𝐶𝐶 𝑓𝑓𝑓𝑓𝑓𝑓 + 𝑔𝑔𝑔𝑔𝑔𝑔 + ℎ𝑑𝑑𝑑𝑑 is
independent of path 𝐶𝐶 if and only if the integrand is exact differential in 𝐷𝐷.
44.5.2 Example
𝑥𝑥𝑥𝑥𝑥𝑥 +𝑦𝑦𝑦𝑦𝑦𝑦
Show that ∫𝐶𝐶 is independent of path of integration which does not pass through the
�𝑥𝑥 2 +𝑦𝑦 2
origin. Find the value of the integral from the point 𝑃𝑃(−1,2) to the point 𝑄𝑄(2,3).
Solution
𝑥𝑥 𝑦𝑦
We have 𝑓𝑓(𝑥𝑥, 𝑦𝑦) = and 𝑔𝑔(𝑥𝑥, 𝑦𝑦) =
�𝑥𝑥 2 +𝑦𝑦 2 �𝑥𝑥 2 +𝑦𝑦 2
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
Now 𝜕𝜕𝜕𝜕 = −𝑥𝑥𝑥𝑥/(𝑥𝑥 2 + 𝑦𝑦 2 )3/2 and 𝜕𝜕𝜕𝜕
= −𝑥𝑥𝑥𝑥/(𝑥𝑥 2 + 𝑦𝑦 2 )3/2
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
Since 𝜕𝜕𝜕𝜕
= 𝜕𝜕𝜕𝜕 , the integral is independent of any path of integration which does not pass
through the origin. Also, the integrand is an exact differential. Therefore, there exists a
function 𝜙𝜙(𝑥𝑥, 𝑦𝑦) such that
𝜕𝜕𝜕𝜕 𝑥𝑥 𝜕𝜕𝜕𝜕 𝑦𝑦
= 𝑓𝑓(𝑥𝑥, 𝑦𝑦) = and = 𝑔𝑔(𝑥𝑥, 𝑦𝑦) =
𝜕𝜕𝜕𝜕 �𝑥𝑥 2 +𝑦𝑦 2 𝜕𝜕𝜕𝜕 �𝑥𝑥 2 +𝑦𝑦 2
Integrating the first equation with respect to 𝑥𝑥, we get 𝜙𝜙(𝑥𝑥, 𝑦𝑦) = �𝑥𝑥 2 + 𝑦𝑦 2 + ℎ(𝑦𝑦).
𝜕𝜕𝜕𝜕 𝑦𝑦 𝑦𝑦 𝑑𝑑ℎ 𝑑𝑑ℎ
Substituting in 𝜕𝜕𝜕𝜕 = = + 𝑑𝑑𝑑𝑑 or 𝑑𝑑𝑑𝑑 = 0 or ℎ(𝑦𝑦) = 𝑘𝑘, constant.
�𝑥𝑥 2 +𝑦𝑦 2 �𝑥𝑥 2 +𝑦𝑦 2
Hence 𝜙𝜙(𝑥𝑥, 𝑦𝑦) = �𝑥𝑥 2 + 𝑦𝑦 2 + 𝑘𝑘
𝑥𝑥𝑥𝑥𝑥𝑥 +𝑦𝑦𝑦𝑦𝑦𝑦 (2,3) (2,3)

Therefore, ∫𝐶𝐶 = ∫(−1,2) 𝑑𝑑(�𝑥𝑥 2 + 𝑦𝑦 2 ) = [�𝑥𝑥 2 + 𝑦𝑦 2 ](−1,2) = √13 − √5
�𝑥𝑥 2 +𝑦𝑦 2
Suggested Readings
Courant, R. and John, F. (1989), Introduction to Calculus and Analysis, Vol. II, Springer-
Verlag, New York.
Jain, R.K. and Iyengar, S.R.K. (2002) Advanced Engineering Mathematics, Narosa
Publishing House, New Delhi.
Jordan, D.W. and Smith, P. (2002) Mathematical Techniques, Oxford University Press,
Oxford.
5
Line Integral
Wylie, C. R. and Barrett, L.C. (2003) Advanced Engineering Mathematics, Tata McGraw-
Hill, New Delhi.
6
Lesson 45
Green’s Theorem in the Plane
45.1 Introduction
The theorem provides a relationship between a double integral over a region and the line
integral over the closed curve C bounding R. Green’s theorem is also called the first
fundamental theorem of integral vector calculus.
45.2 The Main Result

