Applied Mathematics III

Module Introduction

This module consists of five units. The first unit deals with ordinary differential equations. In
this unit we will look briefly at many terms and concepts related to ordinary differential
equations and discuss analytical methods of solving first and second order ordinary differential
equations. The second unit deals with Laplace transforms; Inverse Laplace Transform. The
second unit introduces and briefly discusses various methods of finding Laplace and Inverse
Laplace transforms and applications of Laplace Transform to solve homogenous and non-
homogenous differential equations and Integral equations will also be treated thoroughly in the
second unit. The third unit deals with Fourier series and Fourier integrals; and Fourier
transformations, which are the most useful type of mathematical tools in physical sciences, and
their applications. The fourth unit discusses on calculus of vector valued functions and vector
fields, specifically focuses on the limit, continuity and derivatives of vector functions and vector
fields and their applications. The fifth unit deals with calculus of functions of complex variables
together with their applications. By doing so students will be able to express terms and concepts
related to ordinary differential equations, Laplace Transform, Fourier series and Fourier
integrals; and Fourier transformations, calculus of vector functions, vector fields and functions
of complex variables
SOME FEATURES OF THE MODULE.
Quick Check Exercises: Each exercise set begins with approximately five practice exercises that
are designed to provide students with an immediate assessment of whether they have mastered key
ideas from the section.
Applicability : One of the good feature, primary goals of this module is to link calculus to the real
world and the student’s own experience. This theme is carried through in the examples and
exercises.
Career Preparation: This module is written at a mathematical level that will prepare students for
a wide variety of careers that require a sound mathematics background, including engineering, the
various sciences, and business.
Cooperative learning: One of the primary goals of this module is also to promote cooperative
learning, so that students share knowledge and skills through a lot of group discussions and
group activities given at each of new ideas introduced.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 1

 Applied Mathematics III

Module Objectives:
At the end of this module students will be able to:
 Understand the concept of ordinary differential equations.
 Define ordinary differential equations.
 Correctly identify differential equations of various orders, degrees and type;
 Compute general and particular solutions of linear differential equations
 Demonstrate the application of differential equations together with solving exercises
 Apply and Appreciate the concept of differential equations in solving real life
problems.
 Define Laplace Transforms
 Finding Laplace transforms of some functions using basic principles
 Identify Different Properties of Differential Laplace Transforms
 Apply Laplace transforms to solve differential equations modeling physical
problems
 Distinguish Fourier series and Fourier integrals; Fourier transformation.
 Demonstrate the applications of Fourier series, Fourier integrals, Fourier and Laplace
transformations together with solving exercise.
 Apply the concept of Fourier series, Fourier integrals, Fourier and Laplace
transformations in solving real life problems
 Define scalar and vector fields.
 compute curl and divergence, and know what they have to say about the nature of a
vector field.
 Evaluate line and surface integrals. Identify when a line integral is independent of
path and use the Fundamental Theorem of Line Integrals to solve applied problems
 Find the curl and divergence of a vector field, the work done on an object moving in a
vector field, and the flux of a field through a surface. Use these ideas to solve applied
problems
 Identify and use Green’s Theorem, the Divergence (Gauss’s) Theorem and Stokes’
Theorem.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 2

 Applied Mathematics III

UNIT I

1. ORDINARY DIFFERENTIAL EQUATIONS

Unit Introduction
This unit is divided into three sections. The first section presents Introduction to basic concepts
and ideas of differential equations. In the second section, we discuss various types of First-Order
differential equations that can be solved analytically. Particularly, the second section examines
Separable and non-separable differential equations, Exact and Inexact differential equations
(integrating factors), Linear and Some Special Non-linear Differential Equations of the first
order. Linear second order ordinary differential equations. The third section introduces linear
second order ordinary differential equations. In the third section second order homogeneous and
non-homogeneous ordinary differential equations are discussed and their solutions obtained with
a variety of techniques. Euler-Cauchy equation and systems of differential equations are also
parts of the third section. The learning activities in this unit include self-study, reading
assignments, group discussions, and problem solving.
Unit Objectives: Specific learning Objectives
By the end of this unit, the learner should be able to :
 Define Differential Equations
 Correctly identify differential equations of various orders, degrees and type;
 Form a differential equation by elimination of arbitrary constants;
 Mention Different Physical Examples of Differential equations
 Demonstrate how to differentiate separable and non-separable, exact and non
exact, linear and nonlinear first order differential equations together with solving
exercises.
 Apply the concepts of ordinary differential equations in real life problems.
 Identify homogeneous and non-homogeneous second order differential equations.
 Demonstrate how to compute general and particular solutions of second order
differential equations together with solving exercises.
 Demonstrate how to transform systems of differential equations to second order
differential equations together with solving exercises.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 3

 Applied Mathematics III

1.1 Introduction to Differential Equations

1.1.1 Basic Concepts, terminology and Physical examples of Differential Equations
Definition1.1.1: Ordinary and partial differential equations
A differential equation (DE) is an equation involving derivatives of unknown function of one or
more variables.
A differential equation containing only ordinary derivatives of an unknown function(s) with
respect to a single independent variable is said to be an ordinary differential equation (ODE).
A differential equation involving partial derivatives of one or more unknown functions of two or
more independent variables is called a partial differential equation (PDE).

Examples:

(a) The equations

dy dy d 2 y dy dx dy
 2x  y ,  5 y  e x ,   6 y  0 ,and   2 x  y (an ODE can contain more
dx dx dx dy dt dt
than one unknown function) are examples of ordinary differential equations since the function
y  f (x) depends on only one variable x . In the function y  f (x) , x is called the independent
variable, and y is the dependent variable.
(b) The following equations

 2u  2u  2u  2u u u u y
  0,  2 2  and  2x  z
x 2
y 2
x 2
t t y x x
are partial differential equations since the functions u  f ( x, y) , u  f ( x, t ) , y  f ( x, z ) in each
equations depends on two variables.
Typical Examples of differential equations in applications are the following:
dR
i. The decay of a radioactive substance is described by an equation of the form  kR ,
dt
where k is a constant.
d 2x  dx 
ii.The differential equation m  F  t , x,  describes Newton's law for position xt  of a
 dt 
2
dt
dx
particle acted on by a force F of t, x and ,the velocity.
dt

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 4

 Applied Mathematics III

d 2x
m 2  k1  k 2 x  F t  where F t  is a given function and
dx
iii. The differential equation
dt dt
m, k1 , k 2 are constants, governs the motion of a vibrating mass at the end of a vertical spring.

 2u  2u
iv. c  2
( governs the motion of wave Equation in one dimension)
x 2 t 2
 2u u
v. a  2
(governs the motion of Heat Equation in one dimension)
x 2 t
 2u  2u
vi.   0 ( Laplace Equation in two dimension)
x 2 y 2

d 2q
 q  sin t models the charge q  qt  on a capacitor in an electrical circuit
dq 1
vii. L 2
R
dt dt C
containing an inductor, resistor and a capacitor, where the current is driven by a sinusoidal
electromotive force operating at the frequency  .

Differential equations may be classified further according to their order and degree.

Definition 1.1.2: Order and degree of a differential equation

The order of a DE is the order of the highest ordered derivative involved in the expression. The
degree of an ordinary differential equation is the algebraic degree of its highest ordered–
derivative i.e., the power to which the highest-order derivative is raised, after the equation has
been rationalized to contain only integer powers of derivatives.
Examples:
dy
i.  2 x  y is first order, first degree ordinary DE
dx
y
ii.  2 x  z is first order, first degree partial DE
x
d 2x dx
iii.  2  15 x  0 is second order, first degree ordinary DE
dt 2
t
3
d 3 y  dy  2
iv.     x 2 y  0, is of third order and second degree, since after rationalization it
 dx 
3
dx
2
 d3y 
contains the term  3  .
 dx 

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 5

 Applied Mathematics III

v.  y' '2  2 xy  e x , is a second order and second degree ode

 3

 
2
1      k d y is a second order and second degree ode
dy 2
vi.
  dx   dx 2
 
2
 d3y 
3
d 2 y dy  dy 
vii.  3   2 2  x 2    0 is third order and second degree.
 dx  dx dx  dx 
First order differential equations are occasionally written in differential form as:

M ( x, y)dx  N ( x, y)dy  0 .

   
For example, x  y 2  3 y dx  x 2  3x  y dy  0 is a first order and first degree ode.
Quick Check Group Activity 1.1.1: Using the notes given in section 1.1.1, discuss in a small
group of 3-4 members, order and degree of the following differential equations.
2 3
dy d 3 y  d2y   dy 
  2     3x  1
2
a) 2 5 c)
dx dx 3  dx   
dx
2 5
1 1  d3y   d2y 
d y
2 2
 dy  3 d)  2    2   y  e x
b)  2    x    dx   dx 
 dx   dx 

Instructor's role
Check and give feedback to their answers

All the differential equations to be discussed in the next two sections are ordinary differential
equations.
Notation: Throughout this module ordinary derivatives will be written by using either the
dy d 2 y d 3 y
Leibniz notation , 2 , 3 ,... or in the prime notation y, y ' , y '' ,... .
dx dx dx
In symbols we can express an nth-order ordinary differential equation in one dependent variable
 
by the general form: F x, y, y ' , y '' , y ''' ,..., y n   0 or in Leibnitz notation as

dy d 2 y d 3 y dny
F ( x, y, , 2 , 3 ,..., n )  0 (*)
dx dx dx dx
where F is a real-valued function of n  2 variables: x, y, y ' , y '' ,..., y n  .

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 6

 Applied Mathematics III

On the assumption that it is possible to solve an ordinary differential equation in the form (*)
uniquely for the highest derivative y n  in terms of the remaining n  1 variables, we can write (*)
as:
dny
dx n

 f x, y, y ' , y '' , y ''' ,..., y n1  (**)

The differential equation (**) where f is a real-valued continuous function, is referred to as the
normal form of (*). Thus the normal forms to represent general first- and second-order ordinary
differential equations respectively are:
dy
dx
 f x, y  and
d2y
dx 2

 f x, y , y ' 
For example, the normal form of the first-order equation 4 xy '  y  x is y ' 
x  y  ; the normal
4x
form of the second-order equation y ''  y'6 y  0 is y ''  y'6 y .
Classification by Linearity:

Definition1.1.3: An nth-order ordinary differential equation F x, y ' , y '' , y ''' ,..., y n   0 is said to 
be linear if F is linear in y, y ' , y '' ,..., y n  . This means that an nth-order ODE is linear when it can

be written as an ( x) y n   an1 x y n1  an2 x y n2   ...  a2 x y ''  a1 x y '  a0 x y  g x   0 or

dny d n1 y d2y

an  x       a1 x   a0 x  y  g x 
dy
n
 a n 1 x n 1
 ...  a 2 x 2
dx dx dx dx

In the additive combination on the left-hand side of the above equation we see that the
characteristic properties of a linear ODE are as follows:
 The dependent variable y and all its derivatives y, y ' , y '' ,..., y n  are of the first degree, that
is, the power of each term involving y is 1.
 The coefficients a0 , a1 , a2 ,..., an of y, y ' , y '' ,..., y n  depend at most on the independent
variable x.
A nonlinear ordinary differential equation is simply one that is not linear. Nonlinear functions of
1
the dependent variable or its derivatives, such as sin y , e y , and so on or products of the
y
dependent variable or its derivatives , cannot appear in a linear equation.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 7

 Applied Mathematics III

Examples : Linear and Nonlinear ODEs

(a) The equations
d 3y
 y  xdx  4xydy  0 , y ''  2 y  y  0 and x 3 3
dy
 x  5y  ex
dx dx
are, in turn, linear first-, second-, and third-order ordinary differential equations.
(b) The equations

are examples of nonlinear first-, second-, and fourth-order ordinary differential equations,
respectively.
Two important special cases of linear ODEs are linear first-order (n =1) and linear second order
(n= 2) DEs:
d2y
a1 x   a0 x  y  g x  and a2  x   a1 x   a0 x  y  g x 
dy dy
2
dx dx dx

1.1.2 Nature of Solutions of ODE: Particular and General solution.

Definition 1.1.4 Solution of an ODE
Any functional relation between the dependent variable y and the independent variable x, defined
on some interval I and possessing at least n derivatives that are continuous on I, which satisfies
the differential equation, is said to be a solution or an integral of the equation on the interval.

Interval of Definition: The interval I in Definition is variously called the interval of

definition, the interval of existence, the interval of validity, or the domain of the solution and
can be an open interval (a, b), a closed interval [a, b], an infinite interval (a,  , and so on.
Examples: Verification of a Solutions
1. Verify that the function y  x 2 is a solution of xy  2 y for all x on the interval  ,  .
Solution: One way of verifying that the given function is a solution is to see, after substituting,
whether each side of the equation is the same for every x in the interval. Since y  x 2 

y  2 x , then substituting y  x 2 and y  2 x in xy  2 y we obtain that

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 8

 Applied Mathematics III

Left hand side : xy  x(2 x)  2 x 2

Right hand side: 2 y  2x 2

we see that each side of the equation is the same for every real number x.

Explicit and Implicit Solutions:

A solution in which the dependent variable is expressed solely in terms of the independent
variable and constants as an explicit formula y  f x  is said to be an explicit solution.

Example: y  x 2 is an explicit solution of xy  2 y for all x on the interval  , 

A relation Gx, y   0 is said to be an implicit solution of an ordinary differential equation on
an interval I, provided that there exists at least one function y that satisfies the relation as well as
the differential equation on I.
Example: x 2  y 2  1  0 is an implicit solution of yy   x for all x on the interval  1,1 .

An n th order ordinary differential equation may have more than one solution which can be
represented by a single formula involving an arbitrary constants such a solution is called a
general solution.
Example: By integrating both sides of the first order equation y  cos x we obtain

y sin x  c which is a general solution of y  cos x , where c is an arbitrary constant. By

4
assigning definite values [ c  0 c  3 , c   ] to the constant, each of the functions
5
4
y  sin x, y  sin x  3, y  sin x 
5
is a particular solution of the differential equation y  cos x satisfying the initial conditions

y0  0 , y0  3 , y 0   i.e. respectively.

4
5
3. Again by simple integration of both sides of the second order ordinary differential equation
x3
y ''  2 x twice we obtain y   c1 x  c2 which is a solution of y ''  2 x where c1 and c2 are
3
arbitrary constants.
On a given interval, a solution of a differential equation from which all solutions on that interval
can be derived by substituting values for arbitrary constants is called a general solution of the

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 9

 Applied Mathematics III

equation on the interval. A solution obtained by assigning definite values to the arbitrary
constants in the general solution is called a particular solution.
Although the prime problem in the study of differential equations is finding the solution to a
given differential equation, the converse problem is also interesting.That is the problem of
finding a differential equation which satisfies by a given group of functions, each one of which is
a solution of the ODE. The problem is solved by repeated differentiation and elimination of the
arbitrary constants.

Thus, in general, a group of functions y  f ( x, c1 , c2 ,....cn ) with n parameters satisfies an nth-

order ODE in general (although in some degenerate cases an ODE of less than nth order is
obtained). The intuitive converse of this is that the general solution of an nth-order DE depends
on n arbitrary constants; i.e., the solution y depends on x and the real constants c1 , c2 ,....cn ; for
our purposes, this will be assumed to be valid although a totally general proof is difficult.

A general solution is therefore a family of solutions defined on some interval I that contains all
solutions of the DE that are defined on I. A solution that is not a member of a family of solutions
of the equation that is, a solution that cannot be obtained by specializing any of the parameters in
the family of solutions is called a singular solution.

Solution Curve:
The graph of a solution of a differential equation is called an integral curve or a solution curve
for the equation, so the general solution y  f ( x, c1 , c2 ,....cn ) of a differential equation produces
a family of integral curves corresponding to the different possible choices for the arbitrary
constants. Since y  f x  is a differentiable function, it is continuous on its interval I of
definition. Thus there may be a difference between the graph of the function and the graph of the
solution . Put another way, the domain of the function y  f x  need not be the same as the
interval I of definition (or domain) of the solution .

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 10

 Applied Mathematics III

Group Activity 1.1.1

1. Fill in the blanks for each of the Differential Equations.

Differential equation Ordinary Order Degre Independen Dependen

or Partial e t variable t variable
2
d 2 y  dy 
1 y   5
dx 2  dx 

d 3 y dy
2 
dx 3 dx

u  2 u u
3 4 2 
t dx dy
2
 d 3s   d 2s 
4  3    2   s  3t
 dt   dt 
3

d y   dy   2
2 2
5  1    
dx 2   dx  

6 x  y dx2  2xydxdy  dy 2  0

3. A certain drug is being administered intravenously to a hospital patient. Fluid containing

5mg/cm3 of the drug enters the patient’s bloodstream at a rate of 100 cm3/h. The drug is
absorbed by body tissues or otherwise leaves the bloodstream at a rate proportional to the
amount present, with a rate constant of 0.4 (h)−1.
(a) Assuming that the drug is always uniformly distributed throughout the bloodstream, write a
differential equation for the amount of the drug that is present in the bloodstream at any time.
(b) How much of the drug is present in the bloodstream after a long time?
4. For small, slowly falling objects, the assumption made in the text that the drag force is
proportional to the velocity is a good one. For larger, more rapidly falling objects, it is more
accurate to assume that the drag force is proportional to the square of the velocity.
(a) Write a differential equation for the velocity of a falling object of mass m if the drag force is
proportional to the square of the velocity.
(b) Determine the limiting velocity after a long time.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 11

 Applied Mathematics III

Assessment
 Check students participation in the group activity.
 Give feedback to their answers

1.2 First Order Ordinary Differential Equations

dy
First- order ODEs contain only equated to some function of x and y, and can be written in
dx
either of two equivalent standard forms,

 F x, y  or
dy
dx
M ( x, y)dx  N ( x, y)dy  0
where F x, y  , M x, y  and N x, y  are in general functions of both x and y.
Definition: (Initial value problem)
A first order differential equation together with an initial condition, y( x0 )  y0 , is called the
initial value problem.
For example, y 2 xy  x  1  0, y(1)  3 is an initial value problem.
Theorem 1.2.1: Existence of a unique solution ( Picards theorem)
For a differential equation of first order, viz.

 f  x, y 
dy
dx
which satisfies the initial condition
y( x0 )  y0
there exists a unique solution y   x  , defined for all values of x in a certain region including

x0 , satisfying the given differential equation and taking on the value y0 when x  x0 ,provided

f
that f x, y  and are continuous single valued functions of in the region
y

x  x0   , y  y0  

where   0 and   0 .
We shall, in the remainder of this section, discuss some analytical methods of solving various
types of First Order and First-Degree differential equations. However, despite the apparent
Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 12
 Applied Mathematics III

simplicity of this equation, analytic solutions are usually possible only when F ( x, y) has simple
forms. Which of the two above forms is the more useful for finding a solution depends on the
type of equation being considered.
1.2.1 Separable Differential Equations; Reduction to Separable Form
Definition: A differential equation which can be written in the either form:

 f ( x) g  y  Basic form)
dy
(1.2.1a)
dx

or equivalently, differential equation which can be written in the form:

f ( x)dx  g ( y)dy (1.2.1b)
where f is a function of x alone and g is a function of y alone, is said to be separable
differential equation.
Equation (1.2.1a) and ( 1.2.1b ) is called separable because the variables x and y are separated.

Since the variables x and y are now separable , we have ,from (1.2.1a),
dy
 g ( y)   f ( x)dx , (1.2.1c).

which expresses y implicitly in terms of x .

Similarly integrating both sides of (1.2.1b), we obtain

 f ( x)dx   g ( y)dy  c ,where c is an arbitrary constant.

This is the general solution of the differential equation.
If the solution can be defined explicitly, i.e., it can be solved for y as a function of x, then do it. If
not, the solution can be defined implicitly, i.e., it cannot be solved for y as a function of x.
Examples:
1.Solve the following differential equations.
a) y  1  y 2 b) xy  y  0, y(1)  1

c) x 2 1  y   1  x  y 2  0 d) x  1dy   y  1dx  0 , y0  3

dy
dx
Solutions:
dy
a) y  1  y 2   1  y2
dx

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 13

 Applied Mathematics III

dy
  dx
1 y2
Integrating both sides, we obtain
arctan y  x  c  y  tan(x  c) . 

b) xy  y  0, y(1)  1
 xy   y
dy
x y
dx
dy dx
  .
y x
Integrating both sides we obtain
ln y   ln x  c .
From this we obtain the general solution
c
y .
x
1
By using the initial condition, y(1)  1, we see that c  1 and hence y ( x)  .
x

c) x 2 1  y   1  x  y 2  0
dy
dx
This can be put into the form
dy  y 2   1  x  
  
dx  1  y  x 2 


1 y 1  x 
dy   2 dx
2
y x
where the variables are separated.
Integrating both sides we get
 1 1  1 1
  y 2
 dy     2  dx
y x x
which gives

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 14

 Applied Mathematics III

1 1
 ln y  ln x   ln c ,
y x
where c is an arbitrary constant. Simplification gives
 y 1 1
ln    
 cx  x y
which is the general solution of the differential equation.
d) x  1dy   y  1dx  0 , y0  3 .
Separating the variables we obtain
dx dy

x 1 y 1
Integrating both sides we get
ln x  1   ln  y  1  ln c,
or x  1 y  1  c
Substituting x  0, y  3 in the last equation we obtain c  2.

Hence the solution of the given initial value problem is x  1 y  1  2.

Quick Check Exercises 1.2.1: Solve the following Differential equations
x 1 1
a) y   , y (0) 1 c) y   y  0
2y y

b) y  e y x 2 , y0  ln 2 d) y  x  1(1  y 2 )

Teachers role
 Check and give feedback to their answers
Group Activity: 1.2.1
1. Solve the following differential equations.

a) y  e x y 3 e) y 2 xy  x  1  0, y(1)  3

 
b) x 3  1 dy  ydx  0 f) ( x ln x) y  y
c) yy  sin 2 x, y(0)  3 x
h) y   , y(2)  5
y
7
d) y  y tanh x. y (0) 
i) ( x  1) y' y  1  0, y(0)  1
5 2 2

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 15

 Applied Mathematics III

dy dy
j)  e x y  x 2 e y k)  e 2 x y  x 3e  y
dx dx
 dI 
g. L   RI  0, I (0)  I 0 , where L and R are cons tan ts
 dt 
2. (Newton’s Law of Cooling). A copper ball is heated to a temperature of 1000 c . Then at a
time t  0 it is placed in water which is maintained at a temperature of 300 c . At the end of 3
minutes the temperature of the ball is reduced to 700 c . Find the time at which the temperature of
the ball is reduced to 310 c . (Hint: the rate of change of the temperature T of the ball is
proportional to the difference between T and the temperature of the surrounding medium.)
 dr 
3. The stress P in thick cylinders of radius r is given by r    2 p  2c. Find P in terms of r.
 dp 
Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers

1.2.1.1 Reduction to Separable Form

Certain first order differential equations are not separable but can be made separable by a simple
change of variables.
Method of Transformation of Variables
First order differential equations of the form

y  f (ax  by  c) , (*)

where a , b and c are real constants, can be made separable by substituting

z  ax  by  c .
From this, it follows that
z  a  by .
That is,
z  a
y  .
b
Substituting this in (*) we get,

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 16

 Applied Mathematics III

z  a
 f (z ) .
b
This can be separated as
dz
 dx . 
a  bf ( z )
Examples: Solve the following differential equations

a) y  ( x  y 1) 2  2 b) y  1  x  y  2 , y(0)  2

 cot  y  x   1
dy
c)
dx
Solution:
a) Letting

z  x  y 1 ,
we have
y  z   1 .
Substituting this in the given equation, we get
z  1  z 2  2 .

This implies
dz
z 2
1 
 dx ,

z 1
 ln  2x  c
z 1

z 1
  ce 2 x .
z 1
Hence

1  ce 2 x
z .
1  ce 2 x
But since
z  x  y 1 ,

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 17

 Applied Mathematics III

then
1  ce 2 x
y  z  x 1   x 1
1  ce 2 x

1  ce 2 x
 y ( x)   x 1 . 
1  ce 2 x

b) y  1  x  y  2, y(0)  2
Let
zx y2 (*)
 y  z  x2
y  z  1 .
Substituting this in the given equation, we get
z  z.
This implies
dz
 z
  dx ,

 2 z  xc .
Substituting this in (*) we get
xc
x y2  ,
2

 xc
2

 x y2    .
 2 
Thus

 xc
2

y ( x)     x 2.
 2 
But y(0)  2 implies c  4
Thus

 x 4
2

y ( x)     x2.
 2 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 18

 Applied Mathematics III

 cot  y  x   1 .
dy
c) Solve the differential equation
dx
Putting y  x  v, we obtain by differentiation of both sides we get

dy dv dy dv
1  or  1
dx dx dx dx
The given differential equation now becomes
dv dv
 1  cot v  1 or  cot v
dx dx
in which variables are separable, and we obtain
dv sin v
dx   dv.
cot v cos v
Integration now gives

x   ln cos v   ln c  ln
c c
 ln ,
cos v cosx  y 
which is the required solution.
Quick Check Class Activity 1.2.2: Solve the following differential equations

 cosx  y  1
2
dy dy  x  y  1 
a) c)   .
dx dx  x  y  3 

 sin 2 x  y  1
dy
b) dy
dx d)  4x  2 y  1
dx
Instructor's role
 Check and give feedback to their answers

1.2.2 Homogeneous equations

Any function of the form f x, y  is said to be homogenous function of degree n, if

f tx, yt   t n f x, y 

that is, if x and y be replaced by tx and ty , t n factors out of the resulting function, and the
remaining factor is the original function. For example,
x x y 1
x 2  2 y 2 , sin , ln , x  y  xy ,
y x x y
are homogeneous of degrees 2, 0, 0, 1 and -1, respectively, since

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 19

 Applied Mathematics III

tx 2  ty 2  t 2 x 2  2y 2 
tx x
sin  sin
ty y
tx  ty x y
ln  ln
tx y

yx  ty  txty  t ( x  y  xy )

1 1
 t 1
tx  ty x y
An equation of the form

 f  x, y 
dy
dx
is said to be homogeneous whenever the function f is homogeneous of degree zero and so is a

 f x, y  is homogenous f does not depend on x and

y dy
function of . Thus, if y
x dx
y x
separately, but only on their ratio or .
x y
Thus if an equation is homogenous it can be written as
 y
y  g  . (*)
 x
y
where g is any given function of .
x
y
To solve such equations we set v  , assuming that y and v are functions of x . Then
x
y  vx and hence
y   v' x  v .
Substituting this in (*) we obtain

v ' x  v  g (v) ,
dv
thus x  g (v )  v
dx
This can be separated as

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 20

 Applied Mathematics III

dv dx
 .
g (v )  v x

y
If we integrate this and replace v by , we obtain the general solution of (*) .
x
Examples:
a) Solve the differential equation 2 xyy  y 2  x 2  0 .

Solution: Since
2 xyy  y 2  x 2  0
can be re written as
2 xyy  y 2  x 2
y x
 2 y   . (*)
x y
y
Now let v .
x
 y  vx

 y  v ' x  v .
Then from ( * ) it follows that
1
2(v ' x  v)  v 
v
2dv v2  1
 x 
dx v
2v dv dx
  
v 1
2
x
Integrating both sides yields
ln | v 2  1 |  ln | x |  c
c
 v2 1
x
y
Now, setting v  we obtain
x

y 2  x 2  cx .
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 21
 Applied Mathematics III

dy x 2  xy  y 2
b) Solve the differential equation  .
dx x2

Solution: The equation can be written as

2
dy x 2 xy y 2 y  y
 2  2  2  1   
dx x x x x  x
which is homogeneous type.
y dy dv
Substituting v  or y  vx , we obtain by differentiation v  x and the given
x dx dx
differential equation becomes

 1  v  v 
dv
vx
2

dx

 1  v 
dv 2
or
dx
On separating the variables, this gives
dx dv

x 1  v2
Integrating both sides, we obtain
y
ln x  tan  v  c  tan  c
x
 y
 tan 
Hence x  c1e  x
which is the general solution of the differential equation.
dy 2 xy
c) Solve the differential equation  2 and find a solution subject to the condition that
dx x  y 2
y  1 when x  1.
Solution: The equation can be written as
y
2
dy x
 2
dx  y
1  
x
which is homogeneous type.
dy dv
Putting y  vx and v  x the above differential equation becomes
dx dx

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 22

 Applied Mathematics III

dv 2v
vx 
dx 1  v 2
dv 2v v  v3
x   v 
dx 1  v 2 1  v2

Separating the variables and integrating both sides, we obtain

 
ln x  ln v  ln 1  v 2  ln c
y
cv cxy
Hence x c x 2  2
1 v 2
y x  y2
1 2
x
or x 2  y 2  cy
Substituting x  1 and y  1 in the above, we obtain c  2 . Hence the particular solution is

x2  y2  2 y .
Quick Check Exercises: Solve the following differential equations.
a) ( x 2  y 2 )dx  2 xydy  0 dy y 2
c) x  y
dx x
dy y 3  3x 2 y
b)  dy y y
dx x 3  3xy 2 d)   sin
dx x x
Instructor's role
 Check and give feedback to their answers

1.2.2.1 Equations Reducible to homogenous form

Sometimes the given differential equation may not be homogenous but may be reducible to
homogenous type by means of substitution. Let us consider the differential equation in the form

ax  by  cdx  a' x  b' y  c'dy  0

or ax  by  c dx  a' x  b' y  c'dy  0 ( where
a b
 )
a' b'

We can write the equations as

dy ax  by  c  a b 
 ;  
dx a' x  b' y  c'  a' b' 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 23

 Applied Mathematics III

This can be solved by reducing it to homogenous form by using the transformations

dy dY
x  X  h; y  Y  h, then dx  dX ; dy  dY hence  . Therefore the equation reduces to
dx dX

dy aX  bY  ah  bk  c

dx a' X  b'Y  a' h  b' k  c'

To simplify we choose h and k so that ah  bk  c  0 and a' h  b' k  c'  0 .

For these values of h and k the equation reduces to

Y 
a  b 
dy aX  bY X
 
dx a' X  b'Y Y 
a'  b'  
X
which is homogenous equation.
After solving this equation we replace X  x  h;Y  y  h, to get the solution of the given
equation.
a b a b
When '
 ' ( a case of failure) we put '  '  m ,(say),
a b a b
dy ax  by  c
Then  becomes
dx a' x  b' y  c'
dy ax  by  c

dx m(ax  by )  c'
Now by putting ax  by  v , the equation can be brought to variable separable form.
Examples:
dy x  2 y  3
a) Solve 
dx 2 x  y  3
a b 1 2
Solution: We have '
 ' i.e., 
a b 2 1
Put x  X  h; y  Y  h,
dy X  2Y  h  2k  3
 
dx 2 X  Y  2h  k  3

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 24

 Applied Mathematics III

Choose h  2k  3  0 and 2h  k  3  0 . These two equations give h  1 and k  1 and therefore

x  X  1, y  Y  1 . On substituting in the differential equation, we get

dy X  2Y

dx 2 X  Y
dY dV
which is homogeneous and may be solve by substitution Y  VX and V  X . Then
dX dX
we have
dV 1  2V
V X 
dX 2 V
dV 1  2V 1V 2
X  V 
dX 2 V 2 V


2  V dV 
dX
1V 2
X
Integrating yield
2 V dX
 1V 2
dV  
1V 2
dV  
X
C

1V  1
 ln  
  ln 1  V  ln CX
2

 1V  2
Substituting x  X  1, y  Y  1and simplifying the required general solution is

x  y  2  K x  y 3 .
dy  x  y  1
b) Solve 
dx 2 x  2 y  1
Solution:
a b 1 1
Now we have '
 ' m   m . The equation may be written as
a b 2 2
dy ( x  y)  1

dx 2( x  y )  1
dv dy dy dv
Now by putting x  y  v and  1    1 , the equation can be brought to variable
dx dx dx dx
separable form. Use the method of separation of variables to solve (Class Exercise!!!)
Quick Check Exercises: Solve the following differential equations.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 25

 Applied Mathematics III

dy x  y  2 dy 2 y  x  5
a)  b) 
dx y  x  4 dx 2 x  y  4

Instructor's role
 Check and give feedback to their answers

Group Activity 1.2.2

1. Solve the following differential equations without and with initial conditions

a) y 
y 2 x 3 cos x 2
x

y
 
, y   0.
h) y  sin 2 ( x  y)  0
y2  2 x y
i) y 
 y x2
b) xy  y  x sec 
x
j) y  ( x  y  2) 2
c) ( x3  3xy 2 )dx  ( y 3  3x 2 y)dy  0
dy y y
k.   tan
d) xyy  2 y  4 x , y(2)  4
2 2
dx x x

 y l. ydx  2 y  x dx  0
e) xy   y  x sec , y (1)   .
2

 x m. x  y dx   y  x dy  0
yx n. ( x 2  y 2 )dx  ( xy  x 2 )dy
f) y  , y (1) 1.
yx
dy  x3
x y o. x  y  2 , y (1)  2
g) y  dx y
x
x
p. (ln x  ln y  1)dy  dx, y(1)  e
y

q. ( x  3xy )dx  ( y  3x y)dy  0

3 2 3 2

x y
2. Find the curve that passes through the point ( e , e ) and has the slope 
y x
1  2 y  4x
3. Solve y  . (Hint: use v  y  2 x ).
1  y  2x

Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 26

 Applied Mathematics III

1.2.3 Exact Differential Equations

We consider here a special kind of non-separable differential equation called an exact differential
equation.

Definition:
A first order differential equation of the form
M ( x, y)dx  N ( x, y)dy  0
is said to be exact if there exists a function, denote by U ( x, y) , such that

U U
dU  dx   M ( x, y)dx  N ( x, y)dy .
x y

That is, if the first member of the equation is the exact differential of a function of x and y. The
differential equation then takes the form
dU  0
and its solutions are defined implicitly by
U x, y   C
where C is an arbitrary constant.
U
Note: i)  M  U ( x, y)   M x, y dx  k  y 
x
U
ii)  N  U ( x, y)   N x, y dy  g x 
y
iii) Every Separable equation is exact.

Theorem: The necessary and sufficient condition for the differential equation
M ( x, y)dx  N ( x, y)dy  0 (*)
to be exact is
M N
 .
y x
Proof: Exercise.

Procedure of Solution:
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 27
 Applied Mathematics III

M N
Step 1:Check whether differential equation written in the form (*) satisfies  or not.
y x
M N
Step 2: If for given equation  is satisfied then there exists a function U ( x, y) for
y x
which
U
M
x
Integrating with respect to x, while holding y constant, we get

U
 M  U ( x, y)   M x, y dx  k  y  (**)
x

where the arbitrary function 𝑘(𝑦) is constant of integration.

U
Step 3:Differentiate (**) with respect to y and assume  N we get
y
U 
M x, y dx  k '  y   N x, y 
y 

y
or

k '  y   N x, y   M x, y dx
y 
(***)

Step 4: Integrate (***) with respect to y and substitute this value in (**) to obtain
U x, y   C , the solution of the given equation.

Remark
a) We could just start the above mentioned procedure with the assumption that
U x, y 
 N x, y  and by integrating N(x,y) with respect to y and differentiating the resultant
y
expression, we would find the analogues of (**) and (***) to be, respectively,
U
 N  U ( x, y)   N x, y dy  g x  and
y

g ' x   M x, y   N x, y dy
x 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 28

 Applied Mathematics III

Examples: Determine whether the following differential equations are exact. If they are exact
solve them by the procedure given in this section.

1. Solve (3x 2 y  e y )dx  ( x 3  xe y  2 y)dy  0

Solution: Now M x, y   3x y  e and N x, y   x  xe  2 y

2 y 3 y

M N
 3x 2  e y and  3x 2  e y
y x
Thus
M N

y x
which implies the differential equation is exact.
Apply procedure of solution above
Let
U
 3x 2  e y
x
Integrating with respect to x, we obtain
U x, y   x3 y  xe y  g ( y)
where g(y) is a constant of integration

Differentiating U x, y   x y  xe  g ( y) with respect to y we obtain

3 y

U
 x 3 y  xe y  g ' ( y)
y
This gives
U
N x, y    x 3 y  xe y  g ' ( y)
y
or g '  y   2 y  g  y    y 2

Substituting this value of g  y    y 2 in U x, y   x y  xe  g ( y) we get

3 y

U x, y   x3 y  xe y  y 2  c .

Thus x y  xe  y  c is the solution of the given differential equation.

3 y 2

Example 2: Solve the following differential equations.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 29
 Applied Mathematics III

2 x sin 3 y dx  (3x 2 cos3 y  2 y)dy  0

Solution: Since

M ( x, y)  2 x sin 3 y , N ( x, y)  3x 2 cos 3 y  2 y
and
M N
 6 x cos 3 y ,  6 x cos 3 y
x x
we get
M N

y x
which implies the differential equation is exact.
But since

u( x, y)   M ( x, y)dx  k ( y)   2 x sin 3 ydx  k ( y)

 x 2 sin 3 y  k ( y) ( *)
and
u
N
y

then 3x 2 cos 3 y  k ( y)  3x 2 cos 3 y  2 y

 k ( y)  2 y  k ( y)  y 2  c1 .

From ( * ) it follows that

u( x, y)  x 2 sin 3 y  y 2  c1  c2 .

x 2 sin 3 y  y 2  c ,
This implies
x 2 sin 3 y  y 2  c
where c  c2  c1 is a constant.

Thus x sin 3 y  y  c is the required solution.

2 2

1 x
Example 3: Solve the differential equation dx  2 dy  0 .
y y

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 30

 Applied Mathematics III

x
Solution: Now M x, y   and N x, y   2
1
y y
M 1 N 1
  2 and  2
y y x y
M N
 
y x
which implies the differential equation is exact.
But since
U
 M  U ( x, y)   M x, y dx  k  y    g  y 
x
x y

U  x, y  
x
 g ( y)
y
where g(y) is a constant of integration

Differentiating with U x, y  

x
 g ( y) respect to y we obtain
y
U  x
 2  g ' ( y)
y y
This gives
U  x
N x, y    2  g ' ( y)
y y
or g'y  0  gy  k
Substituting this value of g  y   k in U x, y  we get

U x, y  
x
k c
y
x
Thus,  C is the solution of the given differential equation.
y
Quick Check Class Exercises: Determine whether the following differential equations are
exact. If they are exact solve them by the procedure given in this section.
a) x2 x 2  y 2  x 2  2 y 2 y'  0 b) 2 xy dx  x 2  y 2 dy  0

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 31

 Applied Mathematics III

c) e  y dx  2 y  xe  y dy  0

Instructor's role
 Check and give feedback to their answers

1.2.3.1 Inexact equations; Integrating Factors

Definition: An equation of the form M ( x, y)dx  N ( x, y)dy  0 but for which

M N

y x
are known as inexact equations.

Inexact equations can be made exact by multiplying them by some function F ( x, y)  0 . Thus

F x, y M ( x, y)dx  F x, y N ( x, y)dy  0 become an exact equation. Such a multiplier (or a
function) is called an integrating factor. Although an equation of the form
M ( x, y)dx  N ( x, y)dy  0 has integrating factors, but there is no general method of finding them.
We now use the following Theorems (rules) for finding integrating factors

Rules for Finding Integrating Factors

Theorem 1.2.3.1:

1  M N 
When    is a constant, say k , or a function of x
N  y x 

e e
kdx f ( x ) dx
alone, say f (x) , then an integrating factor, F (x) , will be or .

Proof. Exercise.

Example 1: Solve the differential equation

(4 x  3 y 2 )dx  2 xydy 0 .
Solution: Since

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 32

 Applied Mathematics III

M ( x, y)  4 x  3 y 2 , N ( x, y)  2 xy
we obtain
M N
 6y ,  2y
y x
M N
  .
y x
Thus the differential equation is not exact.
But since

1  M N  1
    by  2 y   2 ,
N  y x  2 xy x
an integrating factor is
dx
2 
F ( x)  e x
 x2 .
Thus


x 2 (4 x  3 y 2 )dx  2 xydy  0 
 (4 x 3  3x 2 y 2 ) dx  2 x 3 ydy  0
is an exact differential equation, since

 
y
 
4 x 3  3x 2 y 2  6 x 2 y 
x
2x3 y .  
Now then, since

u( x, y)   (4 x 3  3x 2 y 2 )dx  k ( y)

 x 4  x 3 y 2  k ( y) (*)

u
and  2x3 y
y
we have
2 x 3 y  k ( y)  2 x 3 y .
This implies
k ( y)  0  k ( y)  c1 ,

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 33

 Applied Mathematics III

where c1 is a constant. Thus from ( * ) it follows that

u( x, y)  x 4  x 3 y 2  c1  c2

 x 4  x 3 y 2  c,
where c  c2  c1 is a constant.
Therefore,
x 4  x3 y 2  c ,
is a required solution. 
 
Example 2: Solve 2 y 2  3x dx  2 xydy  0

Solution: Now M x, y   2 y 2  3x and N x, y   2 xy

M N
 4 y and  2y
y x
M N
The given differential equation is not exact, that is  ,
y x
M y  Nx 4y  2y 1
  is a function of x only.
N 2 xy x
1
 xdx
Hence e  e ln x  x is an integrating factor.
 
By multiplying the given equation by x we get 2 xy 2  3x 2 dx  2 x 2 ydy  0.
This is an exact equation as
 
(2 y 2 x  3x 2 )  (2 x 2 y)  2 xy
y x
Applying the method for solving exact differential equation, if we put
U x  2 xy 2  3x 2

we get
U x, y   x 2 y 2  x 3  g  y  ,

g '  y   0 and g  y   c .

The solution of the differential equation is U x, y   x 2 y 2  x 3  c

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 34

 Applied Mathematics III

Theorem 2:
1  N M 
When    is a constant, say k , or a function of y alone, say f ( y) , then an
M  x y 

e e
kdy f ( y ) dy
integrating factor, F ( y) , will be or .
Proof: Exercise

Example 1: Solve the differential equation 2 xydx  3x dy  0 .

2

Solution: Since
M ( x, y)  2 xy , N ( x, y)  3x 2 ,

M N
we have  2 x,  6x .
y x

This implies
M N

y x
Thus the differential equation is not exact.
But since
1  N N  1
   6 x  2 x   2 ,
M  x x  2 xy y
an integrating factor is
dy
2 
F ( y)  e y
 y2 .
Thus

y 2 2 xydx  3x 2 dy  0 
 2 xy 3 dx  3x 2 y 2 dy  0
is an exact differential equation, since

2 xy 3   6 xy 2   3x 2 y 2 . .
y x

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 35

 Applied Mathematics III

Now then, since

u( x, y)   (2 xy 3 )dx  k ( y)

 x 2 y 3  k ( y) ( *)
and
u
 3x 2 y 2 ,
y
we obtain

3x 2 y 2  k ( y)  3x 2 y 2 .
 k ( y)  0  k ( y)  c1 ,
where c1 is a constant. Thus from ( * ) it follows that

u( x, y)  x 2 y 3  c1  c2

 x2 y3  c ,

where c  c2  c1 is a constant. Therefore, x y  c , is a required solution.

2 3

Example 2: Solve the differential equation of the first-order:

 
xydx  2 x 2  3 y 2  20 dy  0

Solution: Now M x, y   xy and N x, y   2 x 2  3 y 2  20

M N
 x and  4x
y x
M N
The given differential equation is not exact, that is  ,
y x
M y  Nx  3x

N 2 x  3 y 2  20
2

M y  Nx
leads us nowhere, as is a function of both x and y. However,
N
Nx  M y 4x  x 3
  is a function of y only. Hence
M xy y
dy
3
 e 3 ln y  e ln y  y 3 is an integrating factor.
3
y
e

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 36

 Applied Mathematics III

After multiplying the given differential equation by y3 we obtain

 
xy 4 dx  2 x 2 y 3  3 y 5  20 y 3 dy  0
This is an exact differentiation equation. Applying the method of the previous section we get
1 2 4 3 6
x y  y  5y4  C
2 6

Quick Check Group Activity 1.2.3

1. Find integrating factors and solve the following differential equations.

a) ( y  1)dx  xdy  0 c) ydx  xdy  0, y(1)  2

b) y cos xdx  2 sin xdy  0 d) 3x

2
ydx  2 x 3 dy  0
1
e) 3 y dx  xdy  0, y(1) 
2

2
3. Verify that y, xy 3 and x 2 y 5 are integrating factors of ydx  2 xdy  0 and solve.

Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity and Give feedback to their answers
1.2.4 Linear First Order Differential Equations; Reduction to Linear Form
1.2.4.1 Linear First Order Differential Equations

Definition:
A first order differential equation is said to be linear if it can be written in the form

y   f ( x ) y  r ( x) ,

Theorem: The general solution of the linear first order differential equation:
y   f ( x) y  r ( x) ,

where f and r are continuous functions, is given by

where f and r are continuous functions.
 f ( x ) dx   f ( x ) dxr ( x)dx  c
y ( x)  e   e 

Proof: Let us write the equation

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 37

 Applied Mathematics III

y   f ( x) y  r ( x ) ( *)
in the form
 f ( x) y  r ( x)dx  dy  0 ( **)
and, let
M ( x, y)  f ( x) y( x)  r ( x), N ( x, y)  1 ,
M N
then  f ( x),  0.
y x

This implies, for f ( x)  0 ,

M N

y x

Thus, the differential equation ( **) and hence ( * ) is not exact.

But since
1  M N 
    f ( x) ,
N  y x 
the integrating factor is

F ( x)  e 
f ( x ) dx
.

Now, let F ( x)  eh( x ) ,where h( x)   f ( x)dx. Multiplying ( * ) by this factor we obtain

e h ( x ) ( y   f ( x) y )  e h ( x ) r ( x) .
Since h( x)  f ( x) , this may be writer as

d
dx
 
ye h  e h r .

Integrating both sides we have

y( x)e h ( x )   e h ( x ) r ( x)dx  c ,

that is,
y ( x)  e  h ( x ) e h( x)

r ( x)dx  c , h( x)   f ( x)dx . Or
 f ( x ) dx   f ( x ) dx r ( x)dx  c 
y ( x)  e  e .
 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 38

 Applied Mathematics III

This is the general solution of (*). 

Definition 2.3.2: I ( x)  e 
f ( x ) dx
is called the integrating factor of the standard form of a linear
differential equation .
Procedure of Solution :
Step 1:Put the equation in the standard form y  f ( x) y  r ( x) if it is not given in this form.

Step 2:Identify f x  and compute the integrating factor I ( x)  e 

f ( x ) dx

Step 3:Multiply the standard form by I x  .

Step 4:The solution is y.I x    f x .I x dx  c

Example 1 : Solve the following differential equations:

xy   y  4  0
Solution: Since
xy  y  4 0
can be rewritten as
1 4
y  y ,
x x
then
1 4
f ( x)  r ( x)  .
x, x
Thus using the result of the above theorem, we see that

 x   x 4 
dx dx

y ( x)  e  e  dx  c 
  x  

 ln x  1  
 
ln x
e  4 e  dx  c 
  x 


1
x

 4 dx  c 
c
 4  .
x
Therefore,

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 39

 Applied Mathematics III

c
y ( x)   4 
x
is the general solution. 

 3x  1 y  e 3 x
dy
Example 2.: Find the general solution of : x
dx
Solution: Standard form is
dy  1 e 3 x
 3   y 
dx  x x

1 e 3 x
f ( x)  3  r ( x) 
x, x
 1
  3 dx
Integrating factor = (x)  e 
P(x)dx
= e  x   xe 3x

e 3 x
y.xe 3 x    xe 3 x dx  c  x  c
x
 c
or y ( x)  e 3 x 1   for 0<x<.
 x
Quick Check Class Exercises 1.2.7
1. Solve the following linear differential equations given below
dy dy
a)  2 y  8e x c) x y1
dx dx x
dy
b) x3  3x 2 y  cos x
dx
2. Solve the following initial value problem
dy dy
a)  5 y  20, y (0)  2 b) x   y  ex, y(1)  2
dx dx

Instructor's role
 Check and give feedback to their answers

Group Activity 1.4.1:

1. Solve the following differential equations.
a) y tan x  y b) y   y cos x  cos x

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 40

 Applied Mathematics III

c) y  3 y  e 2 x  6 g)
dy
 y  sinh x
dx
xy  2 y  2e x
2
d) dr
h)  r sec θ  cos θ
dθ
y
e) y   2  x 2e x dy
x i)  ( y tan x  cos x)  0
dx
dy 3x 2 y
f) 
dx 1  x 3

2. Solve the following initial value problems.

a) y  y tan x  2 x cos x, y(0)   1

c) xy  (1  x) y , y(2)  6e 2
d) xy  4 y  8x 4 , y(1)  2

  2
e) xy   y  sin x, y  
2 
3. Find the curve having the given slope y  and passing through the

given point ( x0 , y0 ) if

x 1 x
a) y  , (0,2) b) y   , (1, 0)
4  4y 1 y
Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers

1.2.4.2 Reduction to Linear Form (Bernoulli Equation)

Definition:
The differential equation of the form
Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 41
y   p ( x) y  g ( x) y a ,

where a is any real number, is called the Bernoulli equation.

 Applied Mathematics III

Remark: For a  0 and a  1 the equation is linear, and otherwise it is non linear.
For a  0,1 if we set

u ( x)   y ( x) 1  a
the equation will be reduced to the linear form
u   (1 a) p( x) u  (1 a) g ( x) ,
which will be easily solved.
Example 1: Solve the Bernoulli equation
y  y  y 2 .
Solution: Since a  2 , by setting
u  y1 2  y 1 , (*)

we obtain

u   y  2 y .
Thus

( y  y  y 2 ) y  2 .
From this it follows that
yy 2  y 1  1 ,
or
 yy 2  y 1   1 ,
by multiplying both sides by  1 .

Replacing u  y we obtain a linear first order differential equation

1

u  u   1 .
With solution

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 42

 Applied Mathematics III

u ( x)  e    e  dx  c 
dx  dx

 
 1 ce x .
Using (*) we obtain

1 1
y ( x)   .
u ( x) 1  ce x
This is the required general solution. 
dy 1
Example 2: Solve the bernouli equations.  y  3y3
dx x
Solution: Let v  y1n  y 2 (n=3)
dv dy
 2 y 3
dx dx
dy 1 1 dv
or . 3 
dx y 2 dx
Substituting these values into the given differential equation, we get
1 dv 1
  v3
2 dx x
dv 2
or  v  6
dx x
This equation is of the standard form, (2.7) and so the method of Section 2.3 is applicable.

Integrating factor  ( x)  e 
P ( x ) dx

2
where P( x)   . Therefore I x   x 2
x
Solution is given by
v.x 2    6 x 2 dx  c

v.x2  6x1  c

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 43

 Applied Mathematics III

Since v  y 2 we get y 2  6 x  cx 2 or y   1
6xcx2

Quick Check Group Activity 1.4.2 Solve the following Bernoulli equations
1
a) y  y  xy 2 c) 3 y   y  (1 2 x) y 4
x
b) y   xy  xy 1 d) y  x 3 y 2  xy
Instructor's role
 Check and give feedback to their answers
1.2.5 Application Involving First Order Differential Equations
First order differential equations are often used to model rate processes. For example,
1.2.5.1.(Newton's law of Cooling).
This law states that the rate of temperature of a body is proportional to the difference between
the temperature of the body and that of the surrounding medium.

Suppose a body whose temperature is initially 𝜃1 ℃ is allowed to cool in air which is maintained
at a constant temperature of 𝜃2 ℃. It is required to find the temperature of the body as a function
of time 𝑡.
Let the temperature of the body be T  C at time t. Then by Newton's law, we can write

 k T   2 
dT
(*)
dt
where 𝑘 is the constant of proportionality. The initial condition is given by T  1 at 𝑡 = 0.
Solving the equation, we obtain
 kt  logT   2   log C (**)
Initially we have

T 0  1

Substituting this in (**) and simplifying we get

T   2  1   2 e  kt ,

which is the required expression for 𝑇 interms of 𝑡.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 44

 Applied Mathematics III

Example 1. A metal ball is heated to a temperature of 100 C an d at time 𝑡 = 0 it is placed in

water which is maintained at 40 C . If the temperature of the ball is reduced to 60 C in 4
minutes,find the time at which the temperature of the ball is 50 C .
Let the temperature of the body be T  C at time t min. Then the differential equation is given by

 k T  40
dT
(i)
dt

Integration gives;

 kt  logT   2   log C (ii)

At 𝑡 = 0, 𝑇 = 100. This gives

log C   log 60,

and hence (ii) becomes

 kt  log( T 60
40
) (iii)

But T=60 at t=4. Substituting these values in (iii), we obtain

1
k  log 3
4
Hence equation (iii) gives;
t
log 3  log( T 60
 40
)
4
When T=50, we obtain

4 log 6
t  6.5 minutes
log 3
Example 2 Assuming that a building is a homogeneous point source without wind in its
homogeneous fluid surroundings of constant temperature. The internal temperature change of a
building can be calculated using Newton’s Law of Cooling.
The simplest form is the first order differential equation
dT
 k (T  Ts)
dt

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 45

 Applied Mathematics III

where T(t) is the internal temperature of the building, S is the temperature of the surroundings,
and k is a growth constant. If we assume the initial condition T(t0)=T0, after integration and
substitution, we arrive at
T (t )  TS  (T0  TS )e  kt (1)
Therefore,
T ( t f ) TS  k (t f t 0 )
T0 TS
e , (2)

implying
T ( t f ) TS
k (t f  t0 )   ln( T0 TS
). (3)

Thus making it possible to find k with a second condition, 𝑇(𝑡𝑓) = 𝑇𝑧. Through substitution of
this second condition in (3) and after rearranging, we have

k  ( (t f 1t 0 ) )  ln( TT0z TTsS ). (4)

Now we can use the constant k to find the temperature at any given time, by inserting it into (2).
For example, a building with a constant surrounding temperature of 85°F has an initial condition,
T(0)= 75°F. Inserting this information into 2-4, we get
T ( t f ) 85  k ( t f 0 )
T (t )  85  (75  85)e  kt ,  7585 e
T ( t f ) 85
k (t f  0)   ln( 7585 )

Upon adding the second condition, T(5)=70°F, we have

k (5  0)   ln( 7085
7585 ),

k  ( 15 )  ln( 32 )
k  .08109
Then by inserting the k value into (1),
T (t )  85  (75  85)e.08109t
1.2.5.2.(Radioactive decay)
Radioactive decay where the content of radioactive nuclei is denoted by the symbol, n, is
modeled by the following first-order differential equation.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 46

 Applied Mathematics III

dn
 kn
dt
dn
For a positive constant, k, this equation tells us that the rate is negative and proportional to
dt
the amount of radioactive nuclei, n, present. If the initial content at t = 0 is n0 we can multiply
this equation by dt/n to obtain the following form that can be integrated directly.

dn
n
dn
t
n
n n 0 kdt  ln n0   kt  n  n0e
kt
 kdt    k
n 0

The half-life for radioactive decay, t1/2, is defined as the time required for the initial radioactivity,
n0, to decrease to half its original value. Equation shows n  n0 e  kt us how to compute this half
life.
 n0 
  1 ln( 2)
ln  2   ln     ln( 2)  kt1 / 2  t1 / 2 
 n0  2 k
 
With equation , we can rewrite the final version of equation to introduce the half life.
t

n  n0 e t1 / 2 ln(2 )

Though only two applications involving first order equations are discussed in this module there
various laws governed by first order equations. For instance: Kirchhoff's electromotive force law
governing flow of electricity, Hooke's law which relates the tension and extension in elastic
string, the motion for a body falling in a Resisting medium are all governed by First order
equations.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 47

 Applied Mathematics III

1.3 Ordinary Linear Differential Equations of the Second order

Definition: The general linear second order differential equation is of the form

R( x) y" P( x) y'Q( x) y  S ( x) ,
where P, Q, R, S are continuous functions. On any interval where Rx   0 ,we can divide the
equation by Rx  and obtain the special linear differential equation

y  f ( x) y  g ( x) y  r ( x) , (1 )
where f , g and r are continuous functions.