45.2.1 Theorem: (Green’s theorem)
Let C be a piecewise smooth simple closed curve bounding a region R. If f, g, ∂f / ∂y, ∂g / ∂x

are continuous on R, then
 ∂g ∂f 
∫ f ( x, y)dx + g ( x, y)dy =
C
∫∫  ∂x − ∂y dxdy
R
The integration being carried in the positive direction (counter clockwise direction) of C.
Proof: We shall prove Green’s theorem for a particular case of the region R.
Let the region R be simultaneously expressed in the following forms.
R : u1 ( x ) ≤ y ≤ u2 ( x ) , a ≤ x ≤ b
R : v1 ( x ) ≤ x ≤ v2 ( x ) , c ≤ y ≤ d
We obtain
∂g
d  v2 ( y ) ∂g  d
∫∫R ∂x dxdy
= ∫c  v=
∫( x ) ∂x dx dy ∫ [ g (v ( y), y) − g (v ( y), y)]dy
2 1
1  c
d c
= ∫ g (v2 ( y), y)dy + ∫ g (v1 ( y), y)dy =
c d
∫ g ( x, y)dy
C
the integration being carried in the counter clockwise direction.
We obtain
∂f
b  u2 ( y ) ∂f  b
∫∫R ∂x dxdy
= ∫a  u=
∫( x ) ∂y dy dy ∫ [ f ( x, u ( x)) − f ( x, u ( x))]dx
2 1
1  a
b a
∫ f ( x, u2 ( x))dx + ∫ g ( x, u1 ( x))dx =
=
a b
− ∫ f ( x, y )dx
C
the integration being carried in the counter clockwise direction. Therefore
 ∂g ∂f 
∫ f ( x, y)dx + g ( x, y)dy =
C
∫∫  ∂x − ∂y dxdy .
R
45.2.2 Example : Evaluate
∫ (x + y 2 )dx + ( y + 2 x)dy, where C is the boundary of the region in the first quadrant that is
2
bounded by the curves y 2 = x and x 2 = y .
Solution: The curves intersect at (0,0) and (1,1). The bounding curve is C. We have
f ( x, y=
) x 2 + y 2 and g ( x, y )= y + 2 x .
Using the Green’s theorem, we obtain
∫ (x + y 2 )dx + ( y + 2 x)dy= ∫∫ (2 − 2 y)dxdy

2
C R
1 x 1
x
∫ ∫ (2 − 2 y)dydx =
= ∫ (2 y − y ) |
2
2
dx
0 x2 0 x
1
= ∫ (2
0
x − x − 2 x 2 + x 4=
)dx 11/ 30
45.2.3 Example: Find the work done by the force F = ( x 2 − y 3 )i + ( x + y ) j in moving a

particle along the closed path C containing the curves x + y= 0, x 2 + y =
2
16 and y = x in the
first and fourth quadrants.
Solution: The work done by the force is given by
W= W= ∫ F .dr= ∫ ( x − y 3 )dx + ( x + y )dy.

2
C C
The closed path C bounds the region R. Using the Green’s theorem, we obtain
∫ (x − y 3 )dx + ( x + y )dy= ∫∫ (1 + 3 y
2 2
)dxdy.
C R
It is convenient to use polar coordinates to evaluate the integral. The region R is given by
= θ , y r sin θ , 0 ≤ r ≤ 4, −π / 4 ≤ θ ≤ π / 4.
R : x r cos=
Therefore,
2
π /4 4 π /4
 r2 3 4 2  4
∫∫ (1 + 3 y )dxdy =∫ ∫ + θ θ =∫  2 + 4 r sin θ  0 dθ
2 2 2
(1 3r sin ) rdrd
R −π /4 0 −π /4  
π /4 π /4
= ∫ (8 + 192sin θ )dθ = ∫ [8 + 96(1 − cos 2θ )]dθ
2
−π /4 −π /4
π /4
2[104θ − 48sin 2θ ]
= 52π − 96.
=
0
−x
=
45.2.4 Example: Verify the Green’s theorem for f ( x, y ) e= sin y, g ( x, y ) e − x cos y and C
is the square with vertices at (0,0), (π/2,0), (π/2,π/2), (0,π/2).
Solution: We can write the line integral as
 
∫
C
fdx + gdy =  ∫ + ∫ + ∫ + ∫
 C1 C2 C3 C4
 ( fdx + gdy )

where C1 , C2 , C3 and C4 are the boundary lines. We have along C1 : y= 0, 0 ≤ x ≤ π / 2 and
∫e
−x
(sin ydx + cos ydy ) =
0,
C1
: x π / 2, 0 ≤ y ≤ π / 2 and
along C2=
π /2
∫e ∫e
−π /2
−x
(sin ydx + cos =
ydy ) ydy e −π /2 ,
cos=
C2 0
C3 : y π / 2, π / 2 ≤ x ≤ 0 and
along=
∫e ∫ e dx =
e −π /2 − 1,
−x −x
(sin ydx + cos ydy ) =
C3 π /2
: x 0, π / 2 ≤ y ≤ 0 and
along C4=
∫e ∫ cos ydy =
−x
(sin ydx + cos ydy ) = −1.
C4 π /2
Therefore,
π /2 π /2
∫ fdx + gdy =∫∫ (−2e cos y )dxdy =∫ ∫ (−2e cos ydxdy ) =2(e −π /2 − 1).
−x −x
C R 0 0
45.2.5 Example: Now we use the Green’s theorem to show that
3
∂u
∫ ∂n ds= ∫∫ ∇ udxdy,
2
C R
where ∇ 2 is the Laplace operator ∂ 2 / ∂x 2 + ∂ 2 / ∂y 2 and n is the unit outward normal to C.
Solution:
Let the position vector of a point on C, in terms of the arc length r=