If r ( x)  0, then equation (1) becomes

y  f ( x) y  g ( x) y  0 ( 2)
and is said to be homogeneous differential equation.
If r ( x)  0, then equation (1) is said to be non-homogeneous differential equation.

For example, y  4 y  e  x sin x is a non homogeneous linear differential equation, where as

(1 x 2 ) y  2 xy  6 y  0 and y  4 y  0 are homogeneous linear differential equations
Any differential equation of the second order which cannot be written in the form ( 1 ) is said to
be nonlinear.

For example, yy  y  0 and y  y  0 are non-linear differential equations.

Theorem 2: (Existence and Uniqueness Theorem for Linear Initial Value Problems)
Let f , g and r be continuous functions on some open interval I, and let k 0 , k1 be two

given constants. Then there exists a unique function y (x) that satisfies the linear differential
equation

y  f ( x) y  g ( x) y  r ( x) (3)
on I and the two initial conditions

y( x0 )  k0 , y( x0 )  k1 (4)

For some value x0 in I.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 48

 Applied Mathematics III

Remark1: The conditions given in (4) both involve evaluations of the unknown function y

and its derivatives at the same point x0 . This is a crucial requirement for the existence and
uniqueness of a solution.
There is another type of problem involving the differential equations (3) in which conditions at
more than one point are given. For example, we might specify y( x1 )  d1 and y( x2 )  d 2 .
Conditions of this sort are called boundary conditions, and a differential equation together with a
set of boundary conditions is called a boundary value problem.
It is important to note that the existence and uniqueness of a solution guaranteed by Theorem1
for initial value problems, doesn’t hold for boundary value problems.
Example: Observe that the boundary value problem
y  y  0, y(0)  y( )  0 ,
has infinitely many solutions:
y  c sin x , for any constant c .
To check, note that sin 0  sin   0, so the boundary conditions are satisfied, and
(c sin x)  c sin x   c sin x  c sin x  0 .
Where as the boundary value problem

y  y  0, y(0)  0, y( )  1

has no solution. On the other hand, the initial value problem
y   y  0, y(0)  0, y (0)  1
has the unique solution y  sin x ; the initial conditions hold since
sin 0  0, (sin x)  cos x, and cos 0  1 ,and (sin x)  sin x   sin x  sin x  0 .
Remark 2: The requirement that the functions f ( x), g ( x) and r (x) in (3) be continuous is
also an essential requirement for the existence of a unique solution, as the following example
demonstrates.
Example: Substitution shows that the function
y  cx 3  x
is a solution (for any constant c ) of the initial value problem

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 49

 Applied Mathematics III

x 2 y  3xy  3 y  0, y(0)  0, y(0)  1 ,

that is, the problem does not have a unique solution, it has infinitely many solutions. Theorem1
does not apply here, because for the differential equation

x 2 y  3xy  3 y  0, y(0)  0, y(0)  1 ,

that is, for the differential equation
3 3
y  y  2 y  0, y(0)  0, y(0)  1 ,
x x
3 3
f ( x)  and g ( x)  2
x x
are not continuous in an interval containing x0  0 .
Note that an initial value problem for a differential equation of the second order consists of the

equation and two initial conditions

y( x0 )  k1 , y( x0 )  k2 .

1.3. 1 Homogenous Equations

1.3.1.1 Theory of Solutions y  f ( x) y  g ( x) y  0

Theorem 1: (Superposition Principle)

Let y1 and y 2 be solutions of homogeneous linear differential equation 2  on an

open interval I. Then any linear combination y  c1 y1  c2 y2 ,where c1 , c2 are

arbitrary constants, is also a solution of 2  on I. In particular, for such an equation,

sums and constant multiples of solutions are again solutions

Proof: Let y1 and y 2 be solutions of ( 2 ) on I. Then by substituting

y  c1 y1  c2 y2
and its derivatives,
y  c1 y1  c2 y2 , y  c2 y2  c2 y2

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 50

 Applied Mathematics III

into ( 2 ) , we get

y  f ( x) y  g ( x) y  (c1 y1  c2 y2 )  f ( x)(c1 y1  c2 y2 )  g ( x)(c1 y1  c2 y2 )

 c1 y1  c2 y2  f ( x) (c1 y1  c2 y2 )  g ( x) (c1 y1  c2 y2 )

 c1 ( y1  f ( x) y1  g ( x) y1 )  c2 ( y2  f ( x) y2  g ( x) y2 )

 0 , since by assumption y1 , y2 are solutions of ( 2 ) .

This shows that y  c1 y1  c2 y2 is a solution of ( 2 ) on I. 

Remark: The above theorem doesn’t hold for non-homogeneous linear differential
equations or for nonlinear equations.
For example, substitution shows that:
1. The functions y1 1  cos x and y2  1  sin x are solutions of the non-homogeneous

linear differential equation y  y  1, but 2(1  cos x), and (1  cos x)  (1  sin x), are not
solutions of this differential equation.
2. y1  x 2 and y2 1 are solutions of the nonlinear differential equation. yy  xy  0 , but

 x 2 and x 2  1 are not solutions of this differential equation.

The point to taking linear combinations c1 y1  c2 y2 is to obtain more solutions from just two

solutions of equation 2  . However, if y 2 is a constant multiple of y1

c1 y1  c2 y2  c1 y1  c2 y2  c1  kc2 y1

just another constant multiple of y1 . In this event, y 2 is superfluous, providing nothing we did
not know from. This leads us to distinguish the case in which one solution is a constant multiple
of another from the case in which the two solutions are not multiples of each other.
Definition: Linear Dependence and Independence
Two functions, y1 x  and y2 x  , are linearly independent on an interval x0 , x1  ,whenever the

relation c1 y1 ( x)  c2 y2 ( x)  0 for all x in the interval implies that c1  c2  0 . Otherwise,

they are linearly dependent.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 51

 Applied Mathematics III

If two functions y1 and y2 are linearly independent c1 y1 ( x)  c2 y2 ( x)  0 (where c1 and c2 are

not both zero), we may suppose that c1  0. Then

y1 x   y2 x   0 or y1 x    y2 x   Cy2 x 
c2 c2
c1 c1
Therefore: Two functions are linearly dependent on the interval if and only if one of the
functions is a constant multiple of the other for all x in the interval .
Examples:
1) y1  sin x and y2  cos x are linearly independent solutions of y  y  0 , over the real line.
Neither of these functions is a constant multiple of the other. Indeed, if cos x  k sin x for all x ,

  2   2
then in particular cos    k sin   k so k must be 0. But then cos x  sin x for all
4 2 4 2
x , a clear absurdity(for example x  0) . Now we know from theorem 1 above that
a sin x  b cos x

is a solution for any numbers 𝑎 and 𝑏.Since the solutions are linearly independent, this linear
combination provides an infinity of solutions, instead of just constant multiple of one we already
know.
2x 3x
2) since y1  e 5 x  consant , y1  e and y2  e are linearly independent solution of the
y2

differential equation: y  5 y  6 y  0 .

There is a simple test whether two solutions are linearly independent on an interval.

Definition (Wronskian):
Let y1 ( x) and y 2 ( x) be any two solutions to the differential equation
y   f ( x) y   g ( x) y  0 .
The Wronkian of y1 and y1 is defined as

y1 ( x) y 2 ( x)
W ( y1 , y2 )( x)   y1 ( x) y2 ( x)  y1 ( x) y2 ( x) .
y1 ( x) y2 ( x)

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 52

 Applied Mathematics III

Wronskian Test:

Theorem3. (Wronkian Test)

Let y1 and y 2 be solutions of

y  f ( x) y  g ( x) y  0
for x in an open interval I. Let f and g be continues on I. Then

1. Either W ( y1 , y 2 )( x)  0 for all x in I or W ( y1 , y 2 )( x)  0 for all

x in I.
2. y1 and y 2 are linearly independent on I if and only if
W ( y1 , y 2 )( x)  0 for all x in I.
Proof: Exercise.

Example:1 Show that y1  sin x and y2  cos x are linearly independent solutions of y  y  0

Solution: The student may verify that y1  sin x and y2  cos x are solutions of y  y  0 .

Now

sin x cos x
W  y1 , y2 x     sin 2 x  cos 2 x  1  0
cos x  sin x

Thus the solutions are linearly independent.

Example 2. Show that y1  e 3 x and y2  xe 3 x are linearly independent

Solution: Using Wronskian test

W  y1 , y2 x  
e3x xe 3 x
     
 e 3 x e 3 x  3xe 3 x  3e 3 x xe 3 x  e 3 x e 3 x  e 6 x
3e 3x
e  3xe
3x 3x

Since 𝑒 6𝑥 cannot be zero, the Wronskian for these two solutions is non-zero, showing the
linearly independence of the two solutions

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 53

 Applied Mathematics III

Theorem 4:
Let y 1 ( x) and y 2 ( x) be linearly independent solutions of

y  f ( x) y  g ( x) y  0 (*)

and let y3 ( x) be another solution. Then there exist unique constant c1 and c 2

such that

y3 ( x)  c1 y1 ( x)  c2 y2 ( x) .

In other words, any solution of (*) can be written as a linear combination of two
given linearly independent solutions of (*) . Hence (*) does not have singular solution.
Proof: Exercise

Definition: (Fundamental set of Solutions)

Any two linearly independent solutions y1 x  and y2 x  of the homogeneous linear second-order
differential equation (1) on an interval I is said to form basis or a fundamental set of solutions on
the interval.
Examples:
1. y1  e x and y2  3e x are solutions of the differential equation y  y  0 . Now Since

y1
 3  consant y1  e x and y2  3e x are linearly dependent, they don’t form basis .
y2
 2x  3x
2. y1  e and y2  e are solutions of the differential equation y  5 y  6 y  0 .Now

y1
since  e 5 x  consant , y1  e  2 x and y2  e  3 x are linearly independent, thus form a
y2
basis of solutions of the equation for all x .

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 54

 Applied Mathematics III

Definition: (General Solution)

A solution of a differential equation

y   f ( x) y   g ( x) y  0 (1 )

on some open interval I is called a general solution if it is expressed in the form

y( x)  c1 y1 ( x)  c2 y2 ( x),

where c1 , and c2 are constants and y1 and y 2 form a basis (or a fundamental system)

of solutions of equation (1) on I.

Definition: Equation y( x)  c1 y1 ( x)  c2 y2 ( x) , which provides the solution for the

homogeneous equation is called the complementary function.

Example 1: The student may verify that y1  e 2 x and y2  e  3 x solution of the differential
equation:
y  5 y  6 y  0
Solution : Using Wronskian test

e 2 x e 3 x
W  y1 , y2 x   2 x 3 x
   
 e 2 x .  3e 3 x  e 3 x  2e 2 x  e 5 x  0
 2e  3e

y1
Now since y1  e 2 x and y2  e  3 x are linearly independent,  e 5 x  consant , they
y2
form a basis of solutions of the equation for all x and hence a general solution is

y( x)  c1 y1 ( x)  c2 y2 ( x)  c1e 2 x  c2e3 x
[Example 2] The student may verify that y1  sin x and y1  cos x are solutions of y  y  0
Since
sin x cos x
W  y1 , y2 x     sin 2 x  cos 2 x  1  0
cos x  sin x

Thus the solutions are linearly independent. The general solution is then

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 55

 Applied Mathematics III

y( x)  c1 y1 ( x)  c2 y2 ( x)  c1 sin x  c2 cos x

1.3.1.2 Reduction of order: How to Obtain a Basis if One Solution is Known

If y1 is a nonzero solution of the equation

y  f ( x) y  g ( x) y  0

we want to seek another solution y2 such that y1 and y2 are linearly independent

Theorem2: If one solution, y1 of the differential equation

y  f ( x) y  g ( x) y  0 (*)
is known on some interval I, the second linearly independent solution y 2 of the
equation can be obtained by the method of reduction of order and is given by

e 
 f ( x ) dx

y2  uy1  y1 ( x)  2 dx . So that y1 , y 2 are the basis.

y1 ( x)

Proof: Suppose y1 is a known nonzero solution of (*) , to get y 2 , since y1 and y 2 are linearly

independent, y2  ky1 for any constant k . So

y2
 u ( x)
y1

must be a non-constant function of x , and y2  uy1 must satisfy (*) . Now, substituting

y2  u y1 ,

and its derivatives,

y2  uy1  uy1 and y2  uy1  2uy1  uy1

into (*) gives

uy1  2uy1  uy1  f ( x)(uy1  uy1 )  g ( x) uy1  0 .

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 56

 Applied Mathematics III

Collecting terms in u, u and u , we have

uy1  u(2 y1  f y1 )  u( y1  f y1  g y1 )  0

But since y1 is a solution of (*) , the expression in the last parentheses is zero.

Thus

uy1  u(2 y1  f y1 )  0 .

Dividing by uy1 , we can rewrite this equation in the form

u y
 2 1  f .
u y1

Setting z  u , z  u , we obtain

z' y
 2 1  f
z y1

dz dy
  z
  2 1 
y1  f ( x)dx

 ln z   2 ln y1   f ( x)dx .

Exponentiating and simplifying, we obtain

e
1  f ( x ) dx
z ( x)  2
y ( x)
1

But since z  u , we have

1  f ( x ) dx
u( x)  e .
y12 ( x)

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 57

 Applied Mathematics III

That is,

e 
 f ( x ) dx

u ( x)   2 dx .
y1 ( x)

Hence the desired solution is

e 
 f ( x ) dx

y2  uy1  y1 ( x)  2 dx 
y1 ( x)
Example 1: Find general solutions for the differential equation

x 2 y  xy  y  0, x0,

if one solution is y1  x .

Solution: To use the formula, we need to write the differential equation in the following standard
form:

Since x y  xy  y  0 Can be written as

2

1 1
y  y  2 y  0 ,
x x
and y1  x then
1
  x dx
e
y2  x  dx  x ln x , for x  0 .
x2
Further more, since

y1  x , y2  x ln x, x0
are linearly independent, they are the basis for the given differential equation and the general
solution is
y( x)  c1 x  c2 x ln x, x  0 
Example 2: Find general solutions for the differential equation

x 2 y  3xy  4 y  0, x  0 ,
if one solution is y1  x 2 .
Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 58
 Applied Mathematics III

Solution: To use the formula, we need to write the differential equation in the following standard
form:

Since x y  3xy  4 y  0 Can be written as

2

3 4
y  y  2 y  0 ,
x x
and y1  x 2 then
3

  dx
x
e
y2  x 2  dx  x ln x , for x  0 .
x4
Further more, since

y1  x 2 , y2  x ln x, x0
are linearly independent, they are the basis for the given differential equation and the general
solution is

y( x)  c1 x 2  c2 x ln x, x  0 
Quick Check Group Class Activity Find the second linearly independent solution to
a) x y  xy  2 y  0, x  0 if one solution is y1  x sin(ln x)
2

b) x y  3xy  5 y  0, x  0 if one solution is y1  x cos(ln x)

2 2

c) (1  x ) y  2 xy  2 y  0, x  0 if one solution is y1  x

2

d) (1  2 x  x ) y  2(1  x) y  2 y  0, x  0 if one solution is y1  x

2

Discuss your solutions to questions above in small groups and see whether or not they match
with the solution suggested by your instructor.
Instructor's role
 Check and give feedback to their answers
Group Activity 1.3
Discuss your solutions to questions above in small groups and see whether or not they match
with the solution suggested by your instructor.
1. Solve the following differential equations by the method of reduction of order if one
solution, y1 ,is given

a) x 2 y  5xy  9 y  0, y1  x3

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 59

 Applied Mathematics III

3 4
b) y  y  2 y  0, y1  x 2 , for x  0
x x

c)
y  4 y  0, y1  cos(2 x)
d) (2 x 2  1) y  4 xy  4 y  0, y1  x , for x  0
2. Find general solution of
2 2 ln x
y  y  2 y  , x 0
x x x
if the two solutions y1  x and y1  x 2 for the corresponding homogeneous equations are
given.
3. Application of reducible equations (Motion)
a) A small body moves on straight line so that the product of its velocity and acceleration is
constant, say, 1m 2 / sec3 .If at t  0 the body’s distance from the origin is 2 meters and its
velocity is 2m / sec , what are the distance and velocity at t  6 sec?
b) What happens in the above problem (prob.2a), if the acceleration equals the velocity (the
other data being as before)? Will the distance at t  6 be larger?
5. Find the curve through the origin in the xy -plan which satisfies y  2 y and whose
tangent at the origin has slope 1 .
Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers

1.3.1.3 Homogeneous Linear Differential Equations with Constant Coefficients

The general linear homogenous equation discussed is usually difficult to solve and requires
special techniques. However an important and special case occurs when the coefficients are
constants, the equation being called a constant coefficient equation.
Consider a constant coefficient second order homogeneous linear differential equation of the
form

y  by  cy  0 , (1)

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 60

 Applied Mathematics III

where coefficients b and c are constant.

To solve ( 1 ) we remember from Unit One that a first - order linear differential equation
k x
y  ky  0 with constant coefficient k has an exponential function as a solution, y  e .
This gives us the idea to try as a solution of ( 1 ) the function

y  ex . (2)
Substituting ( 2 ) and its derivatives

y  ex and y  2ex

into equation ( 1 ) , we obtain

2  b  c  0 . (3 )
Hence ( 2 ) is a solution of ( 1 ) , if  is a solution of the quadratic equation (3 ) .

Definition: The equation 2  b  c  0 is called the characteristic equation (or auxiliary)

equation) of ( 1 ) .

The characteristic equation 2  b  c  0 of y  by  cy  0 is quadratic. Its roots are

1 
1
2
 
 b  b 2  4c , 2 
1
2

 b  b 2  4c . 
Our derivation shows that the functions

y1  e1 x and y 2  e 2 x
are solutions of ( 1 ) .

But since b and c are real constants, there are three possible situations for the roots of 1 and 2

of the characteristic equation depending on the sign of the discriminant b 2  4c :

Two distinct real roots if b  4c  0 ,

2
Case I.

Two complex conjugate roots if b  4c  0 ,

2
Case II

if b  4c  0 .
2
Case III A real double root

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 61

 Applied Mathematics III

Case I: Two Real Roots

In this case, since b 2  4c  0 , then 1 and 2 are distinct real roots and hence

y1  e1x and y2  e2 x

are the basis and hence the corresponding general solution is

y( x)  c1e1x  c2e2 x
where c1 and c 2 are constants. 

Example 1: Solve the differential equation y  y  2 y  0, y(0)  4, y(0) 1

Solution: The characteristic equation is
 2    2  0,
and hence its roots are

1 
1
2

1  
9  1, 2 
1
2
1  9   2
So that the general solution is

y( x)  c1e2 x  c2e x .
But since
y(0)  4, y(0)  1
and
y( x)   2c1e 2 x  c2e x ,
then
y(0)  c1  c2  4, y(0)   2c1  c2  1 .
From these it follows that
c1  1, c2  3
Therefore, the particular solution is
y( x)  e 2 x  3e x 
The student should check the answer.
Case II: Two Complex Roots
In this case, since b 2  4c  0 , then the roots are complex conjugates,

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 62

 Applied Mathematics III

1  p  iq, 2  p  iq ,
where p and q are real constants and q  0 . But then since

s1  e( p  iq ) x  e px eiqx  e px cos qx  ie px sin qx,

and

s2  e( p  iq ) x  e px e qx  e px cos qx  ie px sin qx
then,
1 1
y1  ( s1  s2 )  e px cos qx , y2  ( s1  s2 )  e px sin qx .
2 2i
which are linearly independent solutions and therefore are the basis. Hence the corresponding
general solution is

y( x)  e px ( A cos qx  B sin qx) ,

where A and B are constants. 
Example 1: Solve the differential equation
y  2 y  10 y  0, y(0)  4, y(0)  1
Solution: The characteristic equation is
 2  2  10  0
and its roots are

1 
1
2
2  
 36  1  3i , 2 
1
2

2 
 36  1  3i

with p  1, q  3 so that the general solution is

y( x)  e x ( A cos 3x  B sin 3x) .

But since
y(0)  4, y(0)  1
and
y( x)  A(e x cos 3x  3e x sin 3x)  B(e x sin 3x  3e x cos 3x),
we have
y(0)  A  4 , y(0)  A  3B  1

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 63

 Applied Mathematics III

Hence
A  4, B   1 .
Therefore, the particular solution is
y( x )  e x (4 cos 3x  sin 3x) . 
Case III: A Real Double Root
In this case, since b 2  4c  0, then we get only one root
b
 
2
b
x
hence only one solution y1  e 2
.

To obtain a second independent solution y 2 (needed for a base), we use the reduction method
(discussed in Theorem 2, Sec. 2.2 ). Since one solution is known,

e bx e bx
b b
 x  x
y2  y1 ( x)  2 dx  e 2
  bx
dx  e 2  dx
y1 ( x) e
b
 x
 xe 2
.
Thus
b b
x x
y1  e 2
, y 2  xe 2

are linearly independent solutions and therefore are the basis. Hence the corresponding general
solution is
b
 x
y( x)  (c1  c2 x)e 2
. 
Example: Solve the following differential equation
a) y  8 y  16 y  0
b) y  4 y  4 y  0, y(0)  3, y(0)  1
Solution:
a) The characteristic equation is

2  8  16  0
Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 64
 Applied Mathematics III

and the root is

1
  (8  0 )   4
2
so that the general solution is

y( x)  (c1  c2 x)e  4 x . 
b) The characteristic equation is

2  4  4  0
and the root is
1
  (4  0 )  2
2
so that the general solution is

y( x)  (c1  c2 x)e2 x .
But from the initial conditions, y(0)  3, y(0)  1 we obtain

c1  3, c2   5 .
Hence the particular solution is

y( x)  (3  5x)e2 x . 
Summary of cases I  III
Roots of Basis of General solution of
Case
 2  b  c  0 y  by  cy  0 y  by  cy  0
Distinct real
I
1 , 2 e1x , e2 x y  c1e1 x  c2e2 x

Complex conjugate
e px cos qx ,
II 1  p  iq y  e px ( A cos qx  B sin qx)
px
2  p  iq
e sin qx

Real Double root

III

b e x , xe x y  (c1  c2 x)ex
2

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 65

 Applied Mathematics III

Quick Check Exercises: Write the auxiliary equations for the following differential equations
and hence solve the equations:
(d) y' '2 y' y  0, y(0)  y' (0)  1.
(a) y //  3 y /  2 y  0

d2y d y
(b) 2
6  9y  0
dx dx
(c) y' '4 y  0, y(0)  1, y' (0)  0.
Instructor's role
 Check and give feedback to their answers
Group Activity 2.3
1. Find general solutions for the following differential equations
a) y  4 y  3 y  0 d) y  16 y  0

b) y  4 y  0, e) y  4 y  0
c) y  2y  10y  0 f) y - 6 y  10y  0
2. Solve the following initial value problems
a) y  2 y  y  0, y(0)  1, y(0)  0

b) y  14 y  49 y  0, y(0)  2 , y(0)   6 2


  
c) y  2 y  2 y  0, y ( )  0, y ( )   2e 2
2 2
3. A particle moves on a straight line so that its acceleration is equal to three times its
3
velocity. At t  0 its displacement from the origin is 1 ft and its velocity is ft / sec . Find the
2
time when the displacement is 10 ft .
4. (Undamped system, harmonic oscillations)
If an iron ball weight w  89 nt stretches a spring 10 cm , how many cycles per minute will
this mass spring system executes? What will the motion be if we pull down the weight an
additional 15 cm ?
5. The three cases of damped motion (Over damping, critical damping and under damping)
How does the motion in exercise 4, above, change if the system has damping given by

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 66

 Applied Mathematics III

a) c  200 kg / sec b) c  180kg / sec c) c 100 kg / sec ?

where c is the constant of damping.
Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers

1.3.1.4 Euler-Cauchy Equation

Definition:
A differential equation of the form

x 2 y  bxy  cy  0 ,
(1 )
where b , c are constants, is called the Euler-Cauchy equation.
To solve the Euler-Cauchy equation, we let

y  xm ( 2)
Substituting y and its derivatives,
y  mxm 1 , y  m(m  1) x m 2
in to (1), we find
x 2 m(m  1) x m2  bxmx m1  cx m

 x m (m(m  1)  x m (bm)  cx m
 x m (m2  (b  1)m  c)  0 .
For x  0 , we obtain the characteristic (or auxiliary) equation

m2  (b  1)m  c  0 ( 3)
for determining m in ( 2 ) .
The Three Cases of Solutions
Case I: Distinct Real Roots.
If the auxiliary equation ( 3 ) has distinct real roots m1 , m2 we get the basis of solutions

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 67

 Applied Mathematics III

y1  x m 1 , y2  x m 2
and a corresponding general solution of (1 ) is

y( x)  c1 x m1  c2 x m2 ,
for all x for which y1 and y 2 are defined. 
Example: Solve the Euler - Cauchy equation
x 2 y  4 xy  6 y  0
Solution: The auxiliary equation is

m2  5m  6  0
The roots are
m1  3, m2  2 .
This gives
y1  x 3 , y2  x 2 ,
and the general solution is
y( x)  c1 x 3  c2 x 2 . 
Case II: Complex Conjugate Roots
If the roots of ( 3 ) are complex, they are conjugate, say,

m1    iv, m2    iv
and this gives

x m 1  x   iv  x  xiv  x  eiv ln x
 x  cos(v ln x)  i sin(v ln x)

x m 2  x   iv  x  x iv  x  e iv ln x

 x  cos(v ln x)  i sin (v ln x.

But Since


1 m1
2

x  x m2  x  cos(v ln x)

and

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 68

 Applied Mathematics III


1 m1
2i

x  x m2  x  sin(v ln x) ,

then we obtain the real solutions

x  cos(v ln x) and x  sin(v ln x)

and the corresponding general solution for all positive x is
y( x)  x  A cos(v ln x)  B sin(v ln x) 
Example: Solve the differential equation
x 2 y  7 xy  13 y  0
Solution: The auxiliary equation is
m2  6m 13  0.
The roots are complex conjugate,

m1   3  9  13   3  2i,

m2   3  9  13   3  2i,
and the corresponding general solution for all positive x is

y( x)  x  3 A cos(2 ln x)  B sin(2 ln x) 

Case III: Double Root
If ( 3 ) has a double root, this root must be
1
(1  b) .
2
We then get a first solution
1
(1 b )
y1 ( x)  x 2
and a second solution y 2 will be obtained by the method of reduction of order. Thus for positive
x,
dx
b x
e b ln x
1 1
(1b ) e (1b )
y2  x 2
 x(1b) dx  x 2
 x(1b) dx

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 69

 Applied Mathematics III

1
(1b ) dx
x 2
 x
dx

1
(1b )
x 2
ln x .
That is,
1
(1b )
y2  x 2
ln x ,
and the corresponding general solution is
1
(1b )
y( x)  (c1  c2 ln x) x 2


Solve x y  3xy  4 y  0 .