( s ) x( s )i + y ( s ) j.
Then, the tangent vector to C is given by
dr dx dy
=
T = i+ j
ds ds ds
and the normal vector n is given by (since n.T = 0 )
dy dx
=
n i− j.
ds ds
Note that n is the unit normal vector. Now
∂u
∫ ∂n ds= ∫ ∇u.nds
C C
since ∂u / ∂n is the directional derivative of u in the direction of n. Therefore, using Green’s

theorem, we obtain
∂u  ∂u ∂y ∂u ∂x   ∂u ∂u 
∫ ∂n ds =
C
∫  ∂x ∂s − ∂y ∂s  ds =
C
∫  − ∂y dx + ∂x dy 
C
∂ u ∂ u
2 2
∫∫R  ∂x 2 + ∂y 2 dxdy =
= ∫∫R ∇ udxdy.
2
Suggested Readings
Verlag, New York.
Oxford.
4
Hill, New Delhi.
5
Lesson 46
Surface Integral
46.1 Introduction
b
The double and triple integrals are the generalizations of the definite integral ∫ f ( x)dx to two
a
and three dimensions respectively. The surface area integral is a generalization of the arc
length integral
b
∫
a
1 + ( y ') 2 dx.
We shall now present a generalization of the line integral ∫ f ( x, y)ds to three dimensions.
C
This generalization is called the surface integral.
Let g ( x, y, z ) be a given function defined in the three dimensional space and let S be surface
which is the graph of a function z = f ( x, y ), or y = h1 ( x, z ), or x = h( y, z ). We assume that
(i) g ( x, y, z ) is continuous at all points on S, (ii) S is smooth and bounded and (iii) the
projection R of the surface S on x-y plane, x-z plane, or y-z plane respectively expressed in
the forms as assumed in the proof of the Green’s theorem. For example, the projection R on
the x-y plane can be expressed in the forms
R : u1 ( x ) ≤ y ≤ u2 ( x ) , a ≤ x ≤ b
or R : v1 ( x ) ≤ x ≤ v2 ( x ) , c ≤ y ≤ d .
The surface integral can be defined in a similar way as the double integral is defined.
Subdivide S into n parts S1 , S 2 ,..., S n of areas ∆A1 , ∆A2 ,..., ∆An . The projection R of S is
therefore partitioned into n rectangles R1 , R2 ,..., Rn . We choose an arbitrary point
Pk ( xk , yk , zk ) on each element of the surface area S k and form the sum
n
=In ∑ g ( x , y , z )∆A .
k =1
k k k k
Let n → ∞ , such that the largest element of the surface area shrinks to a point. This implies
that as n → ∞ ,the length of the longest diagonal of the projected rectangles tends to zero. In
the limit as n → ∞ , the sequence {I n } has a limiting value which is independent of the way S
is subdivided and the choice of Pk on S k . This limiting value is called the surface integral of
g ( x, y, z ) over S.
That is, we define the surface integral as
Surface Integral
∫∫ g ( x, y, z )dA lim ∑ g ( xk , yk , zk )∆Ak .

=
S
|d | → 0
k =1
where | d | is the length of the longest diagonal of the projected rectangles.
The surface integral can be evaluated in any of the following ways.
(i) Let S be represented in parametric form as r = r (u , v). Then we can write
∫∫ g ( x, y, z )dA
= ∫∫ g[ x(u, v), y(u, v), z(u, v)] | r × r | dudv u v
S R*
∫∫ g[ x(u, v), y(u, v), z (u, v)][r rv 2 − (ru .rv ) 2 ]1/2 dudv
2
u
R*
where R* is the region corresponding to S in the u-v plane.
(ii) Let S be represented in the form z = f ( x, y ). Then we can write
∫∫=
g ( x, y, z )dA ∫∫ g[ x, h ( x, z ), z ][1 + f + f y 2 ]1/2 dxdy
2
1 x
S R
where R is the orthogonal projection of S on the x-y plane.
(iii) Let S be represented in the form x = h1 ( y, z ). Then we can write
∫∫ g=
( x, y, z )dA ∫∫ g[h( y, z ), y, z ][1 + (h ) + (h1 ) 2 y ]1/2 dxdz
2
1 x
S R
where R is the orthogonal projection of S on the x-z plane.
(iv) Let S be represented in the form x = h( y, z ). Then we can write
∫∫=
g ( x, y, z )dA ∫∫ g[h( y, z ), y, z ][1 + h + hz 2 ]1/2 dydz
2
y
S R
where R is the orthogonal projection of S on the y-z plane.
If S is piecewise smooth and consists of the surfaces S1 , S 2 ,..., S k , then
∫∫ g=
S
( x, y, z )dA ∫∫ g ( x, y, z )dA + ∫∫ g ( x, y, z ) dA +... + ∫∫ g ( x, y, z ) dA .
S1 S2 Sk
We now present some of the important applications of the surface integrals.
46.2 Mass of a Surface