2
Example:

The auxiliary equation m  4m  4  0 has the double root m  2 . Hence the

2
Solution:
basis of real solutions for all positive x are
x2 , x 2 ln x
and the corresponding general solution is
y( x)  (c1  c2 ln x) x 2 
Summary of cases I  III of Euler-Cauchy Equation

Roots of Basis of General solution of

Case
m 2  (b  1)m  c  0 x 2 y  bxy  cy  0 x 2 y  bxy  cy  0
Distinct real
y1  x m 1 , y2  x m 2 y( x)  c1 x 1  c2 x 2
m m
I
m1 , m2
Complex conjugate
x  cos(v ln x) , y(x) 
II m1    iv,
x  sin(v ln x) x  A cos(v ln x)  B sin(v ln x)
m2    iv
Real Double root
1 1 1
(1  b ) (1  b ) (1b )
III 1 x 2
x 2
ln x y( x)  (c1  c2 ln x) x 2
(1  b) ,
2

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 70

 Applied Mathematics III

Group Activity 2.4

1. Find a real general solution for the following Euler-Cauchy differential equation.
a) x 2 y  5xy  9 y  0 e) x 2 y  20 y  0

b) xy  2 y  0 f) x 2 y  xy  y  0

c) x 2 y   xy   2 y  0 g) x 2 y  7 xy  9 y  0
2. Solve the following Euler-Cauchy initial value problems.
a) x 2 y  xy  9 y  0, y(1)  2, y(1)  0

b) x 2 y  3xy  y  0, y(1)  3, y(1)   4

c) x 2 y  2 y  0, y(1)  3, y(1)  1

3. Show that Euler-Cauchy Equations and Constant-Coefficient Equations can be

transformed into each other by setting x  e

t

Assessment :
 Asking an answer for some of the questions.
 Check students participation in the group activity and Give feedback to their answers
1.3.2 Non Homogeneous Linear Differential Equations

Consider a non-homogeneous linear differential equation

y  f ( x) y  g ( x) y  r ( x) , (1)
where f ( x), g ( x) and r (x) are continuous functions and r ( x)  0, and the corresponding
homogeneous equation

y  f ( x) y  g ( x) y  0 . (2)

The following Theorem states the relation between ( 1 ) and ( 2 ) .

Theorem 1: (Relation between solutions of (1) and ( 2 ) )

a. The difference of two solutions of (1) on some open interval I is a solution of ( 2 ) on I.

b. The sum of a solution of (1) on I and a solution of ( 2 ) on I is a solution (1) on I.

Proof: Exercise

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 71

 Applied Mathematics III

Definition: A general solution of the non-homogeneous equation (1) on some open

interval I is a solution of the form
y ( x)  y h ( x)  y p ( x) , ( 3)
where
yh ( x)  c1 y1 ( x)  c2 y2 ( x)
is a general solution of the homogeneous equation ( 2 ) on I and y p (x) is any particular

solution of (1) on I containing no arbitrary constants.

Remark:
1. To solve the non homogeneous equation ( 1 ) , we have to solve the homogeneous equation

( 2 ) and find any particular solution y p of ( 1 ) .

2. Any particular solution y p of ( 1 ) must be linearly independent with y1 and y 2 .

3. To obtain any particular solution y p (x), there are two methods

 The Method of Undetermined Coefficients and

 The Method of Variation of Parameters
4. The solution to the complete linear differential equation is composed of the sum of the
complementary function and the particular integral
1.3.2.1 The Method of Undetermined Coefficients
Techniques for obtaining the complementary function y c were developed in sections 1.3.2.2 and
1.3.2.3 with many working examples. What remains is only to provide techniques for finding a
particular integral in order to obtain a complete solution. In this section we shall discuss the
technique called the method of undetermined coefficients.
Undetermined coefficients: These are constants to be explicitly determined by solving the
particular integral of a differential equation. The method that does this is called the method of
undetermined coefficients.
Procedure for the techniques of undetermined coefficients
This method applies to forced differential equations of the form:

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 72

 Applied Mathematics III

y  by  cy  r (x) , (*)

with constant coefficients and special r (x) , namely, exponential functions, polynomials,
cosines, sines, or sums or products of such functions. These r (x) have derivatives of a form
similar to r (x) itself.

The general procedure in this technique is to assume the particular integral y p of a form similar

to the right member r (x) in equation ( * ) . Choose for y p a form similar to that of r (x) and

involving unknown coefficients to be determined by substituting that choice for y p into ( * ) .

The rules of the method are as follows

Rules for the Method of Undetermined Coefficients

(a) Basic Rule.

If r (x) in ( * ) is one of the functions in the first column, in the table given below, choose the

corresponding function y p in the second column and determine its undetermined coefficients by

substituting yp and its derivatives into ( * )

(b) Modification Rule.

If a term in your choice for y p happens to be a solution of the homogeneous equation

corresponding to (*) , then multiply your choice of y p by x repeatedly until no term of the
k k
product x y p is a solution of the homogeneous equation. Then use the product x y p to solve

the nonhomogeneous equation.

(c) Sum Rule.

If r (x) is a sum of functions in several lines of first column, (in the table given below), then

choose for y p the sum of the functions in the corresponding lines of the second column.

Note that we always have to solve the homogeneous equation first.

Table below summarizes a general rule for the formulation of the particular integral

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 73

 Applied Mathematics III

Terms in r (x) Choice for y p

i ke ax ce ax
ii kxn (n  0, 1, . . .) kn x n  kn 1 x n 1  . . .  k1 x  k

k cos bx  A cosbx  B sin bx

iii 
k sin bx 

iv kex cosbx e x ( A cos bx  B sin bx)


kex sin bx 
v (c0  c1 x  c2. x 2  ...  cn x n )e rx (C0  C1 x  C2. x 2  ...  Cn x n )e rx

The procedure is best illustrated with an example

Example 1 – Polynomials
a) Find the solution to the equation:

d 2 y dy
  2 y  x2 ,
dx 2 dx

subject to boundary conditions y0  0 and y ' 0  0 .

Solution: We first seek a solution to the homogeneous equation:

d 2 yc dyc
  2 yc  0
dx 2 dx

the subscript c indicating this is the complementary solution. The auxiliary equation is:

2    2  0,

giving 1  1 and 2  2 . The complementary solution is therefore:

yc  Ae x  Be 2x .

We will now find the particular solution. As we have an x 2 on the RHS we will guess that the
particular solution is of the form:

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 74

 Applied Mathematics III

y p  Ax 2  Bx  C.

To substitute this into the function we will also need the first and second derivative of y p :

y p '  2 Ax  B,
y p ' '  2 A.

This gives us:

 
2 A  2 Ax  B  2 Ax 2  Bx  C  x2 .

If we multiply out the LHS and collect powers of x we get:

 2 Ax 2  2 A  2Bx  2 A  B  2C  x2 .

Now we can equate the powers of x on both sides of the equal sign:

 2 A  1,

2 A  2B  0,

2 A  B  2C  0.

This is now a set of three simultaneous equations to be solved. In this case we can use the first
1 1
equation to find A   , then substitute that into the second to find B   and then use the
2 2
3
third equation to find C   . This gives us our particular solution:
4
1 1 3
y p   x2  x 
2 2 4
Note that we must solve the boundary condition with the full solution and not just the
complementary solution! So before solving for A and B we form the full general solution:

yx   yc  y p  Ae x  Be 2 x 
1 2 1 3
x  x .
2 2 4

To solve the second boundary condition we will also need the first differential:

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 75

 Applied Mathematics III

y' x   Ae x  2 Be 2 x  x  .
1
2

Thus the boundary conditions give us:

y 0  A  B 
3
 0,
4
y ' 0  A  2 B   0.
1
2
3 1 1
From the first we find A   B ; using this in the second we find  3B  0 , so B  and
4 4 12
2
A , giving us a final answer of:
3
2 x 1 2 x 1 2 1 3
y e  e  x  x .
3 12 2 2 4

b) Find the solution to the equation:

y  4 y  8x 2
Solution:

Observe that yh ( x)  A cos 2 x  B sin 2 x, and since r ( x)  8x is a polynomial of degree

2

two, then
y p ( x)  k2 x 2  k1 x  k0 .

By substituting y p and y p in the equation we obtain

k2  2, k1  0, k0   1
Thus
y p ( x)  2 x 2  1 ,

and the general solution is

y ( x)  y n ( x)  y p ( x)

 A cos 2 x  B sin 2 x  2 x 2  1 . 

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 76

 Applied Mathematics III

Example 2 – exponentials
a) Find the general solution to:
d2y
 4 y  e3 x
dx 2
Solution:
Again we start by finding solution to the homogenous equation:

d 2 yc
 4 yc  0.
dx 2

Forming the auxiliary equation:

2  4  0,

the roots of this are purely complex,   2i , thus the complementary solution is given by:

yc  A cos2 x   B sin2 x .

We then find the particular solution. The only function which can differentiate to give a term
containing e3 x is Ae 3 x , so we will take this as our guess and substitute it in:

9 Ae 3 x  4 Ae 3 x  e3 x ,

or, after collecting up like terms:

13 Ae 3 x  e3 x

1
so we find that A  and our particular solution is:
13
1 3x
yp  e .
13

As this question didn’t come with any boundary conditions we cannot solve for A and B , so it
is sufficient to give the full general solution:

y  yc  y p ,

y  A cos2 x   B sin 2 x  
1 3x
e .
13

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 77

 Applied Mathematics III

b) Find the solution to the equation:

y  4 y  4 y  e 2 x

Solution: Since

yh ( x)  c1e2 x  c2 xe 2 x ,
2x
and both e and xe 2 x are solutions of homogeneous equation, then by modification rule ,

y p ( x)  a x 2 e 2 x .
1
By substituting y p , yp and yp in the equation we find a  and hence
2
1 2 2x
y p ( x)  xe .
2
Thus
1 2 2x
y( x)  c1e 2 x  c2 xe 2 x  x e 
2
Example 3 – trigonometric
a) Find the general solution to:
d 2 y dy
  2 y  sin x 
dx 2 dx

Solution: We look to solve the homogeneous equation:

d 2 yc dyc
  2 yc  0.
dx 2 dx

Forming the auxiliary equation:

2    2  0,

we find 1  2 and 2  1 , giving the complementary solution:

yc  Ae 2 x  Be x .

This time in our search for a particular solution we will try:

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 78

 Applied Mathematics III

y p  A sinx   B cosx .

We must include both the sin x  and cosx  terms as one can differentiate into the other. In
order to substitute the particular solution into the equation we calculate the first and second
differentials:

y p '  A cosx   B sin x ,

y p ' '   A sin x   B cosx .

Substituting this into the question we find:

 A sinx  B cosx   A cosx  B sinx  2 A sinx  B cosx  sinx,

then collecting on sin x  and cosx  :

 A  B  2 Asinx   B  A  2B cosx  sinx.

Equating coefficients we find:
 3 A  B  1,
A  3B  0.
Solving these simultaneously gives:
3 1
A ,B   .
10 10
Thus our particular solution is:

sin x   cosx .
3 1
yp  
10 10
Hence the full general solution is:

sin x   cosx .
3 1
y  Ae 2 x  Be x 
10 10
Example 4 – Mixed
a) Find the solution to the equation:
y  3 y  2 y  4 x  e3 x

Solution: The corresponding homogenous equation is:

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 79

 Applied Mathematics III

d 2 yc dy
2
 4 c  3 yc  0,
dx dx

with the auxiliary:

2  4  3  0,

giving 1  1 and 2  3 giving the complementary solution:

yh ( x)  c1e 2 x  c2e x ,

and since r (x) is given as the sum of forms (i ) and (ii ) , then

y p ( x)  (ax  b)  ce 3 x

By substituting y p , yp and yp in the equation we find

1
a  2, b  3, c .
2
Hence
1 3x
y p ( x)  2 x  3  e ,
2
and the general solution is
1 3x
y( x)  c1e 2 x  c2 e x  e  2x  3 . 
2
Quick Check Activity:
1. Determine the form of particular solution to the following non homogeneous differential
equation (Individual Activity)
a) y  2 y  2 y  e 2 x cos x  3sin x   
c) y  4 y  x 2  3 sin 2 x

b) y  2 y  5 y  e x cos 2 x
2. Solve the following non homogeneous differential equations by the Method of Undetermined
Coefficients (Group Activity)
(a) x 2 y  4 xy  6 y  x 2  x
(b) y  y  x sin x c) y  y   2 sin x  4 x cos x

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 80

 Applied Mathematics III

Group Activity 2.5.1

1. Solve the following non homogeneous differential equations by the Method of
Undetermined Coefficients.
a) y  3 y  2 y  e x sin x f) y  3 y  28 cosh 4 x

b) y  y  1  x  x 2 g)
y  8 y  16 y  64 cosh4 x

c) y  2 y  y  x  e x h) y  2 y  35 y  12e5 x  37 sin 5x

d) y  y  e x i) y  6 y  9 y  50e cos x
x

e) y  y  2 e x  6e 2 x j) 3 y  10 y  3 y  9 x  5 sin x

2. Solve the following non homogeneous initial value problems by the Method of
Undetermined Coefficients.

a)
y  y  2 y  10 sin x, y(0)  1, y(0)   3.

b)
y  6 y 13 y  4 e 3 x , y(0)  2, y (0)  4 .

c)
y  4 y  e  2 x  2 x, y(0)  0, y(0)  4

c)
y  y  2  2 x  x 2 , y(0)  8, y(0)  1

Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity and Give feedback to their answers
1.3.2.2 The Method of Variation of Parameters
Method of solution by variation of parameters (VOP)
The method of undetermined coefficients discussed in the previous section is limited in its
application. We need another technique with wider application. The technique discussed in this
section is called the method of variation of parameters.
The VOP procedure consists of replacing the constants in the complementary function by
undetermined functions of the independent variable x and then determining these functions so
that when the modified complementary function is substituted into the differential equation,
f (x) will be obtained on the left side.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 81

 Applied Mathematics III

Consider a differential equation of the form

y  f ( x) y  g ( x) y  r ( x) , (1)

where f , g and r are continuous functions on some interval I, and a corresponding

homogeneous equation

y  f ( x) y  g ( x) y  0 . (2)

Recall that the general solution of ( 2 ) is given by

yh ( x)  c1 y1 ( x)  c2 y2 ( x) ,

where y1 , y2 form a basis of solutions, and the method of variation of parameters involves in

replacing the constants c1 and c2 (here regarded as “parameters” in y h ) by functions u (x) and

v(x) to be determined so that the resulting function

y p ( x)  u( x) y1 ( x)  v( x) y2 ( x) ( 3)

is a particular solution of equation ( 1 ) . By differentiating ( 3 ) we obtain

yp  uy1  uy1  vy2  vy2 .

To simplify this expression, it is convenient (but not necessary) to set

uy1  vy2  0 .

Then
yp ( x)  uy1  vy2 .

By differentiating this function we have

yp  uy1  uy1  vy2  vy2 .

Substituting the expressions for y p , yp and yp into ( 1 ) and collecting terms containing u and

terms containing v , we obtain

u( y1  f ( x) y1  g ( x) y1 )  v( y2  f ( x) y2  g ( x) y2 )  uy1  vy2  r .

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 82

 Applied Mathematics III

Since y1 and y 2 are solutions of the homogeneous equation ( 3 ) , this reduces to

uy1  vy2  r .

This gives a second equation relating u(x) and v(x), and we have the simultaneous equations

uy1  vy2  r ,
uy1  vy2  0.

Multiply the first equation by  y2 and the second by y 2 and add to get

u( y1 y2  y2 y1 )   y2 r .

Thus

 y2 r
u  .
y1 y2  y2 y1

Now multiply the first equation by y1 and the second by  y1 and add to get

v( y1 y2  y2 y1 )  y1r .

y1r
Thus v  .
y1 y2  y2 y1
Since y1 , y2 are linearly independent, we have Wronskian of y1 , y2 ,

W ( y1 , y2 )  y1 y2  y2 y1  0 .
By integration,

y2 r y1r
u   dx, v   dx .
y1 y2  y2 y1 y1 y2  y2 y1

Substituting them into ( 3 ) , we obtain

y2 ( x) r ( x) y1 ( x)r ( x)
y p ( x)   y1 ( x)  dx  y2 ( x)  dx .
y1 ( x) y2 ( x)  y1 ( x) y2 ( x) y1 ( x) y2 ( x)  y1 ( x) y2 ( x)

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 83

 Applied Mathematics III

Example: Solve the following non homogeneous differential equations.

a) y  4 y  e
3x

b) x y  4 xy  6 y  21x

2 4

c) y  16 y  17e , y(0)  6, y(0)  2

x

Solution:
a) A basis of solutions of the homogeneous equation on any interval is

y1  e 2 x , y 2  e 2 x .

This gives

y1  2e 2 x , y2   2e 2 x .

and hence

y1 y2  y1 y2   4.

The particular solution is

e 2 x . e3 x e 2 x . e3 x
y p ( x)   e 2 x  dx  e 2 x  dx
( 4) ( 4)

1 2x x 1
 e  e dx  e 2 x  e5 x dx
4 4

1 3x 1 3x
 e  e
4 20

1 3x
y p ( x)  e .
5

But since

yh ( x)  c1e2 x  c2e2 x ,

then the general solution is

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 84

 Applied Mathematics III

1 3x
y( x)  yh ( x)  y p ( x)  c1e 2 x  c2e 2 x  e . 
5

b) Since

y1  x 3 , y2  x 2

are the basis of the solutions of the corresponding homogeneous equation on any interval and

y1  3x 2 , y2  2 x

then

y1 y2  y1 y2   x 4 .

Re writing

x 2 y  4 xy  6 y  21x  4

we get
4 6
y  y  y  21x  6 .
x x

Thus

r ( x)  21x  6 .

Hence the particular solution is

(21x  6 ) x 2 (21x  6 ) x 3
y p ( x)   x 3  dx  x 2
 ( x 4 ) dx
( x 4 )

1
 x 4 .
2
But since

yh ( x)  c1 x3  c2 x 2 ,

then the general solution is

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 85

 Applied Mathematics III

1 4
y( x)  c1 x 3  c2 x 2  x . 
2

c) Since

y1  cos 4 x, y2  sin 4 x

are the basis of the solutions of the corresponding homogeneous equation on any interval and

y1   4 sin 4 x, y2  4 cos 4 x,

then

y1 y2  y1 y2  4 .

Thus the particular solution is

17 17
y p ( x)   cos 4 x  e x sin 4 xdx  sin 4 x  e x cos 4 xdx  e x .
4 4

But since

yh ( x)  A cos 4 x  B sin 4 x ,

then the general solution is

y( x)  A cos 4 x  B sin 4 x  e x .

From the initial condition

y(0)  6, y(0)   2 ,

and

y( x)   4 A sin 4 x  4B cos 4 x  e x

it follows that

y(0)  A 1  6  A5

3
y(0)  4 B  1   2  B .
4

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 86

 Applied Mathematics III

Thus

3
y ( x)  5 cos 4 x  sin 4 x  e x . 
4

Home take Group Activity: Solve the following differential equations by method of variation
of parameters:

(a) Solve x 2 y //  2 xy /  2 y  x 2  2 (d)  

Solve y //  3 y /  2 y  cos e  x

(b) x 2 y //  xy /  8 y  x 4  3 ln x , x  0 (e) y //  5 y /  6 y  x 2

(c)
y
Solve x y'' + y' - x = x ex (f) y //  4 y  3 sin x

g) y ''  y '  cot x

Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers
Group Activity 1.2.5.2
1. Solve the following non homogeneous differential equations by the Method of Variation
of Parameters.

a) y  2 y  y  e x cos x f) y  9 y  sin x

ex g) y  y  sec x
b) y  2 y  y 
x h) y  y  sec x
c) y  2 y  y  e x sin x
i) y  y  e2 x
d) y  5 y  6 y  x e
3 2x

j) y  2 y  y  e x
e)
y   2 y   y  e  x ln x
2. Solve the following non homogeneous Euler- Cauchy differential equations.

a) xy  y  (3  x) x 2e x d) x y  3xy  15 y 

2 1
x
b) 4 x y  8xy  3 y  7 x 0  15x e) x y  3xy  y  3x
2 2 3 2 2

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 87

 Applied Mathematics III

c) x y  4 xy  6 y  7 x sin x f) x y  3xy  15 y  x e

2 4 2 2 x

3. Find the steady-state oscillation of the mass-spring system governed by the equation
given below.

a) y   3 y   2 y  20 cos 2t

b)
y  2 y  5 y   13sin 3t
Assessment
 Asking an answer for some of the questions.
 Check students participation in the group activity.
 Give feedback to their answers

1.4. Systems of Differential Equations

To solve the systems of differential equations, we use the method of solution by elimination. In
this method, unknown functions and their derivatives are successively eliminated until one
arrives at a single higher order differential equation containing only one unknown function and
its derivatives.
Note that here y : y(t ) , x : x(t ) .
Now consider systems of the form
(a) x  a1 x  b1 y  f1

(b) y  a2 x  b2 y  f 2 ,

where f1 and f 2 are given functions and a1 , b1 , a2 , b2 are constants.

Let us now solve this equation.
(a)  b1 y  x  a1 x  f1 (*)

Again from (a ) , we have x  a1 x  b1 y  f1

 a1 x  b1 (a2 x  b2 y  f 2 )  f1 , using (b) .

 a1 x  b1a2 x  b1b2 y  b1 f 2  f1

 a1 x  b1a2 x  b2 ( x  a1 x  f1 )  b1 f 2  f1 , using (*).

 x  (a1  b2 ) x  (b1a2  a1b2 ) x  b2 f1  b1 f 2  f1

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 88

 Applied Mathematics III

 x  ( a1  b2 ) x  (a1b2  b1a2 ) x  b1 f 2  b2 f1  f1  x  bx  cx  r ,


where b   (a1  b2 ) , c  a1b2  b1a2 , r (t )  b1 f 2 (t )  b2 f1 (t )  f1 (t ) . By
solving this differential equation we get x , then y is obtained from
equation (*). 
Example: Solve the systems of differential equations:
(a) x  x  y
(b) y  x  y  et
Solution: From (a ) we get y  x  x (*)
Again from (a ) , we have x  x  y

 x  x  y  et , using (b) .

 x  x  ( x  x)  et , using (*) .

 2 x  et

Thus x  2 x  et

By solving this differential equation we get, x (t )  c1e

2t
 c2e 2t
 et .
Then from (*) we obtain

y  x  x  ( 2c1e 2t
 2c2e  2t
 et )  (c1e 2t
 c2e  2t
 et )

That is, y(t )  c1 ( 2  1)e

2t
 c2 ( 2  1)e  2t
.
Hence the general solutions of the systems of differential equations are

x (t )  c1e 2t
 c2e 2t
 et

. y(t )  c1 ( 2  1)e 2t
 c2 ( 2  1)e  2t

Quick Check Group Activity 2.6

Solve the following systems of differential equations.
x  x  y x  4 x  2 y
a) b)
y  3 x  y  e t y  3x  y  2e3
t

 Asking an answer for some of the questions.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 89

 Applied Mathematics III

1.5 Unit Summary

 Ordinary differential equation is an equation which involves one or several derivatives of
an unspecified function y of x ( y  f ( x)) ; the equation may also involve y itself
and constants.
 A differential equation of the form g ( y)dy  f ( x)dx is called separable.

 A first order differential equations of the form

M ( x, y)dx  N ( x, y)dy  0
is said to be exact if there exists a function u ( x, y ) such that

u u
 M,  N
x y
 The necessary and sufficient condition for the differential equation
M ( x, y)dx  N ( x, y)dy  0
to be exact is
M N
 .
y x
 If an equation of the form M ( x, y)dx  N ( x, y)dy  0 is not exact, it can be made exact
by multiplying by an integrating factor.

1  M N   k , or
 If      ,
N  y x   f ( x)
e  k dx , or

then an integrating factor is F ( x)   f ( x ) dx
e 

1  N M   c, or
    
y   f ( y )
 If ,
M  x
e  c dx , or

then an integrating factor is F ( y )   f ( y ) dy
e 

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 90

 Applied Mathematics III

 The general solution of the linear first order differential equation

y   f ( x ) y  r ( x) ,
where f and r are continuous functions, is given by

 f ( x ) dx   f ( x ) dxr ( x)dx  c 
y ( x)  e 

e

 The differential equation of the form y  p( x) y  g ( x) y a , where a is any real
number, is called the Bernoulli equation.
 A differential equation of the form
y   f ( x) y   g ( x) y  r ( x) , (1 )
where f , g and r are continuous functions, is called a linear differential equation of

the second order. If r ( x)  0, then equation (1) becomes

y   f ( x) y   g ( x) y  0
and is said to be homogeneous differential equation.
If r ( x)  0, then equation (1) is said to be non homogeneous differential equation.

 A solution of a homogeneous differential equation

y   f ( x) y   g ( x) y  0 ,
on some interval I, is said to be general solution if it expressed by the form
y( x)  c1 y1 ( x)  c2 y2 ( x) ,
where y1 and y 2 are linearly independent solutions.

 Consider a homogenous differential equation

y   by   cy  0 (1)
and its characteristic equation
2  b  c  0 (2)

Case Roots of (2) Basis of (1) General solution of (1)

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 91

 Applied Mathematics III

I Distinct real
1 , 2 e1 x , e2 x y  c1e 1x  c2 e 2 x

II Complex conjugate
e px cos qx ,
1  p  iq y  e px ( A cos qx  B sin qx)
1  p  iq e px sin qx

III Real Double root


b e x , xe x y  (c1  c2 x)e x
2

 Consider a homogenous Euler-Cauchy differential equation

x 2 y  bxy  cy  0 (1)

and its characteristic (or auxiliary) equation

m 2  (b  1)m  c  0 (2)

General solution of (1)

Roots of (2) Basis of (1)
Case

Distinct real
m1 , m2 y1  x m 1 , y2  x m 2 y( x)  c1 x m1  c2 x m2
Complex conjugate
x  cos(v ln x) , y(x) 
II m1    iv,
x  sin(v ln x) x  A cos(v ln x)  B sin(v ln x)
m2    iv
1
Real Double root (1  b )
2
x , 1
(1b )
III 1
(1  b) 1
y( x)  (c1  c2 ln x) x 2

2 (1  b )
2
x ln x

 A general solution of a linear non homogeneous differential equation

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 92

 Applied Mathematics III

y   by   cy  r (x) , (1)
is given by
y ( x)  y h ( x )  y p ( x) ,

where yh ( x)  c1 y1 ( x)  c2 y2 ( x) is the general solution of the corresponding homogeneous

equation y  by  cy  0 , and y p (x) is a particular solution of equation (1), given by the

formula

y 2 ( x) r ( x) y1 ( x)r ( x)
y p ( x)   y1  dx  y2  dx .
y1 ( x) y2 ( x)  y1 ( x) y2 ( x) y1 ( x) y2 ( x)  y1 ( x) y2 ( x)

1.7 Review Problems

1. Solve the following differential equations.

 y  y
a) xy   y  3 x 4 cos2   . b) xy   y  x 2 tan   0
 x  x
c) 9 y   ( x  y  1) 2  0

Find the exact differential equation if u  tan(y  x )

2 2
2.
3. Show that the following differential equations are exact and solve them.

a) (cot y  x 2 )dx  x csc2 y dy

2 xy
b) y  ln(1  y 2 )  0, y (2)  e 1
1 y 2

Solve the initial value problem: (e  ye )dx  (e  xe ) dy  0, y(1)  0

y x x y
4.
5. Find integrating factors and solve the following differential equations.

a) 2 x tan ydx  sec2 ydy  0


b) 2Sinydx cos ydy  0, y (0) 
2
6. Solve the following differential equations.

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 93

 Applied Mathematics III

a) y  y  2 sin x b) y  y sin x  ecos x

c) y  (1  3x 1 ) y  x  2, y(1)  e  1
7. Solve the Bernoulli equation: 3 y   y  (1  2 x) y 4

8. Solve the Clairaut’s equation: y  xy   y

9. Solve the following differential equations if one solution y1 is given

 1 
1
1 sin x
a y  y  1  2  y  0, y1 ( x)  x 2 sin x c) xy  2 y  xy  0, y1 
x  4x  x

b) 1 x y  2xy  2 y  0, y ( x)  x

2
1

10. Verify that, for any nonzero constant a , y1 ( x)  e ax is a solution of y  2ay  a 2 y  0 .
Find the general solution.
11. Solve the following differential equations

a) y  y  2 y  sin 2 x g)
y   2 y   y  xe x

e 2 x 2e x
b) y  4 y  4 y  ; x  0 h) y   2 y   y 
x2 x3
13. Find the steady-state oscillation of the mass-spring system governed by the equation:
y  2 y  4 y  sin 0.2t
14. Solve the following systems of differential equations.
a) x  2 x  y  sin t c) y  x  3 cos t , x(0)  1, y(0)   1
b) x  y  cos t  sin t d) y  x  cos t  sin t

Kassahun Nigatu(MSC),Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 94

 Applied Mathematics III

UNIT II

2. LAPLACE TRANSFORMS
UNIT INTRODUCTION
This unit is divided into five sections. The first section presents Definition to Laplace transforms,
Inverse Laplace Transform, discusses Laplace transforms of simple functions and establish
conditions for the existence of Laplace Transforms. The second section establish some useful
properties of Laplace Transforms and Applications of Laplace Transform to solve homogenous
and non-homogenous differential equations and Integral equations will be treated thoroughly in
the third section. Unit step functions and The Second shifting Theorem will be dealt in the third
section. The forth section presents Differentiation and Integration of Transforms. Convolution
Theorem and Integral equations will be studied in the fifth section.
Unit Objectives:
At the end of the unit students will be able to:
 Define Laplace Transforms
 Finding Laplace transforms of some functions using basic principles
 Identify Different Properties of Differential Laplace Transforms
 Find Laplace transforms of some functions using the standard Laplace Transforms
Tables and the properties of Laplace transforms
 Find the Laplace Transforms of first and second order derivatives
 Define Laplace Inverse Transforms
 Find Laplace inverse transforms of some functions using basic principles using
the standard Laplace Transforms Tables and the properties of Laplace transforms
 Solve differential equations using Laplace Transforms
 Apply Laplace transforms to solve differential equations modeling physical
problems

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 95

 Applied Mathematics III

2.1 The Laplace Transform; Inverse Transform

The Laplace transform is a very versatile tool to the engineer and scientist, for it enables one to
find solutions differential equations and corresponding initial and boundary value problems
involving homogenous and non-homogenous equations alike. It owes its present form to a
symbolic method developed by the English engineer Oliver Heaviside. Pierre Simon de Laplace
was a French mathematician who lived during 1749-1827, and was essentially interested to
describe nature using mathematics. The main goal of this chapter is to present those results of
Laplace which are used to find solutions of differential and integral equations. The method can
be extended to solve systems of differential equations, partial differential equations and integral
equations. Its importance lies in the fact that its application is considerably easier than other
available techniques.

Definition 2.1:
The Laplace transform of a given function f (t ) , that is defined for all t  0 , is defined as the
function F (s) such that

L[ f ](s)   e  st f (t )dt  F ( s) (1)
0

for all s such that the integral exists, where L is referred to as the Laplace Transform
operator. Furthermore the original function f is called the inverse Laplace transform of F (s)
and is denoted by
f (t )  L1 F (s) or L1 F (s)  f (t )

Remark 2.1
(a) Laplace transform takes a function f (t ) into a function F (s) of the parameter s .
(b) Original functions are denoted by lower case letters and their transforms by the same
letters in capitals, for instance F (s) denotes the transform of f (t ) . Thus we write

L f (t )  F s  or F s    e -st f t dt


(c) The defining equation for the Laplace transform is an improper integral, which is defined
as

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 96

 Applied Mathematics III

 N

 f x dx  lim  f x dx

a
N 
a

Thus, the existence of the Laplace transform of f depends upon the existence of the limit.
Example: Find the Laplace transform of
a) f (t )  1, t  0 b) f (t )  e a t , t  0

b) f t   t n d) f t   sin at

Solution: (a) By definition, we have


L1   e -st 1dt = lim
T

0
T  
0
e -st dt

 
T
  1  1 -ST
= lim e st  -  = lim - e  e -s.0
T 
  s  0 T   s

=
1
s

- lim e -st  1 = 0  1= , s>0
1
s
1
s

Therefore L f t  = 1 where, s  0 , f t   1 .
s

(b) By Definition we have

  
L tn =
0

e st t n dt

By applying the formula for integration by parts we conclude that

 
 t n e st 
 
L t =
n

 st n
e t dt = -  
n st n1
 e t dt
0
 s 0 s 0
The first term on the right-hand side is equal to zero for n  0 and s  0 , so
 
  e
Lt =n  st
t dt 
n n
s e
 st
t n 1 n
dt  L (t n1 )
s
0 0

Replacing n with n-1 in this equation, we get


 
L t n1 
(n - 1) 
L(t n2 ) 
 s 
  
Combining values of L t n and L t n 1 one can write

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 97

 Applied Mathematics III

 
L tn =
n(n - 1)
s2
L(t n2 )

Continuing in this way one gets

 
L tn =
n(n - 1)(n - 2).....3.2.1.
sn
. Lt 0 

 
Since L t 0 = L1 
1
s
by part (a), we obtain

 
L tn =
n!
s n1
,

where n! nn  1n  2    3  2 1 .

Therefore if n  1 by part (b) we get

Lt  
1
s2

(c)   0

L e at   e st t at dt , by Definition

T
T  1 ( as ) t 
lim 0 e
( as ) t
= dt = lim  a  s e  dt
T  T  0

 1 1 
= lim
T 
-
 a  s a  s 

1 1
=- = provided a  s  0 or s  a .
as sa

Thus the Laplace transform of e at is F s   L e at    1

sa
if s  a .

If may be observed that for s  0 , L(e at ) does not exist :

Let s  0 then the exponent of e is positive for t  0 . Therefore

T
T  e s t   1 s T 1 
lim 
 st
e dt = lim s = lim  - s e  s  = 
T 
0
T   0 T 

which means the integral diverges.

Let s  0 , then integral becomes

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 98

 Applied Mathematics III

dt  lim t T0 = - lim

T
lim 
T 
0
T  T 
T=

(d) By Definition

 T
Lsin at  =  e  st
sin atdt  lim  e  st sin atdt
T 
0 0

Using integration by parts we obtain

T T
 1 
T
1
 0
e st sin atdt = - e st sin at    a e st cos atdt .
 s 0 s 0
Using integration by parts again we obtain
T T
1 a a a2
 0
e st sin atdt   e sT sin T - 2 e sT cos aT  2 - 2  e st sin atdt
s s s s 0
T
a2
Bringing 
s2  0
e st sin atdt on the left hand side we get

a  s2 
T
1  sT a sT

 st
e sin atdt  ( e sin T - 2 e cos aT  2 ) 2 
0
s s s  s  a 2 

By taking the limit as T   in this equation

a  s2 
Lsin at   lim ( e sT sin aT - 2 e sT cos aT  2 ) 2
1 a

T  s s s  s  a 2 

 Lsin at  
a
s  a2
2

Quick Check Class Activity: Use the definition to prove that

a. Lcos at   b. Lsinh at   c. Lcosh at  

s a s
s  a2
2
s  a2
2
s  a2
2

Instructor's role
 Check and give feedback to their answers

2.1.1 Basic Properties of The Laplace Transform

The Laplace Transform and inverse Laplace transform are linear operators. between functions in
the t domain to functions in the s domain.

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 99

 Applied Mathematics III

Theorem ( Linearity ): Suppose L[ f ](s)  F (s) and L[ g ](s)  G(s) are defined for s  a ,
and a, and b are real numbers. Then

La  f t   b  g t   a  L f t   b  Lg t  for s  a .

and
L1 c1 F1 s   c2 F2 s   c1 L1 F1 s   c2 L1 F2 s 

Examples : Using Linearity

1. Find the Laplace Transform of each of the following function
a) f t   2t 2  4t  1 c) f t   sinh at

b) f t   2 sin 3t  e 2t
Solution
a) By Linearity Property we have
L{2t 2  4t  1}  2L{t 2 }  4L{t}  L1

 1!   1  4  4s  s
2
2!
 2( 3 )  4   2     
s s  s s3
b) By Linearity Property we have

L{2 sin 3t  e 2t }  2Lsin 3t L e 2t 

 3  1
 2 2 2 

s 3  s2
e at  e  at
c) Since f t   sinh at  , we have
2
 e at  e  at  1
Lsinh at   L  L e e
at

at

 2  2

Lsinh at   L 
1
2
     1 1
L e at  L e at   
1 

2 s  a s  a
Lsinh at   2
a
s  a2
Example 2 Find the following inverse Laplace transform.
 1  72
(i) L1  2  (ii)
s  2 s5

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 100

 Applied Mathematics III

L1 
2 3
 3  s5 
(iii) (v) L1  2 
( s  1) s
 ( s - 2s - 3 
 - 2s  6   s 1 
(iv) L1  2  L1  2
 ( s  4) 
(vi) 
 ( s - 4s 
Solution

(i) For k  2 and the linearity of the inverse Laplace transform we get

 1  1  2  1
L1  2 = L1  2 = sin 2t
s  2 2 s  2 2

ii. Look for the Laplace transform of basic functions on the right hand side which gives this

 
denominator. The only possibility is L t n =
n!
s n 1
where n must be 4. Therefore for n  4

 72   72.4!  72 1  4!  72 4
L1  5   L1  41 
 L  41   t  3t 4
s   4!.s  4!  s  24
iii. Look for the Laplace transform of basic functions on the right hand side giving a
denominator of ( s  1) and giving a denominator of s 3 .

 2 3  1  1  1  3 2
L1   3   2 L1  t
  3L  3   2e  t
s 1 s   s  1 s  2

iv. Observe that

 - 2s  6  1  - 2 s 6 
L1  2  = L  2  2 
 s 4  s  4 s 4 

 s   1 
=  2 L1  2  + 6 L1  2 
s  4 s  4
By the Linearity of the inverse transform,
6 1  2 
 2 cos 2t  L    2 cos 2t  3 sin 2t
2 s2  4

In (v) and (vi) but, there is a problem however, and that is that the denominator of F s  is not in
form of our transform of basic functions. To proceed, we need the method of partial fractions,
which will allow us to rewrite F s  Y in a form we can use.

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 101

 Applied Mathematics III

v. Since s 2  2s  3  s  3s  1 we get

s5 s5 A B
= = +
s - 2s  3 ( s - 3)( s  1)
2
s -3 s 1

where A and B are constants to be determined. We can write

s5 ( A  B ) s  ( A - 3B )
=
( s - 3)( s  1) ( s - 3)( s  1)

This implies that

s  5   A  Bs   A  3B

This gives A  B  1 and A  3B  5

Solving this simultaneous equation we get B  1 and A  2 . Therefore, we have

 s5  1  2 -1 
L1  2 = L   
 s - 2s - 3  s - 3 s 1 

 1  1  1 
= 2 L1   L   using linearity of L1
 s - 3  s  1
Hence
 s5 
L1  2   2e  e
3t t

 s - 2 s - 3 

s 1 s 1 1 1 5 1
vi. By Partial Fractions = = + 
2
s - 4s s ( s - 4) 4 s 4 s4
Therefore

 s 1  1 1  1  5 1  1  1 5 4t
L1  2  =- L    L   = - .1+ e
 s - 4s  4 s 4 s - 4 4 4

Kassahun Nigatu (MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 102

 Applied Mathematics III

Theorem 1.2 (First shifting theorem, or shifting in the s variable): Let L[ f ](s)  F (s)
for s  b  0 . Let a be any number. Then

 
L e at f t   F ( s  a) for s  a  b
and
L1F s  a   e at f t 

Proof: We obtain F ( s  a) by replacing s by s  a in the integral of the definition of Laplace

Transform, so that we get

 
 at
L e f (t ) =  e
 st
[e f (t )] dt =  e ( s a )t f (t )dt  F ( s - a).
at

0 0

and
 

F ( s  a)  e
s ( s a )t
f (t )dt   e  st e at f (t ) dt  L (e a t f (t )) .
0 0

If F (s) exist (i.e., is finite) for s greater than some k , then our first integral exists for s  a  k

Now take the inverse on both sides to obtain the second formula in the theorem. 

s sa
For instance, Since L[cos(bt )]  , then L[e at cos(bt )]  .
s  b2
2
( s  a) 2  b 2
A short list of Important Transforms
The following table gives a short list of transforms that are basic. Form these transforms we can
obtain nearly all the other transforms that we shall need in this chapter.

SN f (t ) L( f ) f (t ) L( f )
1 w
1 1 8 sin wt
s s  w2
2

1 s
2 t 9 coshat
s2 s  a2
2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 103

 Applied Mathematics III

2
2! a2
3 t 10 sinh at
s3 s2  a2

t n (n  0, 1,) n! n!
4 11 e at t n
s n 1
( s  a) n  1

(a  1) sa
5 ta, a  0 12 e at cos wt
sa  1 ( s  a) 2  w 2
1 w
6. e at sa
13 e at sin wt ( s  a) 2  w 2

s
7 cos wt
s  w2
2

Important note: We must know how to use the Tables of Laplace Transforms
Examples: Using tables to find Laplace Transforms and Inverse Transforms
Example 1: Find the Laplace Transform of each of the following function
(a) f t   e 2t sin 3t c) f t   e 2t sinh 3t

(b) f t   t 2 e t d) f t   2t 2 e 3t  3 sin(3t )  3

Solutions

(a) Since Lsin 3t  

3

, then L e 2t sin 3t   3
s 3
2 2
s  22  32
2! 2!
(b) Since L(t 2 )  , then L(t 2 e t ) 
s 3
s  13
c) Since Lsinh 3t  
3

,then L e 2t sinh 3t 
3
 .
s 3
2 2
s  22  32
d) From the table,
L(t 2 )  3  Lt 2 e 3t   and Lsin 3t   2
2 2 3 1
and 1
s s  33
s  32 s

So using Linearity
 2   3  1
L{2t 2 e 3t  3 sin(3t )  3}  2   3 2
3  2 
 3 
 ( s  3)   s  3   s 
4 9 3
  2 
( s  3) 3
s 3 2
s

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 104

 Applied Mathematics III

Example 2 Find the Laplace inverse of

5s  2 3s  2  
a) F s   b) F s  
4
c) L1  2
s  2 2
9 
s  6s  13
2
 
 s - 4s  20 

Solutions:

 
a) We look at the (s  2) 2  9 term on the bottom and find it in the Laplace Table. We find,


e 2t sin 3t  
3

( s  2) 2  9  

 
 both have (s  2)  9 on bottom.
2
s2
e 2t cos 3t   

( s  2) 2  9  

Now we want to write the top line in the problem, 5s  2 , in terms of the two top lines “3” and
“s - 2” from the table and put
5s  2  H (3)  K (s  2)
This is like our partial fractions problems:
Choose s = 2  8  H (3)  H  83

Equate s 5K
So that
5s  2 H (3)  K ( s  2) ( s  2)
F s  
8 3
  5
s  2 2
9 s  2 2
9 3 s  2  9
2
s  22  9

 L1 F s   8 3 e 2t sin(3t )  5e 2t cos(3t )

b) In this example, there is one quadratic factor on the bottom line. The first thing to do is
complete the square for this factor, i.e.

s 2  6s  13  s  3  32  13  s  3  9  13  s  3  4
2 2 2

3s  2
 F ( s) 
s  3 2
4 
This can now be done like the previous example (get H = 3 and K   72 ).
c) We begin by completing the square of the denominator, which gives
4 4
We have 
s  4s  20 ( s  2) 2  6
2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 105

 Applied Mathematics III

F s  2 
4
( s  2) 2  16

This means we should choose F s  

4
s  16
2

By the first shifting theorem

 
L e-2t sin 4t   F ( s   2)  F s  2
4
=
( s  2) 2  16
and therefore
 
  e sin 4t 
4
L1   2t

 ( s  2)  16 
2

Quick Check Group Activity:

1. Find the Laplace Transform of each of the following function:
a) f t   e 2t sin(6t ) c) f t   e t cos 2t  3 sin 5t 
1 d) f t   e 2t cos(4t )
b) 2t 2 e 5t  e t sin(t )
2
2. Find the inverse Laplace Transform of
4s  1 6s  6 s3
a) Y ( s) 

s  2s  3
2
 b)
( s  1) 2  81
c)
( s  3) 2  16

Home take Indivdual Exercises:

1. Find the Laplace Transform of each of the following function:
(a) f t   cos 2 t 
(c) f t   2  e 2t 
2

(b) f t   1  2t 
2
d) t 5  4 cos(3t )  6t 2
2. . Find the inverse Laplace Transform of each of the following
9 4 s
a.  2 c.
s  16 s  16
2
( s  2)(s  1)
4s 2
b. d.
s  25
2
s( s  3) 2

Before discussing the possible existence of F s    e -st f t dt , it is helpful to define certain
0

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 106

 Applied Mathematics III

terms
Definition 2.2 A function f is said to be piecewise continuous on the closed interval [𝑎, 𝑏] if
the interval can be divided into a finite number of open subintervals (𝑐, 𝑑) = {𝑡 [𝑎. 𝑏]/ 𝑐 <
𝑡 < 𝑑} such that

i. The function is continuous on each subinterval

ii. The function f has a finite limit as t approaches each endpoint from within the
interval; that is, lim f t  and lim f t  exist.
t d  t c 

The condition (ii) means that a piecewise continuous function f may contain finite or jump
discontinuities.
Definition 2.3 A function f is said to be of exponential order if there exist real numbers a,

M  0 and t 0 such that

f t   Me at for t  t 0 .

Now we prove the following basic existence theorem for the Laplace transform of a function f.

Theorem 2.1 (Existence theorem for Laplace transform)

Let f (t ) be a piecewise continuous function on every finite interval in the range t  0 and
satisfies
f t   Me at ,

for all t  0 and for some constants a, and M  0 . Then the Laplace transform of f (t )
exists for all s  a .

Proof: Since the function f is of exponential order, we know that there are constants a, and t 0

and M  0 such that

f t  e  at  M or f t e  at  M for t  t 0 .

It may be noted that e  at is always positive. Multiplying by e  st e at , we have

f t e  st  Me  st e at

Hence

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 107

 Applied Mathematics III


 Me ( s a )t 
L f t    | e
 
 st
f (t ) | dt   Me ( s  a ) t
dt  - 
0 0
 s  a 0

Me  ( s a )t M
= - lim 
t  sa sa
Since first term is zero for 𝑠 > 𝑎, we have

M
 le
 st
f (t )dt  ,sa
0
sa

which implies the existence of the improper integral defining the Laplace transform of f
and completes the proof.
Corollary 2.1 Let f be a piecewise continuous function of exponential order defined on [0,),

and L{ f t } exists. Then lim F s   lim L f t   0 .

s  s 

 
L f t    e
M
Proof :  st
f (t )dt   le st f (t )dt  , as seen in the proof of Theorem 2.1.
0 0
s-a

Taking limit as s   we get lim F s   lim L f t   0 .

s  s 

Group Exercise 2.3.1

1 Find the Laplace transform of f (t )  cos t  sin t .

2. Verify formulas 11 to 13 given above on the table of Laplace transform.
3. Find the Laplace transform of the following functions

a) f (t )  t 2 e  3t b) f (t )  t 2 cost c) f (t )  e t sin 2 t
5
1 1  cos at
d) f (t )  t 2 e) f (t )  2e cos ( t )
t 2
f) f (t )  , a  const.
2 t

h) f (t )  5e sinh 2t
t
g) f (t )  e cos  t i) f (t )  te sin t
2t t

j) f (t )  sinh t cos t k) g(t)  (t  1) e

2 t

4. Find f (t ) if L ( f ) is

s4 1  7s 1
a) b) c)
s2  4 ( s  1)( s  2)( s  3) ( s  1) 2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 108

 Applied Mathematics III

5
s ak
d)
 1
2 e)
4
s  2s  3
2
f) sk
k 1
2

 s   1
 2
3 s9
g) i)
s  6s  18
2
s  6s  13
2

2.2 Laplace Transforms of Derivatives and Integrals

dy
For solving 2nd order differential equations we need to evaluate the Laplace transforms of
dt
d2y
and .
dt 2
2.2.1 Transforms of Derivatives

Theorem 2.3.1: (Laplace transform of the derivative of f (t ) )

Suppose that f (t ) is continuous and satisfies

f (t )  Me t , for all t  0 ,

for some  and M , and has a derivative f (t ) that is piecewise continuous on every

finite interval in the range t  0 . Then the Laplace transform of the derivative f (t )

exists when s  , and

L  f   s Lf   f (0) .
(*) Proof: Exercise

Remark: By applying (*) to the second order derivative f (t ) we obtain

L ( f  )  s L ( f  )  f (0)  ss L (t )  f (0)  f (0)

 s 2 L ( f ) - sf (0) - f (0) .

That is, L  f   s 2 L ( f ) - sf (0) - f  (0) .

Similarly L  f   s 3 L  f   s 2 f (0)  sf (0)  f (0) .

By induction we thus obtain

L  f ( n)   s n L ( f ) - s n - 1 f (0)  s n  2 f (0)    f ( n  1) (0) . 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 109

 Applied Mathematics III

Examples:
(a) Using the Laplace transform of f ' ' find Lsin kt .

(b) Using the Laplace transform of f ' ' find L sin 2 t .  

Solution (a) Let f t   sin kt , then f ' t   k cos kt , f '' t   k 2 sin kt , f 0  0 and

f ' 0  k . Therefore by above theorem since L f " (t )  s 2 F s   sf 0  f ' 0 we have:

 
L f " (t )  s 2 F s   k  L - k 2 sin kt  s 2 F s   k

where F s   L f t   Lsin kt

L{ f '' t }  L{k 2 sin kt}  s 2 F s   k  s 2 Lsin kt  k .

 L{k 2 sin kt}  k 2 Lsin kt  k  s 2 Lsin kt  k

Solving for Lsin kt we get

Lsin kt 
k
s  k2
2

Solution (b) Let f t   sin 2 t , then f ' t   2 sin t cos t  sin 2t , f 0  0 .
Therefore
L f ' (t )  sF s   f 0  L f ' (t )  Lsin 2t  sF s 

 
where F s   L f t   L sin 2 t . But we also know that Lsin 2t 
2
s 4
2

Therefore

  
 L f ' t   L{sin 2t}  sL sin 2 t   2
s 4
2

 
Solving for L sin 2 t L we get

Lsin t 
2

s s 4
2

Quick check Exercises
1. a) Using the Laplace transform of f ' ' find L(cos at )
b) Using the Laplace transform of f ' ' find L(t cos t )
Instructor's role
 Check and give feedback to their answers

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 110

 Applied Mathematics III

2.2.2 Application To Differential Equations; Initial Value Problems

The Laplace transform method solves differential equations and corresponding initial and
boundary value problems. The process of solving consists of three main steps:
 The given “hard” problem is transformed into a “simple” equation (subsidiary equation)
 The subsidiary equation is solved by purely algebraic manipulations.
 The solution of the subsidiary equation is transformed back to obtain the solution of the
given problem.
In this way Laplace transforms reduces the problem of solving a differential equation to an
algebraic problem.
We shall now discuss how the Laplace transform method solves linear initial value problem with
constant coefficients.
Consider the non homogeneous differential equation with constant coefficients
y  by  cy  r (t ), y(0)  k0 , y(0)  k ,
where r is a continuous function.
Now taking the Laplace transforms of both sides, we obtain
L ( y)  b L ( y)  c L ( y)  L (r )
Now by using Theorm1, we have
 s 2 L ( y)  sy(0)  y(0)   b  s L ( y)  y(0)  c L ( y)  L (r ) .

Then
 s 2  bs  c  L ( y)   sy(0)  by(0)  y(0)  L (r ) .

So
( s  b) y(0)  y(0)  L (r )
L ( y)  .
s 2  bs  c
But since
y(0)  k0 , y(0)  k1,
we get
( s  b)k0  k1  L (r )
L ( y)  . 
s 2  bs  c

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 111

 Applied Mathematics III

Three facts are evident from this equation.

a. Initial conditions must be given so that the first two terms in the numerator are
determined.
b. The function r must have a Laplace transform so that the last term in the numerator
is determined.

c. We must be able to find L 1 of the right hand side of the expression in order to
determine the solution y of the initial value problem.
Thus Laplace transform methods are primarily intended for the solution of the linear initial value
problems with constant coefficients.

Example: Solve the following initial value problems by means of Laplace transform

a) y  4 y  1, y(0)  1
b) y  3 y  2 y  4t  6, y(0)  1, y(0)  3
Solution:
a) Taking the Laplace transform of both sides we get
L ( y)  4 L ( y)  L (1),
that is,
1
s L ( y)  y  0   4 L ( y)  , .
s
Hence
1 1 5 1 11
L ( y)     .
s  4 s( s  4) 4 s  4 4 s
Taking the inverse on both sides we get
5  1  1 -1  1 
y (t )  L -1   L   .
4  s 4 4  s
Or
1
y (t )  (5e4t  1) . 
4
b) Taking the Laplace transform in both sides and using the differentiation property, we
have,

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 112

 Applied Mathematics III

 s L ( y)  s  3  3 s L ( y) 1  2 L ( y)  s4
2
2

6
s
.

So

4 6 s 3  6s  4
 
s  3s  2 L ( y)  s  2  
2

s s s2
.

Hence, factoring the numerator and denominator, we obtain

s 3  6s  4
L ( y)  2 2  2

 s  2  s 2  2s  2 

s s  3s  2 
s  s  2  s  1

s 2  2s  2 1 2
   2.
s ( s 1)
2
s 1 s
That is,

 1  1  1 
y (t )  L 1    2L  2  .
 s 1  s 
Or

y(t )  et  2t . 

Example 3

Find i(t) using Laplace transform method for t>0

Solution:
(1) Before switched from 1 to 2 at t=0
4
i   2 A  i (0  )  2 A
2
(2) System equation (t>0)
di(t )
L  Ri (t )  0 ( L  1H ) ( R  2ohm)
dt
KVL:
di(t )
  2i(t )  0
dt

(3) Solve system equation using Laplace transform

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 113

 Applied Mathematics III

 di (t )   di (t ) 
L  2i (t )  L   2 L[i (t )]
 dt   dt 
 sI ( s )  i (0  )  2 I ( s )
 ( s  2) I ( s )  2  0
2
 I ( s) 
s2
 i (t )  2e 2t u (t ) A

Quick Check Group Activity

Problem solving: Solve the following initial value problems by means of Laplace transform

a) y  4 y  1, y(0)  1
b) y  3 y  2 y  4t  6, y(0)  1, y(0)  3

c) y // (t )  y / (t )  1 , y(0)  1, y / (0)  0

d) y // (t )  3 y / (t )  2 y(t )  4 , y(0)  1, y / (0)  0

2.2.3 Laplace Transforms of Integrals .

Theorem 2: (Laplace transform of Integration of f (t ) ).