Let ρ ( x, y, z ) denote the density of a surface S at any point or mass per unit surface area.
Then, the mass m of the surface is given by
2
Surface Integral
m = ∫∫ ρ ( x, y, z )dA.
S
46.3 Moment of Inertia

Let ρ ( x, y, z ) denote the density of a surface S at any point. Then, the moment of inertia I of
the mass m with respect to a given axis l is defined by the surface integral
I = ∫∫ ρ ( x, y, z )d 2 dA.
S
where d is the distance of the point (x,y,z) from the reference axis l. If the surface is
homogeneous, then ρ ( x, y, z ) = constant and ρ ( x, y, z ) =m/A, where A is the surface area of
S. Then,
m
I= ∫∫
A S
d 2 dA
46.3.1 Example: Find the mass of the surface of the cone z = 2 + x 2 + y 2 , 2 ≤ z ≤ 7, in the
first octant, if the density ρ ( x, y, z ) at any point of the surface is proportional to its distance
from the x-y plane.
Solution: The density is given by ρ ( x, y, z ) = cz , c is constant. We have
x y
z=f ( x, y ) =+
2 x2 + y 2 , f x = , fy =
x2 + y 2 x2 + y 2
x2 y2
dA = 1 + f x + f y dxdy = 1 + 2 + dxdy = 2dxdy.
2
2
x + y 2 x2 + y 2
The projection of S on the x-y plane is given by R : x 2 + y 2 =

25, in the first quadrant.
Therefore, mass of the surface is given by
m= ∫∫ czdA = ∫∫ c[2 +
S R
x 2 + y 2 ] 2dxdy
= c 2 ∫∫ [2 + x 2 + y 2 ]dxdy
R
=
Substituting θ , y r sin θ , 0 ≤ θ ≤ π / 2, we obtain
x r cos=
5 π /2 π /2 5
 2 r3 
m= c 2 ∫ ∫ (2 + r )rdrdθ= c 2 ∫  r +  dθ
0 0 0  3 0
 125  π 100 2
= c 2  25 +  = π c.
 3 2 3
3
Surface Integral
46.3.2 Example: Evaluate the integral ∫∫ ydA where

S
S is the portion of the cylinder
x= 6 − y in the first octant bounded by the planes=

2
x 0,=
y 0,=
z 0 and z = 8.
Solution: The equation of the surface is in the form x = h( y, z ). Here h( y, z )= 6 − y 2 and

g ( x, y, z ) = y. We have
hy =
−2 y, hz =
0, (1 + hy 2 + hz 2 )1/2 =
(1 + 4 y 2 )1/2 .
The projection of S on the y-z plane is the rectangle OABC with sides=
y 0,=
y =
6, z 0
and z = 8. Therefore,
6 8
∫∫ ydA =
∫∫ y(1 + 4 y ) dydz =
∫ ∫ y(1 + 4 y ) dydz
2 1/2 2 1/2
S R 0 0
6
 (1 + 4 y 2 )3/2  2 248
= 8 =  =
[(25)3/2 − 1] .
 8(3 / 2)  0 3 3
46.3.3 Example: Evaluate the surface integral ∫∫ F .ndA

S
where F = 6 zi + 6 j + 3 yk and S is
the portion of the plane 2 x + 3 y + 4 z =

12, which is in the first octant.
Solution: Let f ( x, y, z ) = 2 x + 3 y + 4 z − 12 = 0 be the surface. Then
gradf 1
grad f = 2i + 3 j + 4k , n = = (2i + 3 j + 4k ).
| gradf | 29
Consider the projection of S on the x-y plane. The projection of the portion of the plane ABC
in the first octant is the rectangle bounded by=
x 0,=y 0 and 2 x + 3 y = 12. We have
dxdy dxdy
=
dA = .
n.k 4 / 29
1
Therefore, ∫∫=
S
F .ndA ∫∫
S 29
(12 z + 18 + 12 y )dA.
From the equation of the surface, we get 4 z = 12 − 2 x − 3 y. Hence,
4
Surface Integral
1 1
∫∫ F .ndA
S
= ∫∫
S 29
(54 − 6 x + 3 y )=
dA
4 ∫∫
R
(54 − 6 x + 3 y )dxdy
1  
6 (12 − 2 x )/3 6
1
= ∫  ∫ (54 − 6 x + 3 y )dy = dx ∫ (360 − 102 x + 7 x 2 )dx
 y 0
4 x 0=
=  60
1 7 36
= 360 x − 51x 2
+ x = 138.
6  3  0
Suggested Readings
Verlag, New York.
Oxford.
Hill, New Delhi.
5
Lesson 47
Stokes’s Theorem
47.1 Introduction
Let C be a curve in two dimensions which is written in the parametric form r = r ( s ) . Then,
the unit tangent vector to C is given by
dx dy
=
T i+ j
ds ds
Let v be written in the form =

v gi − fj.
 dx dy  dx dy
Then v.T =
( gi − fj ).  i + j =g −f .
 ds ds  ds ds
By Green’s theorem , we have