Let F (s) be the Laplace transform of f (t ) . If f is piecewise continuous and satisfies the
inequality f (t )  Me t , for all t  0, then

 t

L  f (u )ds 
0
1
s
F ( s)

Or if we take the inverse transform on both sides,

t 1 

0
f (u )du  L1  F ( s) 
s 
.

Proof: Exercise

1 1
Example: Find f (t ) if a) L( f )  b) L( f ) 
s s
2
s s
3

Solution:
a) Since
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 114
 Applied Mathematics III

1 1 1 
  
s  s s  s 1 
2

and

 1 
L 1  t
e ,
 s 1 
we get

 1  
1 1  1 
 t x
f (t )  L 1  2     e dx  1 e .
t
  L   
 s  s   
s s  1  0
b) Observe that
1 1

s  s s ( s 1)
3 2

and

 1 
L 1  2   sinh t.
 s 1 
Hence

 1   1 1  t
f (t )  L 1  3   L 1   2     sinh xdx
 s s   s  s 1   0
 cosht 1 .
Group Exercises 2.3.2
1) Solve the following initial value problems by Laplace transform

a) y  2 y  e t , y(0)  1
b) y  3 y  10 sin t , y(0)  0
c) y  9 y  t 2 , y(0)  0  y(0)
e) y  y  2cos t , y(0)  3, y(0)  4
g) y  4 y  3 y  6t  8, y(0)  0, y(0)  0
2. Find f (t ) if L ( f ) is given
1 4 1
a) b) c)
s(s   ) s ( s 1)  s  1
2 2 2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 115

 Applied Mathematics III

2.3 Unit Step Function and Second Shifting Theorem

Definition:
The unit step function (or Heaviside function) H (t  a) is defined by

0 if t  a
H (t  a)   . (1)
1 if t  a
In particular if a  0 , then
0 if t  0
H (t )   .
1 if t  0

Note that H (t  a) has a discontinuity at t  a and is not defined at t  a .

The Laplace Transform of H (t  a) :

Let H (t  a) be the unit constant function for t  0 (also called the Heaviside Function

H (t  a)
By definition, we have
 a 
e a s


L  u (t  a)    e u (t  a)dt   e u (t  a)dt   e dt   e dt 
 st  st  st
, st
s0
0 0 a a s
e  as e 0.s 1
In Particular if a  0 , L{H t  a }  LH t =  
s s s
a) The Heaviside function is used to “switch on” inputs.
If 𝑓(𝑡) is defined for all t values then H t  f t   0 if 𝑡 < 0 and H t  f t   f t  if t  0 i.e.
𝑓(𝑡) has been ignored if 𝑡 < 0 and switched on for t  0 . This is a very handy function for
looking at the response of a system to an input.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 116

 Applied Mathematics III

 0 if t  a  0 if t  a
-- On-off effect: H (t  a) g (t )   , H (t  a) g (t  a)  
 g (t ) if t  a  g (t  a) if t  a

b) The Unit step function can also be used to write piecewise defined functions in a compact
form(single formula).

A piece wise defined functions of the form

 g t ,0  t  a
f t   
 ht , t  a

is the same as f t   g t H t   ht   g t H t  a 

Similarly

 g t , a  t  b

f t    ht , b  t  c
k t , t  c


is the same as f t   g t H t  a   ht   g t H t  b  k t   ht H t  c 

Theorem (Second shifting theorem, or shifting in the t variable):

Let L[ f ](s)  F (s) for s  b . Then L[ H (t  a) f (t  a)](s)  e  as F (s) for s  b and

L[ H (t  a) f (t )](s)  e  as L f t  a .

Taking the inverse laplace transform on both sides we have

L1[e  as F (s)]  H (t  a) f (t  a)

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 117

 Applied Mathematics III


Proof LH (t - a) f (t - a) =  e  st H (t - a) f (t - a) dt
0


=  e  st f (t - a) dt
a

because H (t  a)  0 for t < a and H (t  a)  1 for t  a. Now let u  t  a in the last integral.
We get

L H (t - a) f (t - a) =  e  s
( a u )
f (u )du
0


= e as  e  st f (u )du  e as F s 
0

Examples:
1) Compute L  f (t )  when

a) f (t )  sin a(t  b)u(t  b)

e t 0  t  2 ,
b) f (t )   t
e  cos t , t  2
Solution:
1. a) By the Second Shifting Theorem (Theorem 1) we get
L  f (t )   L  sin a(t  b)u(t  b)   eb s L (sin at )
ae bs
 .
s2  b2
b) Since f (t ) has a jump discontinuity at t  2 and

0, t  2 ,
u (t  2 ) cos(t  2  
cos(t  2 )  cos, t  2 ,
we have
f (t )  e t  u(t  2 ) cos(t  2 ) .
Thus, by the second shifting theorem, we get
L  f (t )   L (et )  L  u(t  2 )cos(t  2 ) 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 118

 Applied Mathematics III

1
  e2 s L (cos t )
s 1

1 se 2 s
  . 
s 1 1  s 2
Example 2: Compute

 
s
 b) L  t 2u (t  3) 
1 e 2
a) L 1  
 1 s 2 
 
Solution: a) Observe that

 
s
   2s 
L 1 
1 e 2   L 1  1   L 1  e 
 1 s 2   1 s 2   1 s 2 
 
   
 
= sin t  u (t  )sin(t  )
2 2

= sin t  u (t  ) cos t . 
2
b) Let f (t  3)  t .Then, since
2

t  (t  3)  3  (t  3)  3,
we have

f (t )  (t  3) 2  t 2  6t  9 .

 2 6 9
Thus L (t 2 u (t  3))  e 3s L (t 2  6 t  9)  e 3s  3  2   .
s s s

 0,0  t  2
Example 3. Compute Lg t ,where g t    2
t  1, t  2

Solution; Using Heaviside Function we may write g t   H t  2 t 2  1 .  

Using L[ H (t  a) f (t )](s)  e  as L f t  a  we have

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 119

 Applied Mathematics III

 
Lg t   L H t  2 t 2  1   e 2 s
s
 
L t  2  1
2

Lg t  
e 2 s
s

L t 2  4t  5 e  as
s

L t 2  4 Lt  5L1 
e 2 s  2 4 5
Lg t    
s  s 3
s 2 s 

 2,0  t  

Example 4. Compute L f t  where f t   0,   t  2
sin t , t  2


Solution; Using Heaviside Function we may write

f t   2.H t   0  2H t     sin t  0H t  2   2H t   2H t     sin tH t  2 

L{ f t }  2L{H t }  2L{H t   }  L{sin tH t  2 }

But since sin t  sint  2  we have

L{ f t }  2L{H t }  2L{H t   }  L{sint  2 H t  2 }

Using Linearity and L[ H (t  a) f (t  a)](s)  e  as F (s) we have

L{ f t }  2L{H t }  2L{H t   }  L{sint  2 H t  2 }

s
e 2s
L{ f t }   2
2 e

s s s
The unit step function can be used as a building block in the construction of other functions; for
example,
f1 (t )  u(t  a)  u(t  2a)  u(t  3a), a 0 ,
Yields a three - step staircase. By the linearity property of Laplace transforms we obtain
L  f1 (t )   L  u(t  a)   L  u(t  2a)   L u(t  3a  , a 0

  e  e2 as  e3as  .
1  as
s
Consider an infinite staircase

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 120

 Applied Mathematics III

f (t )  u(t )  u(t  a)  u(t  2a)  u(t  3a)  , a  0.

Since e  1 , if as  0 we can use the formula for the sum of geometric series:
 as


1
x
n0
n
1  x  x 2  
1 x
, x  1.

Then, for s  0 , the Laplace transform is given by

L ( f )  L  u(t )   L  u(t  a)   L u(t  2a)   L u(t  3a)   .


1
s
1 e as  e2 as  e3as  . . .


1
s
1  e  a s  (e  a s ) 2  ( e  a s ) 3  . . .  
1
s 1  e as 
. 

Unit Impulse Function

The unit impulse function (also called the Dirac delta function)  (t  a) is loosely described as a
“function” that is zero every where except at t  a and has the property that


  (t  a)dt  1


Note that L ( (t  a))  e a s .

Example: solve the following initial value problems
1, 0  t  1,
a) y  y  f (t )   y (0)  y(0)  0
0, t  1,
b) y   2 y   y   (t 1), y(0)  2, y (0)  3
Solution:
a) Since f (t )  u(t )  u(t 1), then taking Laplace transforms, we get

1 e s
s L ( y)  L ( y)  
2
.
s s
Or
s

 s 1 L ( y)  1se
2
.

Hence
1 e s
L ( y)  ,
s( s 2  1)

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 121

 Applied Mathematics III

1 1 s
and since   2
s( s  1) s s  1
2

We have
1 s e s se s
L ( y)   2   2 .
s s 1 s s 1
Using the second shifting theorem (Theorem1), we obtain
y(t )  1 cos t  u(t 1)  u(t 1) cos(t 1)

1 cost  u(t 1)1 cos(t 1) . 

b) Taking Laplace transforms, we obtain

 s 2 L ( y)  2s  3  2  s L ( y)  2  L ( y)  e s ,
Or

s 2
 2s 1 L ( y)  2s  7  e s .

Hence we get
2s  7  e s 2( s 1) 5 e s
L ( y)    
s 2  2s 1 ( s 1)2 ( s 1)2 ( s 1)2

2 5 es
   .
s 1 ( s 1) 2 ( s 1) 2

1
Since L (tet )  , it follows from Theorem 1 that
( s 1)2

e s
 e s L  te t   L  (t 1)e ( t 1) u (t 1)  .
( s 1) 2

Thus

y(t )  2et  5tet  (t 1)e (t 1)u(t 1)

 et  2  5t  e(t 1)u(t 1) . 

Group Exercises 2.3.3

1. Use the Second Shifting Theorem to show the given equation.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 122

 Applied Mathematics III

 1 1 2 2 1
a) L  tu (t 1)   e 
s

s
2
 
s
2
 s
b) L t u (t 1)  e 
s
3
 2  
s s
2. Compute the inverse of the given Laplace transform

e  s e 3 s e 3 s s  se  s
a) b) c) d)
1 s2 s3 ( s 1) 3 1 s 2
3. Using the Laplace transform, solve the following problems

a) y  6 y  8 y  e3t  e5t , y(0)  0, y(0)  0

4t , 0  t 1,
b) y  3 y  2 y  f (t ), y(0)  0, y(0)  0 where f (t )  
8, t 1.
0, 0  t  3
c) y  4 y  f (t ), y(0)  0, y(0)  0, where f (t )   d)
t , t  3
8sin t , 0  t   ,
y  9 y  f (t ), y(0)  0, y(0)  4, where f (t )  
0, t   .
e) y  y  6 y   (t  2), y(0)  1, y(0)   2
f) y  4 y 13 y   (t 1), y(0)  0, y(0)  3

2.4 Differentiation and Integration of Transforms

In this section we consider differentiation and integration of transforms F (s) and find out the

corresponding operations for original function f (t ) .

2.4.1 Differentiation of Transforms

It can be shown that if f (t ) satisfies the conditions of the existence theorem, then the derivative
F (s) of its transform


F ( s)  L ( f )   e st f (t )dt .
0

with respect to s can be obtained by differentiating under the integral sign with respect to s ;
thus
dF ( s)  d st 
F ' ( s)    e f (t )    e  st (tf (t )) dt   L (tf (t )) .
ds 0 ds 0

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 123

 Applied Mathematics III

We have proved the following Theorem.

Theorem:
If L ( f )  F ( s) exists for s  a, then L tf (t )  exists for s  a and

L tf (t )    L  f (t )    F ' ( s) .
d
ds
(1)
d 
Or L 1  L  f (t )     t f (t ) .
 ds 
Proof: Exercise

Note that differentiation of the transform of a function corresponds to the multiplication of the
function by  t .
Using equation (1) repeatedly, we obtain

L t f (t )    L t f (t )   (1) L t f (t )     (1)
2 n
d 2 d n d
n n -1 n-2
L f(t) .
ds ds 2 ds n
s s2 - k 2
For instance: Since L[cos(kt)]  , then L[t cos( kt)] 
s2  k 2 (s 2  k 2 ) 2
Examples:
1. Find the laplace transform of
a) L[t sin t ] b) L[t 2 cos t ] c) L[te 2t cos 5t ]

Solution:
1
a) Since L[sin t ]  . Hence
s 1
2

d  1 
L[t sin t ]   1
2s
 2  2

ds  s  1  s  1 2 
s
b) We have L[cos t ] 
s 1
2

Hence Using property (c)

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 124

 Applied Mathematics III

d 2  s  2s 3  6s
L[t 2 cos t ]   1 
2


ds 2  s 2  1  s 2  1 3 
c) By shifting Property
s2
L[e 2t cos 5t ] 
s  22  25
Hence using (C)

d  s2  s  22  25
L[te 2t cos 5t ]   1  

ds  s  22  25  s  22  25 2 
2.Find
s-a 2ks
a. ln b.
s -b (s - k 2 ) 2
2

Solution:

a. Let F s   ln  ln s  a   ln s  b  . Then
s-a
s -b

F ' s  
1 1

s a s b
Therefore
 1 1 
L1{F ' s }  L1     e e
at bt

s  a s  b
Using Inverse Laplace Transform of derivatives property, we have
 
L-1 F ' s   (1)tf t 

 1 1 
L1{F ' s }  L1     e  e  tf t 
at bt

s  a s  b

e bt  e at
 f t  
t

b. Let Gs  
2ks
.
(s - k 2 ) 2
2

Then we know that

d  s 
Gs     F s 
2ks d
  2 2 
2 2 2
(s - k ) ds  s  k  ds

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 125

 Applied Mathematics III

where F s  
k
.
s -k2
2

Therefore
 d  k  1 1  k 
L1 Gs    L1{F ' s }   L1   2 2 
t L  2 2 
 t sinh kt where
 ds  s  k  s  k 
Example 2: Solve the differential equation.
ty   ty   y  0, y(0)  0, y (0)  3
Solution: Since

L ( y)   s 2 L ( y)  sy (0)  y(0) 

d d
L (ty)  
ds ds
d
  s2 L ( y )  2sL ( y)  y(0)
ds
d
  s2 L ( y )  2sL ( y ) ,
ds
and
d d
L (ty)   L ( y)   sL ( y)  y (0) 
ds ds
d
s L ( y)  L ( y) .
ds
Substituting these expressions into the Laplace transform of the original equation yields
L ( y  ty  y)  L ( y)  L (ty)  L ( y)
d d
 ( s 2 L ( y)  2sL ( y) )  ( s L ( y)  L ( y) )  L ( y)  0 .
ds ds
Rearranging and canceling terms, we have
d
( s 2  s) L ( y )  2sL ( y )  0 .
ds
We now divide both sides by s  s  s( s 1) to obtain
2

d 2
L ( y)  L ( y)  0 .
ds s 1
Separating variables, we have

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 126

 Applied Mathematics III

dL ( y )  2
 ds ,
L ( y) s 1
and an integration yields
ln L ( y)   2 ln s  1  c1 .
Or
c
L ( y)  .
( s 1) 2
That is,

y(t )  c t e t .

But since y(0)  0 and y (0)  3, then c  3 and hence

y(t )  3tet .
Remark: Observe that all variable coefficients equations can not be solved by this method.
It works only when
a) The coefficients are polynomials in t ;
b) The differential equation involving L( y) can be solved; and
c) The inverse transform of L( y) can be found.
Quick Check Group exercises
1. Find the Laplace Transform of the following functions
a. te 2t sin t d. t cosh at
 
b. t 2  3t  2 sin 3t e. te t sinh t

c. t sinh at  
f. t 2  4 cos 2t
2. Find the Laplace inverse transform
a 1 (s 2  b 2 ) sa
a. arctan   b. ln c. cot 1  
s 2 (s 2  a 2 )  b 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 127

 Applied Mathematics III

2.4.2 Integration of Transforms

Theorem:
f (t )
If L ( f )  F ( s) exists for s  a , and lim exists, then
t 0 t
 f (t )  
L    F ( s )ds ( s  a) .
* *
(2)
 t  s
  f (t )
Or L 1   F ( s*)ds * 
s  t

Proof: From the definition it follows that


   s*t


s F (s*)ds *  s  0 e f (t )dt  ds * .
Reversing the order of integration we get
 
 s*t * 
s F ( s*) ds *  0 f (t ) s e ds  dt .

 1   1  1
e ds *  lim   e  s*t   lim   e  s*t   e  st .
 s*t
But since
s
s*  
 t  s*  s  t  t
 
f (t )  f (t ) 
 F (s*)ds *   e dt  L  ( s  a) .
 st
Then , 
s 0 t  t 
t

Remark: If g (t )   f (u ) du ,
0
then

a) g (t )  f (t ) at all points of continuity of f (t ) ;

b) L  f (t )  L (g(t))  sL g (t )   g (0);

c) g (0)  0 .

1
Finally, using parts (b) and (c), we get L  g (t )   L  f (t )  .
s
t  1
Thus we conclude that L   f (u )du   L  f(t)  .
0  s
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 128
 Applied Mathematics III

Example: Find the Laplace transform of the following functions.

t
sin t sin u
a) f (t )  b) 0 u du
t
Solution:
a) Since
1
L (sin t )  ,
s 1
2

we obtain

 sin t   du
L  2  lim(arctan u )  lim(arctan u)
 t  s u 1 u  u s


  arctan s . 
2
 t sin u  1  sin t  1 
b) L  du   L     L (sin t )dt
0 u  s  t  ss

1 du
  2
s s u 1


1

limarctan u  limarctan
s u  u s
u 
1 
   arctan s  . 
s 2 
Group Exercise 2.3.4

1) Find a) L  t 2 sin at  b) L  t cos at 

c) L  teat sin bt  d) L  tet sinh 2t 

2. Using the Laplace transform solve the following initial value Problem.
a) y  3ty  6 y  1, y(0)  0, y(0)  0
b) y  2 ty  4 y 1, y(0)  y(0)  0
2) Find the Laplace transform of

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 129

 Applied Mathematics III

1  cos au
t

a)
2
t sin 2t b)
sin 3t
t
c)
1 cos at
t
d) 
0
u
du

2) Using equations (1) or (2) find the inverse transform

1 s s
a) b) c)
( s  3) 3 ( s  a) 2
2
( s  4) 2
2

2.5.Convolution and Integral Equations

2.5.1. Convolution

Definition:
If f and g are piecewise continuous functions, then the convolution of f and g ,

written f  g , is defined by
t
 f  g t    f ( x) g (t  x)dx .
0

Using the change of variables u  t  x, we see th

t 0

( f  g )(t )   f ( x) g (t  x)dx    f (t  u) g (u)du

0 t

  f (t  u) g (u)du
0
t

  g (u ) f (t  u )du  ( g  f )(t ) .
0

Hence ( f  g )(t )  ( g  f ) (t ) .
Note that we can take the convolution in either order without altering the result.

Theorem: (Convolution Theorem for Laplace Transform)

If L ( f )  F (s) and L ( g )  G(s) exist, then L ( f  g )(t )  exists and

L ( f  g )(t )  F (s)G(s)  L ( f ) L ( g ) (1)

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 130

 Applied Mathematics III

Proof: By definition

  su    sv 
F ( s)G( s)  L ( f ) L ( g )    e f (u )du   e g (v)dv
0  0 


 e
s (u  v )
f (u ) g (v)dvdu (2)
0 0

If we make the change of variables t  u  v, then v  t  u and dv dt, and integral (1) is
equal to


F ( s)G( s)  e f (u ) g (t  u )dtdu

 st
(3)
0 u

u=t

Region of integration
u

Changing (reversing) the order of integration and noting that

  t

  dtdu    dudt ,
0 u 0 u

(See the above figure), we have the integral in (3) equal to



F ( s)G( s)  e f (u ) g (t  u )dtdu

s t

0 u

 t

e f (u ) g (t  u )dudt
s t
=
0 0


t 
= e   g (t  u ) f (u )du dt
s t

0 0 
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 131
 Applied Mathematics III


=  e  s t ( g  f )(t )dt
0


=  e  s t ( f  g )(t )dt . = L ( f  g ) .
0

Hence L ( f  g)  L ( f )L (g) . 

Note that if L 1 ( F (s))  f and L 1 (G(s))  g then

L 1 ( F (s)G(s))  f  g .

That is, L 1 ( L ( f ) L ( g ))  f  g .
This result can sometimes be used to find the inverse Laplace transform of a function that is a
product of factors whose inverse transforms are known
Note: It is no true that
L 1 F (s)G(s)  L 1 ( F ) L  1 (G) .

1 1 1
For example, if F ( s )  and G ( s )  2 then F ( s ) G ( s)  3
s s s
but

1 t
2

L 1
( FG)  L  3   ,
1
L 1 ( F )  1, L  1 (G)  t
s  2
and, clearly L 1 ( FG)  L 1 ( F ) L 1 (G) . 
Examples:

 1   s 
1. Compute a) L
1
 
2 
b) L
1
 2 
2 
 s( s  4)   (s 1) 

2. Solve the initial value problem

y  2 y  8 y  f (t ), y(0)  1, y(0)  0 .

Solution:
1. a) Since

 1  1  1 1 
L 1  
2 
 L 
 s ( s  4) 2 
 s( s  4)   
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 132
 Applied Mathematics III

and

1  1 
L 1    1, L 1    te 4t ,
2 
s  ( s  4) 
then

 1  1  1 1  t

L 
1
  L 
2 
  1 * te   xe 4 x dx
2 
4t

 s( s  4)   s ( s  4)  0

1 1 1
= te 4t  e 4t  . 
4 16 16
b) Since

 s   s s 
L 1  2   L 1  2
2 

 ( s  1)   s  1 s 2
 1 
and

 s   1 
L 1  2   cost , L 1  2   sin t ,
 ( s  1)   s  1 
then

 s   s s 
L 1  2   L 1  2
2 
  cost * sin t
 ( s  1)   s  1 s 2
 1 
t
=  cos x sin(t  x)dx
0

t sin t
= . 
t
2. Taking Laplace transforms
L  y  2 L ( y)  8 L ( y)  L ( f ) .
That is
s 2 L ( y )  s  2s L ( y )  2  8 L ( y )  L ( f ) .
Or
( s 2 - 2s - 8) L ( y)  L ( f )  s  2 .
Thus

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 133

 Applied Mathematics III

s2
L y 
1
L( f )  2
s  2s  8
2
s  2s  8

1 1 1 1 1 1 2 1
= L( f )  L( f )   .
6 s-4 6 s2 3 s4 3 s  2

Now using the convolution Theorem we obtain.

6
 1
6
  
y(t )  L  1 e 4 t L f   L  1 L(e  2t )L f   e 4 t  e  2 t
1 1
3
2
3

1 1 1 2
 e 4t  f (t )  e  2 t  f (t )  e 4 t  e  2 t .
6 6 3 3
4t 2 t
This solution is valid for any function f (t ) having a convolution with e and e 
2.5.2 Integral Equation
Although the convolution theorem is obviously very useful in calculating inverse transforms, it
also helps in solving certain integral equations, that is, equations in which the unknown function
y (t ) appears under the integral (and perhaps also out side of it).
Example: Solve
t

a) y (t )  1   y ( x)dx .
0
t
b) y(t )  sin 2t   y( x) sin 2(t  x)dx .
0

Solution:
t
a) y (t )  1   y ( x)  1 
0
y (t )  1

Taking transforms, we have

L ( y)  L (1)  L  y(t )  1
1
=  L ( y ) L (1) .
s
Thus
1 1
L ( y)   L ( y) .
s s

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 134

 Applied Mathematics III

Or
1
L ( y)  .
s 1
From this it follows that

 1 
y (t )  L - 1    e t . 
 s 1 
t
b) y (t )  sin 2t   y ( x) sin 2(t  x)dx  sin 2t  y (t )  sin 2t .
0

Taking transforms, we have

L ( y)  L (sin 2t )  L ( y) L (sin 2t ) .
2 2
  2 L ( y) .
s 4 2
s 4
Or
2
L ( y)  .
s 22

Thus

y(t )  2 sin 2t . 

Group Exercise 2.4

1. Use the convolution Theorem
a) To determine the Laplace transform of
t

e
2 t 2x
e cos(3x)dx.
0

b) To show that
1 
t x

L  2 (s)     f ( x)dx
1

s  0 0

2. Solve the initial value Problem

y  2  y  4 y  1, y(0)  y(0)  0
3. Use the convolution theorem to show that

t  F(s)  t

L   f ( x)dx   , where F (s)  L ( f ) .  H int :  f ( x)dx  (1 f )(t )
0  s  0

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 135

 Applied Mathematics III

4. Find L ( f ) if
t
a) f (t )   (t  x) 3 sin xdx
0

t
b) f (t )   e ( t  x ) cos 2 xdx
0

t
c) f (t )   e17( t  x ) x19 dx
0

5. Solve the following initial value problems by Laplace transform

1, t  
a) y  4 y  13 y  f ( x), y(0)  y(0)  0, where f (t )  
0, t  
2, t  
b) y  6 y  18 y  f ( x), y(0)  y(0)  0, where f (t )  
0, t  
6. Find the inverse Laplace transform of each of the following functions by using
convolution.
2 1 1
a) b) c)
s ( s 2  4)
2
( s  1) ( s 2  9)
2
( s  1) 2
2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 136

 Applied Mathematics III

2.6 Unit Summary

 The Laplace transform of a given function f (t ) , that is defined for all t  0 , is defined
by


L ( f )  F ( s)  e
 st
f (t )dt
0

for all s such that the integral exists.

 Suppose L[ f ](s)  F (s) and L[ g ](s)  G(s) are defined for s  a , and a, and b are
real numbers. Then
La  f t   b  g t   a  L f t   b  Lg t  for s  a .

and
L1 c1 F1 s   c2 F2 s   c1 L1 F1 s   c2 L1 F2 s 
 Let L[ f ](s)  F (s) for s  b  0 . Let a be any number. Then

 
L e at f t   F ( s  a) for s  a  b and L1F s  a   eat f t 

 Let L[ f ](s)  F (s) for s  b . Then L[ H (t  a) f (t  a)](s)  e  as F (s) for s  b and

L[ H (t  a) f (t )](s)  e  as L f t  a .and L1[e  as F (s)]  H (t  a) f (t  a)

 The unit step function (or Heaviside function) u(t  a) is defined by

0, t  a
u (t  a)   (a  0)
1, t  a
 Then the Laplace transform of the derivative f (t ) exists when s  , and

L  f   s Lf   f (0) .
 If f and g are piecewise continuous functions, then the convolution of f and g ,
t
written f  g , is defined by  f  g t    f ( x) g (t  x)dx .
0

 If L ( f ) and L (g ) exist, then L  f  g  exists and

L  f  g   L ( f )L (g) .