 ∂f ∂g 
∫ v.dr= ∫ v.Tds= ∫ gdx − fdy= ∫∫ −  ∂x + ∂y dxdy= ∫∫ (∇ × v).kdxdy. \
C C C R R
This result can be considered as a particular case of the Stokes’s theorem. Extension of the
Green’s theorem to three dimensions can be done under the following generalizations.
(i) The closed curve C enclosing R in the plane → the closed curve C bounding an
open smooth orientable surface S (open two sided surface).
(ii) The unit normal n to C → the unit outward or inward normal n to S.
(iii) Counter clockwise direction of C → the direction of C is governed by the direction
of the normal n to S. If n is taken as outward normal, then C is oriented as right
handed screw and if n is taken as inward normal, then C is oriented as left handed
screw.

We now state the Stokes’s theorem.
47.2.1 Theorem (Stokes’s Theorem): Let S be a piecewise smooth orient able surface
bounded by a piecewise smooth simple closed curve C. Let
v( x, y, z ) =v1 ( x, y, z )i + v2 ( x, y, z ) j + v3 ( x, y, z ) k be a vector function which is continuous and
has continuous first order partial derivatives in a domain which contains S. If C is traversed
in the positive direction, then
∫ v.dr= ∫ (v.T )ds= ∫∫ (∇ × v).ndA

C C S
where n is the unit normal to S in the direction o orientation of C.
Stokes’s Theorem
In terms of components of v we have
∫ [v ( x, y, z )dx + v ( x, y, z )dy + v ( x, y, z)dz=] ∫∫ (∇ × v).ndA.

C
1 2 3
S
47.2.2 Remark: As in divergence theorem, the theorem holds if the given surface S can be
subdivided into finitely many special surfaces such that each of these surfaces can be
described in the required manner.
47.2.3 Remark: To prove the Stokes’s theorem, it is not necessary that the equation of the
surface should be simultaneously written in the =
forms z f=
( x, y ), y g ( x, z ) and x = h( y, z )
. For example, if we take the question of the surface as z = f ( x, y ) and assume that f ( x, y )
has continuous second order partial derivatives then the theorem can be easily proved.
47.2.4 Remark: (Physical interpretation of curl)
We know that in rigid body rotation, if v denotes the tangential (linear) velocity of a point on
it, then curl v represents the angular velocity of the uniformly rotating body. We also know
that a line integral of a vector field v around a simple closed curve C defines the circulation
of v around C. For example, if v denotes the velocity of a fluid, then circulation gives the
amount by which the fluid tends to turn the curve by rotating or circulating around C.
Therefore, circulation (line integral) is closely related to curl of the vector field. To see this,
let Cr be a small circle with centre at P* ( x* , y* , z * ) . Then, by Stokes’s theorem, we have
∫ v.dr = ∫∫ curlv.ndA
Cr Sr
where S r is a small surface whose bounding curve is Cr . Let P ( x, y, z ) be any arbitrary point
on Cr . We approximate curlv( P ) ≈ curlv( P* ). Then, we have
∫ v.dr ∫∫= [curlv( P* ).n( P* )]∫∫ dA

* *
[curlv( P )].n( P )dA
Cr Sr Sr
= [curlv( P* ).n( P )] Ar
*
where Ar is the surface area of S r . Let the radius r of Cr tend to zero. Then, the
approximation curlv( P ) ≈ curlv( P* ) becomes more accurate and in the limit as r → 0, we get
1
curlv( P* ).n( P ) = lim ∫ v.dr.
*
r →0 Ar Cr
The left hand side of the above equation is the normal component of curl v. The right hand
side of equation is circulation of v per unit area. The left hand side is maximum when the
circle Cr is positioned such that the normal to surface, n( P* ) points in the same direction as
curlv( P* ).
2
Stokes’s Theorem
47.2.5 Remark: Stokes’s theorem states that the value of the surface integral is same for any
surface as long as the boundary curve, bounding the projection R on any coordinate plane, is
the same curve C. Hence, in the degenerate case, when S coincides with R, we can take n=k
or j or i depending on whether the projection is taken on the x-y plane or x-z plane or y-z
plane.
47.2.6 Example: Verify Stokes’s theorem for the vector field v = (3 x − y )i − 2 yz 2 j − 2 y 2 zk ,
where S is the surface of the sphere x 2 + y 2 + z 2= 16, z > 0.
Solution: Consider projection of S on the x-y plane. The projection is the circular region
x 2 + y 2 ≤ 16, z =
0 and the bounding curve C is the circle z = 0, x 2 + y 2 = 16.
We have
∫ v.dr = ∫ (3x − y)dx − 2 yz dy − 2 y zdz = ∫ (3x − y)dx