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 137

 Applied Mathematics III

2.7 Review Problems

1. Find the Laplace transform of the following functions
1  cos au

t
a) g (t )  du, a is a cons tan t. b) f (t )  et cosh 3t
0
u
2. Solve the following initial value problems by Laplace transform
a) y  4 y  0, y(0)  1, y(0)  2
b) y  4 y  4 y  cos t , y(0)  1, y(0)   1.
c) y  5 y  4 y  e2t , y(0)  1, y(0)  0
d) y  2 y  3 y  6e2t , y(0)  2, y(0)   14

3. Find  
a) L sin t
2

b) L cos
2
t 

c) L  te sin t  t
 2
d) L t cos3t 
4. Find f (t ) if L ( f ) is given

9  s 1  1  s 1 
 
s 2  s 2  9 
a) b)
s2  s 1 

s4 4
c) d)
s2  4 s  2s 2
3

1 1
e) f)
( s  3)3 s5  s3
1 e 4 s s2
g) h)
s5 s2  3
5. Using the Laplace transform, solve the following problems

a) 4 y  4 y  37 y  0, y(0)  3, y(0)  10

1, 0  t  1,
b) y  y  f (t ), y(0)  1, y(0)  0, where f (t )  
0, t  1,
 0, 0  t  4
c) y  4 y  f (t ), y(0) 1, y(0)  0 ,where f (t )  
 3, t  4

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 138

 Applied Mathematics III

d) y  4 y  5 y   (t 1), y(0)  0, y(0)  3

e) y  5 y  6 y  u(t 1)   (t  2), y(0)  0, y(0)  1
6. Find L ( f ) if
t t
a) f (t )   e15 ( t  x ) x 25dx b) f (t )   (t  x) 4 x 7 dx
0 0

7. Solve the following integral equations, using the Laplace transformation a)

t

a) y (t )  sin t   y( x) sin(t  x)dx

0

b) y (t )  t   sin(t  x) y ( x)dx
3

8. Find the inverse Laplace transform of each of the following functions by using convolution

ea s a s
d) e) f)
s ( s  2) s (s  a 2 )
2 2
(s   2 ) 2
2

e  3s 3 s
g) h) j)
s3 s 4 ( s 2 1) ( s 2  2) 2

9. Solve the following integral equations, using the Laplace

transformation.
t
a) y (t )  1   (t  x) y ( x)dx
0

t
1
6 0
b) y (t )  t  (t  x) 3 y ( x)dx

t
c) y (t )  2t  4  y ( x) sin 2(t  x)dx
0

t
d) y (t )  1 sinh t   (1  x) y (t  x)dx
0

t
e) y (t )  e  2 cos (t  x) y ( x)dx
t

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 139

 Applied Mathematics III

UNIT III

3. FOURIER SERIES; FOURIER INTEGRALS AND

FOURIER TRANSFORMATIONS

Unit Introduction
This unit deals with Fourier series and Fourier integrals; and Fourier transformations. The unit is
divided into three sections. The first section presents periodic functions and trigonometric
Fourier series. In the second section we will deal with the Fourier series. Fourier Integrals and
The Fourier Transforms are treated in the third and fourth sections respectively.. By doing so
students will be able to express terms and concepts related to Fourier series and integrals; and
Fourier transformations.
Unit Objectives:
At the end of this unit students will be able to:
 Define Fourier series, Fourier integrals, Fourier transformations
 Identify Fourier series of even and odd functions
 Demonstrate how to differentiate Fourier series of even and odd functions
together with solving exercises
 Distinguish Fourier series and Fourier integrals; Fourier transformation and
demonstrate together with solving exercises.
 Compute Fourier cosine transform, Fourier sine transform and Fourier transform of a
function.
 Demonstrate the applications of Fourier series, Fourier integrals, Fourier and
transformations together with solving exercise.
 Apply the concept of Fourier series, Fourier integrals, Fourier and Laplace
transformations in solving real life problems.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 140

 Applied Mathematics III

3.1 Periodic and Orthogonal Funcions

3.1.1. Periodic Functions and Trignometric Series

Definition:
A function f is said to be periodic if it is defined for all real number x and if there is
some positive number p such that

f ( x  p)  f ( x) for all x. (1)

The number p is called a period of f .

Remark: If a periodic function f has a smallest period p  0 it is known

as the primitive period of f (or the fundamental period of f ).

The graph of such functions is obtained by periodic repetition of its graph in any interval of
length p.

y
T

Figure 3.1-1 A T-periodic function

Examples:
1. f ( x)  sin x, g ( x)  sin 2 x are periodic functions with 2 and  their primitive
periods, respectively.
2. f ( x)  c = constant is a periodic function with a period any positive real number p .
Remarks:
1. If p is a period of periodic function f then 2 p, 3 p, 4 p, . . . are also periods of f ,

since from (1) we have

f ( x  2 p)  f (( x  p)  p)  f ( x  p)  f ( x) ,
f ( x  3 p)  f (( x  2 p)  p)  f ( x  2 p)  f ( x), etc.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 141

 Applied Mathematics III

and for any integer n ,

f ( x  np)  f ( x) , for all x .
Thus 2 p, 3 p, 4 p, . . . are periods of f .

2. If f and g have period p , then the function

h( x)  af ( x)  bg( x), (a, b cons tan ts)
also has the period p .
3. The definite integral of a T-periodic function is the same over any interval of
length T. If f has the period T then
a T T

 f t dt   f t dt
a 0

Example 3.1-1 will use this property to integrate a 2-periodic function shown in Figure 3.1-2.
Example 3.1-1. Let f be the 2-periodic function and N is a positive integer. Compute
N
N
f 2 ( x )dx if f(x) =  x + 1 on the interval 0  x  2

y
1

x
-3 -2 -1 1 2 3

-1
Figure 3.1-2. A 2-periodic function.

Solution

N  N 2  N 4 N

N
f 2 ( x )dx = N
f 2 ( x )dx + 
 N 2
f 2 ( x )dx +  + 
N 2
f 2 ( x )dx

2
N N 2  1 
 = N f ( x )dx = N  (  x  1) dx = N  (  x  1) 3 
2 2 2
f ( x )dx
N N 2 0
 3 0
N N 2

N
f 2 ( x )dx =
3
[ 1  1] = N
3

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 142

 Applied Mathematics III

The most important periodic functions are those in the (2-period) trigonometric system

1, cos x, cos 2x, cos 3x, , cos mx, ,

sin x, sin 2x, sin 3x, , sin nx, ,
Since
1, cos x, sin x, cos 2 x, sin 2 x,  , cos nx, sin nx, ,
are periodic functions with period 2 , the series that will arise in this connection will be of the
form
a0  a1 cos x  b1 sin x  a2 cos 2 x  b2 sin 2 x    an cos nx  bn sin nx  

 a0   (an cos nx  bn sin nx) , (2)
n 1

where a0 , a1 , a2 , . . ., b1 , b2 . . . are real constants.

Such a series is called a trigonometric series, and a n and bn are called the coefficients of the
series.
Note that each term of the series (2) has the period 2 . Hence if the series (2) converges, its
sum will be a function of period 2 .

Activity1.1
1. Find the smallest positive period p of

cos x, sin x, cos 2 x, sin 2 x, cosx, sin x, cos 2x, sin 2x
2. If f (x) and g (x) have period p, show that h( x)  af ( x)  bg( x) ,
(a, b cons tan ts) also has the period p .
3. Show that the function f ( x)  c , where c is constant is periodic function of period p
for every positive p .

4. If f is periodic function of period p , show that f (ax) , a  0 , is a periodic function of

period p a , and f ( x b) , b  0 , is periodic function of x of period bp .Verify these results for

f ( x)  cos x, a  b  2 .

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 143

 Applied Mathematics III

3.1.2. Orthogonal functions

To discuss Fourier series, we need the following integrals. If m and n are integers, then

 cos mx cos nxdx

= 0 for m  n

=  for m = n
= 2 , m  n  0

 sin mx sin nxdx

= 0 for m  n

=  for m = n
= 0, m  n  0

 cos mx sin nxdx

= 0 for all m and n
b
Definition: If 
a
f ( x ) g ( x )dx = 0 then f and g are said to be orthogonal over the interval [a, b].

In general, if any two members ψn, ψm of a set of functions {ψi} satisfy the condition
b

  x  x dx  0
a
n m

then ψn and ψm are said to be orthogonal, and (1.7) is known as the orthogonal condition in the
interval between a and b. The set {ψi} is an orthogonal set over the same interval.
Thus if the members of the set of trigonometric functions are
1, cos x, sin x, cos 2x, sin 2x, cos 3x, sin 3x, . . . ,
then this is an orthogonal set in the interval from −π to π.

3.2. Fourier Series of Functions

One of the most useful tools of mathematical analysis is Fourier series, named after the French
mathematical physicist Jean Baptiste Joseph Fourier (1768–1830). Fourier analysis is ubiquitous
in almost all fields of physical sciences.
In 1822, Fourier in his work on heat flow made a remarkable assertion that every function f(x)
with period 2π can be represented by a trigonometric infinite series of the form

f  x   a0   (a n cos nx  bn sin nx)
n 1

It is a tool in abstract analysis and electromagnetism and statistics and radio communication.
People have even tried to use it to analyse the stock market. (It didn't help.) The representation of
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 144
 Applied Mathematics III

musical sounds as sums of waves of various frequencies is an audible example. It provides an

indispensable tool in solving partial differential equations, and this chapter will show some of
these tools at work.
3.2.1. Fourier Series of Functions with Periodicity 2
The Fourier Series Coefficients
Fourier series arise from the practical task of representing a given periodic function f (x) in
terms of cosine and sine functions. These series are trigonometric series whose coefficients are
determined from f (x) by the Euler formulas, which we drive below.

Assume that f (x) is periodic function of period 2 is integrable over a period and can be
represented by a trigonometric series,

f ( x )  a0 
n 1
 (a n cos nx  bn sin nx) ; (1)

that is, we assume that this series converges and has f (x) as its sum. Given such a function

f (x) ,we want to determine the coefficients a n and bn of the corresponding series (1) . Because
of orthogonality between the trig functions, one has an easy way to evaluate the a n and bn in
equation (1).
Determination of a0 .

Integrating on both sides of (1) from   to , we get

 
 

 f ( x)dx    a0   (a n cos nx  bn sin nx)  dx
 n 1 
 
  

 a0  dx    n

n  1
a cos nxdx  bn  snnxdx

   
 2 a0 ,
if term-by-term integration of the series is allowed.
Thus

1
a0 
2  f ( x)dx .


Determination of a n .

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 145

 Applied Mathematics III

Multiplying (1) by cosmx, where m is any fixed positive integer, and integrating from
  to  :
 
 

 f ( x) cos mxdx    a0   a n cos nx  bn sin nx cos mxdx
 n 1 

   

 ao  cos mxdx    an  cos nx cos mxdx  bn  sin nx cos mxdx  ,
 n  1   
by integrating term-by term.
a  if n  m
 m .
0 if n  m
Thus

1
am 
  f ( x) cos mxdx,

m  1, 2,  .

Determination of bn .

Multiplying (1) by sin mx, where m is any fixed positive integer, and then integrating from
  to  :
 
 

 f ( x) sin mxdx    a0   a n cos nx  bn sin nx sin mxdx
 n 1 

   

 a0  sin mxdx   an  cos nx sin mvdx  bn  sin nx sin mxdx
 n  1   
b  , n  m
 m .
0, otherwise
Thus

1
bm 
  f ( x) sin mxdx,

m  1, 2,  .

Writing n in place of m , we obtain the Euler formulas

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 146

 Applied Mathematics III

1
a0   f ( x)dx
2 

1
an 
  f ( x) cos nxdx,

n  1, 2,  , (2)

1
bn 
  f ( x) sin nxdx,

n  1, 2,  ,

These numbers, given by (2) , are called the Fourier coefficients of f (x) .

Definition:
The trigonometric series

a0 
n 1
 a n cos nx  bn sin nx

with coefficients given by (2) is called the Fourier series expansion of f (x) .

Since the series in (1) hold good for any function of period 2 , the above formula also hold
good for the range 0 to 2 .
Example 1: Find the Fourier series of

 k if    x  0
f ( x)   f ( x  2 )  f ( x), k  0
 k if 0  x  ,

Solutions:
y

k
x
-3 -2 - 2 3
-k

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 147

 Applied Mathematics III

Since the value of f (x) at a single point does not affect the integral, we can leave f (x)

undefined at x  0 and x    .

From (2) we obtain

 
1 1  0

a0 
2 

f ( x)dx 
2  

 
k dx  k 
0
dx  0 .

 
1 1 0

an 
 

f ( x) cos nxdx   
  
k  cos nxdx  k 
0
cos nxdx  0 .

 
1 1 0 
bn   f ( x) sin nxdx   k  sin nxdx  k  sin nxdx
     0 

1   cos nx   cos nx  
0 

 k  k
   n    n  0 

2k
 1  cos n 
n
 4k
 for n odd,
  n .

0 for n even
Hence
4k 4k 4k
b1  , b2  0, b3  , b4  0, b5  .
 3 5
and therefore, the Fourier series of f (x) is

4k  1 1 
 sin x  sin 3x  sin 5 x  .
  3 5 
4k sin(2n  1)

 
 n  1 2n  1
. 

Example 2: Find the Fourier series of f x     x,0  x  2 where f x  2   f x  .

Solution: The coefficients are:

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 148

 Applied Mathematics III

2 2
1 1 1 1 2  2
a0 
 0 f(x) dx 
 0 (  - x ). dx 
 
x - x
2  0
1
 (2 2  2 2 )  0

2 2
1 1 nx  x cos n
an 
 0 f(x) dx 
 0 (  - x ).cos 
dx
1
-1 sin nx
2 n
1

 0 (  - x ).cos nx dx -1
0 cos nx
n2

1   x 1  2
  sin nx - cos nx 
  n n2  0

1
 (1  1)  0
n 2
2 2
1 1 nx
bn 
 0 f(x) dx 
 0 (  - x ).sin 
dx

2
1   x sin nx

 0 (  - x ).sin nx dx -1
-1 cos nx
n
1   x 1  2
  cos nx - sin nx  -1
  n n 2  0
0
n
2
sin nx

1 2
 ( 2 ) 
n n
and therefore, the Fourier series of f (x) is

 sin 2 x 1 1 
2 sin x   sin 3x  sin 4 x  .
 2 3 4 


sin nx
 2 . 
n  1 nx

Quick Check Class Exercises

1. Obtain a Fourier expansion for 1 cos x in the interval    x   .

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 149

 Applied Mathematics III

2. Find the Fourier series of the function

0 if    x  0
a. f ( x)   f ( x  2 )  f ( x),
 1 if 0  x  ,

0 if    x  0
b. f ( x)   f ( x  2 )  f ( x),
 t if 0  x  ,

2 if    x  0
c. f ( x)   f ( x  2 )  f ( x),
 1 if 0  x  ,

0 if    x  0
d. f ( x)   f ( x  2 )  f ( x),
 sin t if 0  x  ,
Instructor's role
Check and give feedback to their answers

We must be careful at this stage not to conclude that we have proved that every periodic function
f (x) has a Fourier expansion that converges to it. The convergence problem can be solved by
the famous theorem of Dirichlet.

3.2.2 Convergence and Sum of Fourier Series

Definition:
A function f (x) is piecewise continuous on a finite interval a  x  b if it is defined
on that interval and is such that the interval can be subdivided into finitely many intervals in
each of which f (x) is continuous and has finite limits as x approaches either endpoint of the
interval of subdivision from the interior.

It follows from this definition that finite jumps are the only discontinuities that a piecewise
continuous function may have.
Example 1: Consider the function
 4,    x  0
f ( x)  
 4, 0 x

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 150

 Applied Mathematics III

It is a piece wise continuous and has a jump at x  0 . Its left - hand limit there is  4 and its
right - hand is 4 . So the average of these limits is 0 . Thus the Fourier series of f converge to
this value when x  0 because then all its terms are 0 . Similarly for other jumps.
Definition:
A function f t  is said to satisfy the Dirichlet conditions on an interval I if and only if

f t  is bounded and has at most a finite number of local maxima and minima and a finite
number of discontinuities on I.

Theorem: (Representation by a Fourier series)

If a function f is bounded periodic function with period 2 which in any one period
has at most a finite number of local maxima and minima and a finite number of points of
discontinuity ( piecewise continuous) in the interval [ ,  ] and has a left - hand and

right-hand derivatives at each point of that interval, then the Fourier series (1) of f

(with coefficients (2) ) is convergent. Its sum is f (x) , except at a point x0 at which

f (x) is discontinuous and the sum of the series is the average of the left and right-hand
limits of f (x) at x0 .

Proof: Exercise
The conditions of Theorem above under which a periodic function possesses a valid Fourier
expansion are referred to collectively as the Dirichlet conditions. Note that this theorem gives a
sufficient condition for a Fourier series to converge to f .
Example 1: Obtain the Fourier expansion of

f x     x  in - < x < 
1
2
Solution: We have,

1  1 1
   2 (  x)dx
a0  f ( x)dx 
  


1  x2 
= x   
2  2  

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 151

 Applied Mathematics III

 
1 1 1
an 
 

f ( x) cos nxdx 
  2 (  x) cos nxdx


Here we use integration by parts, so that


1    cos nx 
  x 
sin nx
an   (1) 
2  n  n 2  


1
0  0
2


1 1
bn 
 
 2 (  x) sin nxdx

  cos nx   sin nx 
  x  n
1
  (1) 
2   n
2
  
(1) n

n
Using the values of a0 , an and bn in the Fourier expansion, since f x  is continous

a0  
f ( x)    an cos nx   bn sin nx
2 n 1 n 1

we get,

(1) n 
f ( x)    sin nx
2 n 1 n
This is the required Fourier expansion of the given function.
Example 2. Obtain the Fourier expansion of
  ,  x  0
f(x) = 
 x,0  x  
Deduce that
2 1 1
 1   ......
8 32 52

Here,

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 152

 Applied Mathematics III


1  
0
a0     dx   xdx   
   0  2

1 
0
an     cos nxdx   x cos nxdx 
   0 

1
n
2

(1) n  1 

1 
0
bn     sin nxdx   x sin nxdx 
   0 

1
n
1  2(1) n 
Fourier series is

f(x) =
 1  1
  (1)  1 cos nx  
n



1  2(1) n 
sin nx

4  n 1 n 2 n 1 n
Note that the point x=0 is a point of discontinuity of f(x). Here f(x+) =0, f(x-)=- at x=0. Hence

[ f ( x  )  f ( x  )]  0    
1 1
2 2 2
The Fourier expansion of f(x) at x=0 becomes
  1  1
   [(1) n  1]
2 4  n 1 n 2
2 
1
or  2
[(1) n  1]
4 n 1 n
Simplifying we get,
2 1 1
 1   ......
8 32 52
Quick Check Exercises
Q1 Find the Fourier series to represent the function f (x) , given by

x for 0  x  
f ( x)  
2  x for   x  2
1 1 1 2
Deduce that    ............ 
12 2 2 32 8
Q2 Obtain the Fourier expansion

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 153

 Applied Mathematics III

  , if    x  0 2 
1
f ( x)   and hence deduce that  1
 x, if 0 x  8 n 1 ( 2n  1)
2

0, if    x  0
Q3 If f ( x)   . Prove that
 sin x, if 0 x 

1 sin x 2  cos 2mx

f ( x)     . Hence show that
 2  m1 4m 2  1
1 1 1 1
   .............  (  2) .
1.3 3.3 5.7 4
Instructor's role
Check and give feedback to their answers

3.2.3 Fourier Series of Functions of Any Period p  2L

The functions considered so far had period 2 , for simplicity. In applications periodic functions
will generally have other periods. But the transition from period p  2 to period p  2 L is
quite simple. It amounts to a stretch (or contraction) of a scale on the x - axis. If a function
f (x) of period p  2L has a Fourier series, we claim that this series is


n n
f ( x)  a0   (a
n 1
n cos
L
x  bn sin
L
x) (1)

with the Fourier coefficients of f (x) given by the Euler formulas:

1 L
2 L L
(a) a0  f ( x)dx

1 L n
(b) an  
L L
f ( x) cos
L
xdx n,  1, 2, 3, . (2)

1 L n
(c ) bn   f ( x) sin xdx, n  1, 2, 3, .
L L L

Remark: We may replace the interval of integration by any interval of length p  2 L , for

example, by the interval 0  x  2L .

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 154
 Applied Mathematics III

Example 1: Find the Fourier series of the function

0 if  2  x   1

a) f ( x)  k if  1  x  1 , p  4
0 if 1  x  2


b) f ( x)  0 if  2  x  0 , p 4
1 if 0 x2

Solution:
y
a)

x
-3 -2 -1 1 2 3

Since p  2L  4, then L  2 and

12 1  1 1 2
 k
a0   f ( x)dx    0dx   kdx   0dx  ,
4 2 4 2 1 1  2

12 nx k1 nx 2k n
an  
2 2
f ( x ) cos
2
dx  
2 1
cos
2
dx 
n
sin
2
,

12 nx k1 nx
bn   f ( x) sin dx   sin dx  0 .
2 2 2 2 1 2
Then the Fourier series is
n
sin
k 2k  2 cos nx
f ( x)  
2

 n 1 n 2

k 2k   1 3 1 5 
   cos x  cos x  cos x     . 
2   2 3 2 5 2 
b) Since p  2L  4 then L  2 and

12 12 1
a0   f ( x)dx   dx  ,
4 2 40 2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 155

 Applied Mathematics III

12 nx 12 n
an  
2 2
f ( x ) cos
2
dx  
20
cos
2
xdx  0,

 2
12 nx 12 n 1 (1) n  , n  odd
bn   f ( x) sin dx   sin xdx    n .
2 2 2 20 2 n  0, n  even

Then the Fourier series is

 sin(2n 1) x
1 2
f ( x)    2 . 
2  n 1 2n  1

Example 2. Find the Fourier series of the function

8 if 0  x  2
f ( x)   p 4
 8 if 2  x  4

Solution : Y
8

-4 -2 0 2 4 X
-8

1  
4 2 4
1
a0 
2 
0
f ( x)dx   8dx    8dx   0 .
2  0 2 

1 
4 2 4
1
an 
 
0
f ( x) cos nxdx  
 0
8 cos nxdx  
2
8 cos nxdx   0 .


1 
4 2 4
1
bn 
 
0
f ( x) sin nxdx   8 sin nxdx   8innxdx 
 0 2 

1   cos nx   cos nx  
2 4

 8  8
   n  0  n  2 

2k
 1  cos n 
n

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 156

 Applied Mathematics III

 4k
 for n odd,
  n .

0 for n even
Hence
4k 4k 4k
b1  , b2  0, b3  , b4  0, b5  .
 3 5
and therefore, the Fourier series of f (x) is

4k  1 1 
 sin x  sin 3x  sin 5 x  .
  3 5 
4k sin(2n  1)

 
 n  1 2n  1
. 

Quick Check Exercises

Find the Fourier series of the periodic function f (x) of period p  2 L .

 1 if  1  x  0
a) f ( x)   , p2
 1 if 0  x  1
b) f ( x)   x,  2  x  2, p4
Instructor's role
Check and give feedback to their answers

Group Activity 1.2

1. Find the Fourier series of the function f (x) , which is assumed to have the period 2 ,
where

y y
a) f (x) equals b) f (x) equals
k
1
x x

c) f ( x)  x,    x   d) f ( x)  x 2 ,    x  

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 157

 Applied Mathematics III

e) f ( x)  x, 0  x  2 f) f ( x)  x  x ,    x  

     
 1 if 2  x  2  x if 2  x  2
g) f ( x)   h) f ( x)  
 1 if   x  3 0 if   x  3
 2 2  2 2
2 1 1 1
i) f ( x)  x ,    x   . Show that 1    
8 9 25 49
2. Find the Fourier series of the periodic function f (x) of period p  2 L .

 x if  1  x  0
a) f ( x)   , p2
 x  2 if 0  x  1

0 if  2  x  0
b) f ( x)   , p4
2 if 0  x  2
 x if 0  x  2
c) f ( x)   , p4
3 if 2  x  4
d) f ( x)  x ,  2  x  2, p4

e) f ( x)  3x 2 ,  1  x  1, p2

1 1
  x if   x  0

f) f ( x)   2 2 , p 1
1
  x if 0  x  1

2 2
3. Find the Fourier series of the periodic function that is obtained by passing
the voltage v(t )  vo cos100 t through a half - wave rectifier.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 158

 Applied Mathematics III

3.2.4 The Fourier Series of Even and Odd Functions

Unnecessary work and corresponding source of errors in determining Fourier coefficients of a
function can be avoided if the function is odd or even.

Definition:
A function f is said to be even if f ( x)  f ( x) , for all x .

Example: f ( x)  cos nx is an even function.

Note that a graph of even function is symmetric with respect to y -axis.

-L X
L

Even function

Definition:
A function g is said to be odd if g ( x)   g ( x), for all x .

Example: g ( x)  sin nx is an odd function.

X
-L L

Odd function
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 159
 Applied Mathematics III

Remarks:
1. If f is an even function on [ L, L] , then
L L

 f ( x)dx  2 f ( x)dx
L 0

2. If f is an odd function on [ L, L] , then

L

 f ( x)dx  0
L

3. If f (x) is given over [0, L] , then we can extend it as even function:

 f ( x), 0  x  L
g ( x)  
 f ( x),  L  x  0
Similarly we can extend it as odd function:

 f ( x), 0  x  L
h( x )  
 f ( x),  L  x  0.
4. The product q  fg of an even function f and an odd function g is odd. Hence if

 n 
f (x) is even, then f ( x) sin x  is odd, which implies
 L 
 n 
L
1
bn 
L  f ( x) sin
L
x dx  0 .
L 
1 L  n 
Similarly, if f (x) is odd then a0  
2L L
f ( x)dx  0 and f ( x) cos x  is odd, which intern
 L 
implies
1 L  n 
a n   f ( x) cos x dx  0 .
L L  L 
Thus we proved the following theorem.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 160

 Applied Mathematics III

Theorem 1: (Fourier cosine series and Fourier sine series)

The Fourier series of an even function f (x) of period p  2L is a Fourier cosine series

n
f ( x)  a0  a
n 1
n cos
L
x, (1)

with coefficients
1L 2L n
L 0 
a0  f ( x)dx, an  f ( x) cos xdx, n  1, 2 ,  .
L0 L
The Fourier series of an odd function f (x) of period p  2L is a Fourier sine series


n
f ( x)  b
n 1
n sin
L
x,

with coefficients
2L n
bn  
L0
f ( x) sin
L
xdx, n  1, 2 ,  .

In particular, the Fourier series of an even function f of period 2 is a Fourier cosine series

f ( x)  a 0  a
n 1
n cos nx

with coefficients
1 2
a0 
  f ( x)dx,
0
an 
  f ( x) cos nxdx,
0
n  1, 2,  .

Similarly, the Fourier series of an odd function f of period 2 is a Fourier sine series

f ( x)  b
n 1
n sin nx ,

with coefficients
2
bn 
  f ( x) sin nxdx,
0
n  1, 2,  .

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 161

 Applied Mathematics III

Theorem 2: (sum of functions)

The Fourier coefficients of a sum f1  f 2 are the sums of the corresponding Fourier

coefficients of f 1 and f 2 . The Fourier coefficients of cf are c times the corresponding

Fourier coefficients of f .

Proof: Exercise
Example 1: Find the Fourier series of the function
a) f ( x)  x if    x   and f ( x  2 )  f ( x)

b) f x   x,  L  x  L
Solution:
a) Since
f ( x)   x   f ( x), for all x,
thus f is odd and hence

a0  an  0,

2 cos n 2
bn 
  x sin nxdx   2
0 n
 (1) n  1 .
n
Hence the Fourier series is

(1) n  1
f ( x)  2  sin nx
n 1 n

 1 1 
 2 sin x  sin 2 x  sin 3x     . 
 2 3 
b) Since this is an odd function, the Bns will = 0.