2 2
C C C
=
since z=0. Setting cos θ , y 4sin θ , we obtain
x 4=
2π 2π
3 1 
C∫ (3x − y)dx =
∫0 4(3cos θ − sin θ )(−4sin θ )dθ =
−16 ∫  sin 2θ − (1 − cos 2θ )  dθ
0 
2 2 
1
= 16
=   2π 16π .
2
i j k
∇ × v = ∂ / ∂x ∂ / ∂y ∂ / ∂z = i (−4 yz + 4 yz ) − j (0) + k (1) = k
3 x − y −2 yz 2 −2 y 2 z
Now,
2( xi + yj + zk ) 1 z
=n = ( xi + yj + zk ), (∇ × v=
).n .
2 x2 + y 2 + z 2 4 4
Therefore,
z z dxdy z dxdy
∫∫ (∇ × v=
S
).ndA ∫∫=
4
dA ∫∫ = ∫∫ = ∫∫=
4 n.k
S
4 ( z / 4)
R
dxdy
R R
16π
which is the area of the circular region in the x-y plane. Hence, Stokes’s theorem is proved.
3
Stokes’s Theorem
∫ (2 y dx + x dy + zdz
3 3
47.2.7 Example: Evaluate where C is the trace of the cone
C
=z (x + y )
2 2
intersected by the plane x=4 and S is the surface of the cone below z=4.
Solution: We have v = 2 y 3i + x3 j + zk and
i j k
curlv =∂ / ∂x ∂ / ∂y ∂ / ∂z =i (0) − j (0) + k (3 x 2 − 6 y 2 ).
2 y3 x3 z
If the outward normal to S is taken, then it points downwards. Then, the orientation of C is
taken in the clockwise direction. Alternatively, if the inward normal to S is taken, then C is
oriented in the counter clockwise direction.
Let f ( x, y, z=
) x 2 + y 2 − z= 0 be taken as the equation of the surface. Then, the normal and
unit normal are given by
xi + yj xi + yj − zk ( xi + yj − zk ) / z xi + yj − zk
=N = −k = and n = except at the
x2 + y 2 z ( x2 + y 2 + z 2 ) / z 2 2z
origin.
(3 x 2 − 6 y 2 ) (3 x 2 − 6 y 2 ) dxdy
We have ∫∫ (∇ × v).ndA =
S
∫∫ −
S 2
dA = − ∫∫ −
S 2 (−1/ 2)
=
since dxdy = (n.k )dA. Therefore, substituting cos θ , y r sin θ , we obtain
x r=
4 0
∫∫ (∇ × v= ∫∫ (3x − 6 y )= ∫ ∫π (3cos θ − 6sin 2 θ )r 3 drdθ

2 2 2
).ndA dxdy
S R r =0 2
4 0 4 0
3 3
= ∫ ∫
2 0 2π
[(1 + cos 2θ ) − 2(1 − cos 2θ )]r 3 drd
=θ ∫ ∫
2 0 2π
(3cos 2θ − 1)r 3 drdθ
3  r 4  4  3sin 2θ  0
=    −θ  192π .
=
2  4 0 2  2π
=
The bounding curve C is given by x 2 + y 2 = 16, z = 4. Now setting cos θ , y 4sin θ ,
x 4=
4
Stokes’s Theorem
∫ 2 y dx + x dy + zdz
= ∫ 2 y dx + x dy
3 3 3 3
C C
0
= ∫π 64[2sin θ (−4sin θ ) + cos3 θ (4 cos θ )]
3
2
We obtain 2π π /2
−256 ∫ [cos θ − 2sin θ ]dθ =
= 4
−1024 ∫ (cos 4 θ − 2sin 4 θ
4
0 0
3 1 π  3 1 π 
=
−1024  . . − 2  . .   =192π .
4 4 2  4 2 2 
Hence, the theorem is verified.
Suggested Readings
Verlag, New York.
Oxford.
Hill, New Delhi.
5
Lesson 48
Divergence Theorem of Gauss
48.1 Introduction
Let C be a curve in two dimensions which is written in the parametric form r = r ( s ) . Then, the
unit tangent and unit normal vectors to C are given by
dx dy dy dx
T = i+ j, n = i − j.
ds ds ds ds
Then,
 dx dy   dy dx 
fdx + gdy = f + g  ds =( gi − fj ).  i − j  ds =(v.n)ds
 ds ds   ds ds 
where =
v gi − fj. Also
∂g ∂f  ∂ ∂ 
− = i + j  .( gi − fj ) =
∇.v
∂x ∂y  ∂x ∂y 
Hence, Green’s theorem can be written in a vector form as
∫ (v.n)ds= ∫∫ (∇.v)dxdy
C R
The result is a particular case of the Gauss’s divergence theorem. Extension of the Greens’
theorem to three dimensions can be done under the following generalisations.
(i) A region R in the plane → a three dimensional solid D

(ii) The closed curve C enclosing R in the plane → the closed surface S enclosing the
solid D
(iii) The unit outer normal n to C → the unit outer normal n to S.
(iv) A vector field v in the plane → a vector field v in the three dimensional space
(v) The line integral ∫ (v.n)ds
C
→ a surface integral ∫∫ (v.n)dA
S
(vi) The double integral ∫∫ ∇.vdxdy

R
→ a triple (volume) integral ∫∫∫ ∇.vdV .
D

The above generalizations give the following divergence theorem.
Theorem: (Divergence theorem of Gauss) Let D be a closed and bounded region in the three
dimensional space whose boundary is a piecewise smooth surface S that is oriented outward. Let
v( x, y, z ) =v1 ( x, y, z )i + v2 ( x, y, z ) j + v3 ( x, y, z ) k be a vector field for which v1 , v2 and v3 are
continuous first order partial derivatives in some domain containing D. Then,
∫∫ (v.n)dA =∫∫∫ ∇.vdV =∫∫∫ div(v)dV

S D D
where n is the outer unit normal vector to S.
Remark: The given domain D can be subdivided into finitely many special regions such that each
region can be described in the required manner. In the proof of the divergence theorem, the
special region D has a vertical surface. This type of region is not required in the proof. The
region may have a vertical surface. For example, the region bounded by a sphere or an ellipsoid
has no vertical surface. The divergence theorem holds in all these cases. The divergence theorem
also holds for the region D bounded by two closed surfaces.
Remark: In terms of the components of v, divergence theorem can be written as
 ∂v1 ∂v2 ∂v3 

∫∫ v dydz + v dzdx + v dxdy
S
1 2 = ∫∫∫ 
3 +
 ∂x ∂y D
+
∂z 
 dxdydz
or as
 ∂v1 ∂v2 ∂v3 

∫∫ (v cos α + v
S
1 2 cos β + v3 cos γ )dA
= ∫∫∫  ∂x + ∂y
D
+
∂z 
 dxdydz .
Example: Let D be the region bounded by the closed cylinder x 2 + y 2 = 16, z = 0 and z = 4.
Verify the divergence theorem if v =3 x 2i + 6 y 2 j + zk .
Solution: We have ∇.v = 6 x + 12 y + 1. Therefore,
2
4
4 =y 16 − x 2
∫∫∫ (∇.v)dV
=
D
∫∫ ∫
z =0
x=
−4 y =
− 16 − x 2
(6 x + 12 y + 1)dydxdz.
4
4 =y 16 − x 2
∫∫∫ (∇.v)dV
=
D
∫∫ ∫
z =0
x=
−4 y =
− 16 − x 2
(6 x + 12 y + 1)dydxdz.
Since x, y are odd functions, we obtain
4 4
=y 16 − x 2
∫∫∫ (∇.v)dV= (4)(2)(2) ∫
D z =0
∫ y =0
dydx= 16 ∫ 16 − x 2 dx
0
1 16  x  4
= 16  x 16 − x 2 + sin −1 =  64π .
2 2  4  0
The surface consists of three parts, S1 (top), S 2 (bottom) and S3 (vertical),
On S1 =
: z 4,=
n k
∫∫ (v.=
S1
n)dA ∫∫=
zdA 4 ∫∫
S1
= dA
S1
4 (area of circular region with radius 4)=64π.
On S 2 : z = 0, n = −k .
∫∫ (v.n)dA =
S2
=∫∫ − zdA =
0.
S2
2 xi + 2 yj 1
On S3 : x 2 + y 2 = 16, n= = ( xi + yj )
2 x2 + y 2 4
1
∫∫ (v.n)dA =
S3
= ∫∫
4 S3
(3 x3 + 6 y 3 )dA.
Using the cylindrical coordinates, we= cos θ , y 4sin

write x 4= = θ , dA 4dθ dz.
Therefore,
4 2π
1
∫∫S (v.n)dA 4=z∫0=θ∫0 [192 cos θ + 348sin θ ]4dθ dz
= 3 3
2π
= 192 ∫ [(cos 3θ + 3cos θ ) + 2(3sin θ − sin 3θ )]
0
3
Hence, ∫∫ (v.n)=
S
dA ∫∫∫ (∇.v)dV .
D
Green’s Identities (formulas)
Divergence theorem can be used to prove some important identities, called Green’s identities
which are of use in solving partial differential equations. Let f and g be scalar functions which
are continuous and have continuous partial derivatives in some region of the three dimensional
space. Let S be a piecewise smooth surface bounding a domain D in this region. Let the
functions f and g be such that v=f grad g Then, we have
∇.( f ∇g ) = f ∇ 2 g + ∇f .∇g
By divergence theorem, we obtain
∫∫ (v.n)dA = ∫∫ f (∇g.n)dA = ∫∫∫ ∇.( f ∇g )dV

S S D
∫∫∫ ( f ∇ g + ∇f .∇g )dV .

2
Now, ∇g .n is the directional derivative of g in the direction of the unit normal vector n.
Therefore, it can be denoted by ∂g / ∂n. We have the Green’s first identity as
∂g
∫∫ f (∇g.n)=
dA ∫∫ f = ∫∫∫ ( g∇ f + ∇g .∇f )dV .
2
dA
S S
∂n D
Interchanging f and g, we obtain
∂f
∫∫ g (∇f .n)=
dA ∫∫ f = dA ∫∫∫ ( g ∇ f + ∇g .∇f )dV .
2
S
∂n
S D
Subtracting the two results, we obtain the Green’s second identity as
 ∂g ∂f 
∫∫ ( f ∇g − g∇f ).ndA
= ∫∫  f − g  dA= ∫∫∫ ( f ∇ g − g∇
2 2
f )dV .
S 
S
∂n ∂n  D
Let f=1. Then, we obtain
4
∂y 2
∫∫ ∇g.ndA =
S
∫∫ S
∂n
dA =∫∫∫
D
∇ gdV .
If g is a harmonic function, then ∇ 2 g =

0 and we have
∂y
∫∫ ∇g.ndA= ∫∫ ∂n dA=
S S
0.
This equation gives a very important property of the solutions of Laplace equation, that is of
harmonic functions. It states that if g ( x, y, z ) is a harmonic function, that is, it is a solution of the
equation
∂2 g ∂2 g ∂2 g
+ + =
0
∂x 2 ∂y 2 ∂z 2
Then, the integral of the normal derivative of g over any piecewise smooth closed orient able
surface is zero.
Suggested Readings
New York.
House, New Delhi.
New Delhi.
5
****** ☺******
This Book Download From e-course of ICAR
Visit for Other Agriculture books, News, Recruitment,
Information, and Events at
www.agrimoon.com
Give Feedback & Suggestion at info@agrimoon.com

Send a Massage for daily Update of Agriculture on WhatsApp
+91-7900 900 676

Disclaimer:
The information on this website does not warrant or assume any legal liability or
responsibility for the accuracy, completeness or usefulness of the courseware contents.
The contents are provided free for noncommercial purpose such as teaching, training,
research, extension and self learning.
Connect With Us:
AgriMoon App AgriVarsha App

App that helps the students to gain the Knowledge App that helps the students to All Agricultural
about Agriculture, Books, News, Jobs, Interviews of Competitive Exams IBPS-AFO, FCI, ICAR-JRF,
Toppers & achieved peoples, Events (Seminar, SRF, NET, NSC, State Agricultural exams are
Workshop), Company & College Detail and Exam available here.
notification.

Engineering Mathematics I

Uploaded by

Copyright:

Available Formats

Engineering Mathematics I

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Engineering Mathematics I

Uploaded by

Copyright:

Available Formats

Engineering

Prof. C. Nahak, Prof. J. Kumar & Prof. S. Kumar

Mathematics Department, IIT Kharagpur, Kharagpur 721 302)

AgriMoon App AgriVarsha App

Lesson 25: Introduction 229-233

Rolle’s Theorem, Lagrange’s Mean Value Theorem , Cauchy’s Mean Value

Theorem 1. 1 [Rolle's Theorem]: Let be continuous on the closed interval

and differentiable on the open interval . If , then there

exists at least one number in such that .

Proof: Assume . If and , then we

consider instead of . Since is continuous on it attains

its bounds: Let and be both maximum and minimum of on . If

, then is throughout i.e., is constant on

for all in . Thus at least one such that .

Suppose . If varies on then there are points where

or points where Without loss of generality assume and the

function takes the maximum value at , so that . It is to be noted

that if , , which is a contradiction. Now as

is the maximum value of the function, it follows that ,

both when and .

when . Since it is given that the derivative at exists, we get

when and when . Combining the two

exists at least one zero o i.e., clearly is continous on [-1,1]

Example 1: Verify the Roll's theorem for .

(i) , (ii) is differentiable on , so all conditions of

Roll's theorems are satisfying. Hence implies and

, is continuous. But is not differentiable at .

Note that , for which is differentiable. As , for

is an odd degree polynomial, hence it has at least one

real root as complex roots occurs in pair.

Suppose two real roots such that , then on , all

properties of Roll's theorem satisfied, hence , such that ,

1.2. Mean Value Theorems

Theorem 1.2 [Lagrange's Mean Value Theorem]: If a function is

continuous on , differentiable , then there exists at least one point ,

such that . Hence Lagrange's mean

value theorem can be written as

Geometrical Representation: If all points of the arc there is a tangent line,

chord connecting the points and .

1.2.1 Cauchy's Mean Value Theorem

Theorem 1.2 [Cauchy's Mean Value Theorem]: It states that if functions

open interval and then there exists some , such

1.2.2 Another form of the statement: If and are derivable in

and for any , then there exists at least one

number such that

Example 4: Write the Cauchy formula for the functions ,

1.2.3 The Intermediate Value Theorem It states the following: If is

continuous on , and is a number between and , then there is a

1.2.4 Applications of the Mean Value Theorem to Geometric properties of

Let be a function which is continuous on a closed inteval and assume

has a derivative at each point of the open interval . Then we have

1. (i) If , is strictly increasing on .

Intermediate value Theorem for Derivatives: If exists for ,

with then for any number between and there is a

Application: If exists with , on any interval then has a

differentiable inverse, there.

Converse of Rolle’s theorem : - (need not true).

Example 1.5 Let be continuous on and differentiable . If

such that , does it follow that ?

No: Take for example on , implies .