1 L  n x  2 L  n x 
An 
L  L
x  sin 
 L 
 dx 
L 0
x sin
 L 
dx

n x
Let  y,
L

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 162

 Applied Mathematics III

2 n  L   L  2L n

L 0  n  
An   y  sin y dy   y sin y dy
  n  n  2 0

n

2L
sin y  y cos y
n 2
0


2L
n 2
0  n  1  0  0  n2L   1
n n 1

n 1

 n x 
 f x   x    1
2L
sin 
n 1 n  L 

Graph of 1st term and the sum of the first 4 terms of equation.
Example 2. Obtain the Fourier expansion of f(x) = x2 over the interval (-, ). Deduce that
2 1 1
 1   ......  
6 22 32
Solution:
The function f(x) is even. Hence
 
1 2
a0 =
 
 f ( x)dx =   f ( x)dx 0

 
2 2  x3 
=  x dx    2

0   3 0

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 163

 Applied Mathematics III

2 2
a0 
or 3


1
an 
 
 f ( x) cos nxdx

2
=
  f ( x) cos nxdx,
0
since f(x)cosnx is even


2
x
2
= cos nxdx
 0

Integrating by parts, we get


2   sin nx    cos nx    sin nx 
an   x 2    2 x   2 
  n   n
2
  n
3
 0
4(1) n

n2


1
Also, bn 
  f ( x) sin nxdx  0

since f(x)sinnx is odd.

Thus
2 
(1) n cos nx
f ( x)   4
3 n 1 n2
 2 
1
2   4 2
3 n 1 n

1 2
1 n 2  6
2 1 1
Hence,  1   .....
6 22 32
3. Obtain the Fourier series of f(x) = 1-x2 over the interval (-1,1).

Solution:
The given function is even, as f(-x) = f(x). Also period of f(x) is 1-(-1)=2

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 164

 Applied Mathematics III

Here
1 1
1
a0 =  f ( x)dx = 2 f ( x)dx
1 1 0

1
1
 x3 
= 2  (1  x )dx  2 x  
2

0  3 0

4

3
1
1
1 1
an  f ( x) cos(nx)dx

1
 2 f ( x) cos(nx)dx as f(x) cos(nx) is even
0

1
= 2 (1  x 2 ) cos(nx)dx
0

Integrating by parts, we get

1
  sin nx    cos nx    sin nx 
an  2 1  x 2     (2 x)   (2) 
 n   (n )   (n )  0
2 3


4(1) n1
=
n 2 2
1
1
bn   f ( x) sin(nx)dx =0, since f(x)sin(nx) is odd.
1 1
The Fourier series of f(x) is
2 4  (1) n 1
f(x) =  2  cos(nx)
3  n 1 n 2
Quick Check Class Exercises
1. Obtain the Fourier series for the function f(x) given by
 2x
1   , if    x  0
a) f ( x)  
 1  2 x , if 0  x  
 
Hence deduce that

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 165

 Applied Mathematics III

1 1 1 2
   ............ 
12 32 5 2 8
b) f ( x)   x,  2  x  2, p4
c) f ( x)  1  x ,  2  x  2, p4

Instructor's role
Check and give feedback to their answers

3.2.5 Half Range Expansions

In various physical problems there is a practical need to use Fourier series in connection with
function f that is defined on some interval, say 0  x  L , as in ( fig .a) ,below. This function

f can be the displacement of a violin string (undistorted) of length L or the temperature in a

metal bar of length L , and so on. On this half interval we want to represent f by a Fourier
series. If f (x) is given on some interval only, say, 0  x  L. (1) We can expand f (x) periodically
with period L, this is full range expansion. (2). We can do better this by choosing the period
p  2L because then and always get a cosine series by first extending f (x) from 0  x  L as
even function on the range  L  x  L ( fig .b) ,below,, and then extend this new function as a
periodic function of period 2L and, since it is even, represent it by a Fourier cosine series. (3) Or
we can extend f (x) from 0  x  L as an odd function on  L  x  L , and then extend this new
function as a periodic function of period 2L ( fig .c) ,below. and, since it is odd, represent it by a

Fourier sine series. These two series are called the two half - range expansions of f , which is
given only on “ half the range” (half the interval). The form of these series is given in Theorem
1 (sec. 1.2.4).

X
-L L

, the given function

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 166
 Applied Mathematics III

X
-L L

, extended as an even periodic function of period

X
-L L

, extended as an odd periodic function of period

Observe the above figures.
(a) Function f (x) given on an interval 0  x  L
(b) Even extension to the full interval  L  x  L (heavy curve) and the periodic extension
of period 2 L to the x-axis.
(c) Odd extension to the full interval  L  x  L (heavy curve) and the periodic extension
of period 2 L to the x-axis.

Example1
(d) : Represent f (x) by a Fourier series, f (x) = x, 0 < x < L.

Solution: (1) Full range expansion y

L
x
2L L L 2L

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 167

 Applied Mathematics III

a0  2nx 2nx
Let f ( x )    ( an cos  bn sin )
2 n 1 L L
2 L 2 L
L 0
a0  f ( x )dx   xdx  L
L 0

2 L 2nx 2 L 2nx
an  
L 0
f ( x ) cos
L
dx   x cos
L 0 L
dx

2nx 2nx 2nx

L L
1 L L
 ( x sin  sin dx )  2 2 cos 0
n L 0
0 L 2n  L 0

2 L 2nx 2 L 2nx
bn  
L 0
f ( x ) sin
L
dx   x sin
L 0 L
dx

2nx 2nx 2nx

L L
1 L L L L
 ( x cos  cos dx )    2 2 sin 
n L 0
0 L n 2n  L 0 n
L L 
1 2nx
f ( x)    sin
2  n 1 n L

1
n =3
n =5
0.8

0.6

n =1
0.4

0.2

0
0 0.2 0.4 0.6 0.8 1

y
L
(2)Fourier cosine series x
2L L L 2L

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 168

 Applied Mathematics III

a0  nx
Let f ( x )    a n cos
2 n 1 L
2 L
L 0
a0  xdx  L

nx nx nx

L
2 L 2 L
a n   x cos dx  ( x sin   sin dx )
L 0 L n L 0
0 L
nx
L
2L 2L 4L
 cos  [( 1) n  1]  
n 
2 2
L 0 n 
2 2
( 2n  1) 2  2
L 4L 
1 ( 2n  1) x
f ( x)   2  cos
2  n 1 ( 2n  1) 2
L

y
L
2L L L x
(3)Fourier sine series
2L
L


nx
Let f ( x )   bn sin
n 1 L
nx nx nx
L
2 L 2 L
bn   x sin dx   ( x cos   cos dx )
L 0 L n L 0
0 L
nx ( 1) n 1 2 L
L
2L 2L
 ( 1) n  2 2 sin 
n n  L 0 n
n 1
2L 
( 1) nx
f ( x)  
 n 1 n
sin
L

1
5 terms Cosine
0.8
Full range

0.6

0.4

Sine
0.2

0
0 0.2 0.4 0.6 0.8 1

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 169

 Applied Mathematics III

Example 2: Find the two half- range expansions of the function

f ( x)  k , 0  x  3, k  0 .

k
x

Solution: (a) Even periodic extension (Fourier cosine series).

From Theorem 1 we obtain
1L 13
L 0 3 0
a0  f ( x ) dx  kdx  k ,

2L n 23 n
an  
L0
f ( x ) cos
L
xdx  
30
k cos xdx  0 .
3
Hence the Fourier cosine series is f ( x)  k.

x
6

(b) Odd periodic extension (Fourier Sine series)

Similarly from Theorem 1 we obtain
 4k
2L n 23 n  , n  odd
bn   f ( x) sin xdx   k sin xdx   n .
L0 L 30 3  0 , n  even

Hence the Fourier sine series is

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 170

 Applied Mathematics III

4k 
1 (2n 1)
f ( x) 

 2n 1 sin
n 1 3
x

4k   1 1 5 
  sin x  sin x  sin x    .
  3 3 5 3 

k
x
-k

Quick Check Exercises

Q1 Obtain the half-range Sine series for f ( x)  2  x for 0 < x <2.
Hence deduce that
1 1 1 
   ............ 
1 3 5 4
Q2 Express
1 1
 4  x, if 0  x  2
f ( x)  
 x  3 , if 1  x 1
 4 2
as the Fourier Series of sine terms.
Q3 Prove that 0  x   ,
2 cos 2 x cos 4 x cos 6 x 
x(  x)   2    ............
6  1 2 2
3 2

Instructor's role
Check and give feedback to their answers

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 171

 Applied Mathematics III

3.2.6. Complex Exponential Fourier Series

Complex Fourier Series

a0  a 
e inx  e  inx e inx  e  inx
f ( x)    ( a n cos nx  bn sin nx )  0   ( a n  bn )
2 n 1 2 n 1 2 2i
a 0  a n  ibn inx a n  ibn inx 
   [( )e  ( )e ]  c0   ( cn e inx  c n e inx ]
2 n 1 2 2 n 1

where
a0 1 
2 2   
c0   f ( x )dx

a  ibn 1  1 
cn  n
2

2  f ( x )(cos nx  i sin nx )dx 
2   
f ( x )e inx dx

a  ibn 1  1 
cn  n
2

2   
f ( x )(cos nx  i sin nx )dx 
2   
f ( x )e inx dx

Hence

f ( x)  c e
n  
n
inx
 x

where
1 
2   
cn  f ( x )e inx dx

For a function of period of 2L, the complex Fourier series is

 nx

 cn e
i
f ( x)  L
 L  x  L (2.8)
n  

where

nx
1 L i
cn  
2 L L
f ( x )e L
dx

Example 1: Find the complex Fourier series of f (x) = ex,    x   if and f (x + 2  ) = f (x).

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 172

 Applied Mathematics III


Solution : Let f ( x)  c e
n  
n
inx
  x  

1  1  1  1
  f ( x )e  e e  e
inx inx (1in ) x
cn  dx  x
dx  dx  e (1in ) x
2  2  2  2 (1  in ) 

1  in
 [e (1in )  e (1in ) ]
2 (1  n )
2

1  in
 [e  (cos n  i sin n )  e  (cos n  i sin n )]
2 (1  n 2 )
1  in 1  in
 (1) n (e   e  )  (1) n sinh 
2 (1  n )
2
 (1  n )
2

sinh  
1  in
ex 

 (1  n
n  
2
)
(1) n e inx   x  

Example 2: Let f (x) = x for 0  x  2  , find the complex form of Fourier series of f (x) on the


Solution : Let f ( x)  c e
n  
n
inx
, where
2
1 2 1 2 1 x2 1 4 2
c0 
2  0
f ( x)dx 
2  0
xdx  
2 2
 
2 2

0

2 2 1  inx 2
  e inx dx 
1 1 1 2
cn 
2  0
f ( x)e inx dx 
2  0
xe inx dx    xe
2  in  0 0 
2
1 1  i 2 n e inx  1 1  e i 2 n  1  1 i
  2e     2    
2  in   in 0  2  in  in   in n
 

1
f ( x)    i  e inx , 0  x  2
n   n
n0

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 173

 Applied Mathematics III

Group Activity 3.2.4:

1. Are the following expressions even or odd?
a) Sums and products of even functions.
b) Product of even and odd functions.
c) Absolute value of odd functions
d) f ( x)  f ( x) and f ( x)  f ( x) , for arbitrary f (x) .
2. Find all functions that are both even and odd.
3. By stating whether the given function is even or odd, find its Fourier
series.
  
k if 2  x  2 1 1 1 
a) f ( x)   Show that 1       .
0 if   x  3 3 5 7 4
 2 2
1 2
b) f ( x)  x ,   x  
2
1 1 1 1 2
Show that i) 1      
4 9 16 25 6
1 1 1 2
ii) 1     
4 9 16 12
 2 x if    x  0
c) f ( x)  
 2 x if 0 x  x  
4. Give the Fourier cosine and the Fourier sine series of the function
a) f ( x)  1, 0  x  L d) f ( x)  e x , 0  x  L

b) f ( x)    x, 0 x e) f ( x)  e x , 0  x  2

c) f ( x)  x 2 , 0  x  L f) f ( x)  x, 0  x  L
1. a) Show that the complex Fourier series of a periodic function f
of period 2 is given by

f ( x)  c e
n  
n
inx
,

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 174

 Applied Mathematics III


1
where cn   f ( x)e dx, n  0,  1,  2, 
 inx

2 

is the complex Fourier coefficient.

b) Similarly, show that the complex Fourier series of a periodic function f of period 2 L is
given by
 in x

f ( x)  c
n  
n e L
,

where
in x
1 L 
cn  
2L L
f ( x) e L dx, n  0,  1,  2, 

is the complex Fourier coefficient.

c) Find the complex Fourier series of
i) f ( x)  e x ,    x   , f ( x  2 )  f ( x)

ii) f ( x)  x,    x   , f ( x  2 )  f ( x)

  1,    x  0
iii) f ( x)   , f ( x  2 )  f ( x)
 1, 0  x  

 0,    x  0
iv) f ( x)   , f ( x  2 )  f ( x)
 1, 0  x  
d) Show that the complex Fourier Coefficients of an even function are pure
imaginary.

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 175

 Applied Mathematics III

3.3. The Fourier Integrals

Fourier series are powerful tools in treating various problems involving periodic functions. But
since many practical problems involves non periodic functions, it is desirable to generalize the
method of Fourier series to include non periodic function. Thus for non-periodic functions, the
method of Fourier series is given by Fourier integral

Theorem 1. (Fourier integral)

If f (x) is piecewise continuous in every finite interval and has a right-hand derivative and a

left-hand derivative at every point and if f (x) is absolutely integrable, that is, the integral




f ( x) dx

exists, then f (x) can be represented by a Fourier integral,



f ( x)   A( ) cos x  B( ) sin  xd

0
(1)

where
 
1 1
A( ) 
  f (v) cos vdv,

B( ) 
 
f (v) sin  vdv (2)

with  taking all values. At appoint where f (x) is discontinuous the value of the Fourier

integral equals the average of the left - and right - hand limits of f (x) at that point.

Proof: Exercise.
Example: Find the Fourier integral representation of the function

 1,    x  0

f ( x)   1, 0x 
 0, x  

Solution: Since

1 
10 

A( )   f (v) cosvdv    (1) cosv dv   cos vdv  0 ,
    0 
Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 176
 Applied Mathematics III

and

1 
10 
B( ) 
 

f (v) sin  vdv 
 
( 1) sin  vdv  0 sin  vdv

2 1 cos 
 ,
 
the Fourier integral representation is
2  1 cos
f ( x) 
 
0 
sin xdw. 

3.3.1 Fourier Cosine and Fourier Sine Integrals

For an even or odd function the Fourier integral becomes simpler. The simplifications follow
immediately from our previous formulas.

Indeed, in Theorem 1 above, if f (x) is an even function, then B( )  0 in (2) , and

2
A( ) 
  f (v) cosvdv .
0
(3)

The Fourier integral (1) then reduces to the Fourier cosine integral


f ( x)   A( ) cos xdw ( f even). (4)

0

Similarly, if f (x) is odd, then in ( 2) we have A( )  0 and

2
B( ) 
  f (v) sin  vdv .
0
(5)

The Fourier integral (1) then reduces to the Fourier sine integral

f ( x)   B( ) sin  xd ( f odd). (6)
0

Observe that these simplifications are quite similar to those in the case of a Fourier series
discussed earlier.
Remark: The Fourier integral representations may also be used for evaluating integrals.
Examples:
1. Find the Fourier cosine and sine integrals of

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 177

 Applied Mathematics III

1, 0  x  1

f ( x)  2, 1  x  4
0, x  4

2. Find the Fourier integral representation of the function
 x
f ( x)  e ,
and evaluate the integral

1
1 
0
2
d .

Solution: 1.
(a) Fourier cosine integral.
Since

2 2 1 4

A( ) 
 
0
f (v) cosvdv    cosvdv  2 cosvdv
0 1 


2
2 sin 4  sin  .

the Fourier cosine integral representation is

2  2 sin 4  sin 
f ( x) 
 
0 
cos xd .

(b) Fourier sine integral.

Similarly, since

2 2 1 4

B( ) 
 
0
f (v) sin vdv    sin vdv  2 sin vdv 
0 1 
2
 (1 cos  2 cos 4 )

the Fourier sine integral representation is

2 1  cos   2 cos 4
f ( x) 
 
0

sin  xd . 

2. Since

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 178

 Applied Mathematics III

 x e  x , x  0
f ( x)  e   x
e , x  0
and
f ( x)  f ( x) , for all x ,

f ( x)  e
 x
is even and so B( )  0 .
Thus
 
2 2
 
A( )  f (v) cos vdv  e v
cos vdv
0 0

2 1
 .
 1  2
Therefore, the Fourier integral of f (x) is the Fourier cosine integral

2 cos  x
f ( x)   d .
 0 1  2
But since

2  d
f (0)  1   ,
 0 1  2
we obtain

d 
0 1   2 2 .
 

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 179

 Applied Mathematics III

Group Activity 3.3.1

1. Find the Fourier integral representation of the function

1, x 1 
sin 
b) f ( x)  
0, x  1
, and evaluate the integral 
0 
d .

 x,    x  
c) f ( x)  
 0, | x |  
 k ,  10  x  10
d) f ( x)   , k  cons tan t
 0, | x |  10

2. Find the Fourier cosine and sine integrals of f ( x)  e  x ,Show that


cos x  x
0 1   2 d 
2
e for x  0 .

3. Find the Fourier sine and cosine integral of

1 if 0  x  1  x if 0  x  a
a) f ( x)   c) f ( x)  
0 if x  a 0 if x  a
e x if 0  x  1
b) f ( x)  
0 if x  1
5. a) Show that the complex Fourier integral of a non periodic function f is given by


 c( )e
ix
f ( x)  d , where



1
 f (v )e
iwv
c( w)  dv .
2 

 
1
  f (v ) e
i ( x v )
f ( x)  dvd .
2
Or it can be given by
 

b) Find the complex Fourier integral (if it exists) for

a) f ( x)  e
 x
sin x,  5  x  5
c) f ( x)  
 x 0, x  5.
b) f ( x)  xe

Kassahun Nigatu(MSC), Mesfin Teshome(MSC) and Yitagesu Daba(MSC) 180

 Applied Mathematics III

3.4 Fourier Transformations

3.4.1 Fourier Cosine and Sine Transformations
An integral transform is a transformation that produces from given functions new functions that
depend on different variable and appear in the form of an integral. The Fourier and Laplace
transformations are the important integral transformations in engineering and physics. They are
important in solving ordinary differential equations, partial differential equations and integral
equations and they also help in handling special functions. First let us consider the Fourier
transformation, in particular the Fourier Cosine and Sine Transformations, and we shall see that
the Fourier transformation can be obtained from the Fourier integral representations.
3.4.1.1 Fourier Cosine Transformation
For an even function f (x) , the Fourier integral is the Fourier Cosine integral


f ( x)   A( w) cos wxdw , (1)

0

where
2
A( w) 
  f (v) cos wvdv
0

2 2 2  2 
  f (v) cos wvdv   f (v) cos wvdv .
  
 0 
0 
We now set

2
fˆc ( w)   f (v) cos wvdv ,
 0

where c represent cosine and hence

2 ˆ
A( w)  f c ( w) .

Writing v  x , we have

2
Fc ( f )  fˆC ( w)   f ( x) cos wxdx . (2)
 0

Which is called the Fourier cosine transform of f (x) , and from (1) , we obtain

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 181

 Applied Mathematics III

2 ˆ
f ( x) 
 f C (w) cos wxdw .
0

This formula gives us back f (x) from fˆC ( w) , and we therefore call f (x) the inverse

Fourier Cosine transform of fˆC ( w) .

The process of obtaining the transform Fc  fˆc from a given function f is called the Fourier
cosine transform.
3.4.1.2 Fourier Sine Transformation
For an odd function f (x) , the Fourier integral is the Fourier sine integral

f ( x)   B( w) sin wxdw (3)
0

where
2
B( w) 
  f (v) sin wvdv
0

2 2 2  2 
  f (v) sin wvdv   f ( v ) sin wvdv .
 0    0  

We now set

2
fˆs ( w)   f (v) sin wvdv ,
 0

Where s represent sine and hence

2 ˆ
B( w)  f s ( w) .

Writing v  x , we have

2
Fs ( f )  fˆs ( w)   f ( x) sin wxdx .
 0

Which is called the Fourier cosine transform of f (x) , and from (3) , we obtain

2 ˆ
f ( x) 
 f s (w) sin wxdw .
0

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 182

 Applied Mathematics III

This is called the inverse Fourier sine transform of fˆs ( w) .The process of obtaining Fs  fˆs ( w)

from f (x) is called the Fourier sine transform. Similarly for the inverse process.
1
Note that Fc 1 and F s are the inverses of Fc and Fs , respectively where “ c ” and “ s ”

indicates cosine and sine, respectively.

Examples:
a) Find the Fourier cosine and sine transforms of the
function

 x, 0  x  a
f ( x)  
0, x  a
b) Find Fc (e  x )

Solution:

2a 2  a sin wa cos wa  1 

a) Fc ( f )  x cos wxdx    .
0  w w2 
2a 2  sin wa a cos wa 

Fs ( f )  x sin wxdx    . 
0   w2 w 

b) By using integration by parts and recursion we obtain

Fc ( f ) 
2

e
0
x
cos wxdx 
2 1
 1 w 2 x 

lim e  x w sin wx  cos wx  1 
2 1
 . 
 1  w2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 183

 Applied Mathematics III

3.4.1.3 Linearity Property and Transforms of Derivatives


If f (x) is absolutely integrable on the positive x-axis (that is, 
0
f ( x) dx exists) and piecewise

continuous on every finite interval, then the Fourier cosine and sine transforms of f (x) exist.

Theorem:
The Fourier cosine and sine transforms are linear operators,
that is,
a) Fc (af  bg)  aFc ( f )  bFc ( g )
b) Fs (af  bg)  aFs ( f )  bFs ( g )
where a and b are constants.

2
Proof: a) Fc (af  bg) 
  af ( x)  bg( x)cos wxdx
0

2 2
 a
  f ( x) cos wxdx  b
0   g ( x) cos wxdx
0

 aFc ( f )  bFc ( g ) .
Thus
Fc (af  bg)  aFc ( f )  bFc ( g ) . 
Similarly for (b).

Theorem: (Fourier cosine and sine transforms of derivatives)

Let f (x) be continuous and absolutely integrable on the x -axis, let f (x) be piecewise

continuous on each finite interval, and let lim f ( x)  0 . Then

x 

Fc  f ( x)   wFs  f ( x)  
2
f (0) , ( 4a )

Fs  f ( x)   wFc  f ( x) (4b)

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 184

 Applied Mathematics III

Proof: This follows from the definitions by integration by parts,

2
Fs  f ( x)    f ( x) cos wxdx
 0

2 

  lim f ( x) cos wx  f (0)  w f ( x) sin wxdx
  x  0 
2 

  lim f ( x) cos wx  f (0)  w f ( x) sin wxdx
  x  0 

 wFs  f ( x)  
2
f (0) . 

Similarly,

2
Fs  f ( x)    f ( x) sin wxdx
 0

2 

  lim f ( x) sin wx  w f ( x) cos wxdx
 x  
0 
  wFc  f ( x) . 

Formula (4) with f  instead of f gives

Fc  f ( x)    w 2 Fc  f ( x)  
2
f (0) ,


Fs  f ( x)    w 2 Fc  f ( x)  
2
w f (0) .

Example: Find the Fourier cosine transform of

f ( x)  e  ax , a  0 .
Solution: By differentiation,

f ( x)  (e  ax )  a 2e  ax  a 2 f ( x) .
Thus

a 2 Fc ( f )  Fc ( f )   w2 Fc  f ( x)  
2
f (0)


Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 185

 Applied Mathematics III

 w2 Fc  f ( x)   a
2
.

From this it follows that

2
(a 2  w 2 ) Fc ( f )  a .

2 a 
Or Fc ( f )    , a  0. 
  a 2  w2 
x2 s2
 1 
Q18. Show that Fourier cosine transform of e 2
is e 4 .
2
Activity 3.4.3
1. Find the Fourier cosine and sine transforms of
cos x, 0  x  b
a) f ( x)  
0, xb

b) f ( x)  e  x , x  .

c) f ( x)  xe a x , a  0

d) f ( x)  xe  x cos x.

1, 0  x  1  1  cos w 
e) f ( x)   , show that  Sinwdw 
0, x  1 w 4
0

2. Find the Fourier sine transform of

a) f ( x)  xe ax
 x , 1  x  1
b) f ( x)  
0, otherwise

sin x, | x |  
c) f ( x)  
0, | x |   .

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 186

 Applied Mathematics III

3.4.2 Fourier Transformation

The preceding sections concerned two transforms obtained from the Fourier cosine and sine
integrals. We now consider a third transform, which is called the Fourier transform, and is
obtained from the complex Fourier integral.
Note that the complex Fourier integral of f (x) is given by
 
1
f ( x)    f (v ) e
iw ( x  v )
dvdw . (1)
2  

Writing the exponential function as a product of two exponential functions, we obtain

1  
1  1
 
 iwx
f ( x)   f (v)e iw( x  v ) dvdw   
iwv

2
f ( v ) e dv e dw . (2)
  2    2 
The expression in brackets is a function of w , which is denoted by F ( f )  fˆ ( w) ,and is called

the Fourier transform of f ;writing, v  x , we have


1
F ( f )  fˆ ( w)   f ( x )e
iwx
dx .
2 

With this (2) becomes


1
f ( x)   fˆ (w)e
iwx
dw .
2 

And is called the inverse Fourier transform of F ( f )  fˆ ( w) .

Note: The following two conditions are sufficient for the existence of the Fourier transform
(3) of the function f defined on the x–axis.

1. f is piecewise continuous on every finite interval



2. f is absolutely integrable on the x - axis (that is, 


f ( x) dx exists)

a x
Example: Find the Fourier transform of f ( x)  e , a  0.
Solution:

a x e  ax , x0
f ( x)  e   ax , a 0.
e , x0

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 187

 Applied Mathematics III

Then

1  0 ( a iw) 

 0
a x
F (e ) e dx  e  ( a iw )
dx
2  


1
2
 1

 a  iw 1  xlim


e ( a  iw) x 
1
 
lim e ( a  iw) x  1  .
a  iw x   
But since

lim e( a  iw) x  lim e axe  iwx

x x 

 lim e ax cos wx  i sin wx  0, a  0.

x  

and

lim e  ( a iw) x  lim e  axe  iwx

x x

 lim e  ax cos wx  i sin wx  0, a  0.

x

Hence

a x 1  1 1 
F (e ) 
2  a  iw a  iw 


1  2a 
  . 
2  a  w
2 2

Exercises:


 e dy   , to prove that the Fourier transform of e  x

y 2 2
/2
1. Use the integral is


e s / 2 .
2

2. Find the fourier transform of x exp (-ax2), where a > 0.

Linearity and Fourier Transform of Derivatives

Theorem 1: (Linearity of the Fourier transform)

If a and b are constants, then
F af  bg   aF ( f )  bF ( g ) .
Provided the Fourier transforms of f and g exist.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 188

 Applied Mathematics III

Proof: This follows from the fact that integration is a linear operation.


F af ( x)  bg( x)    af ( x)  bg( x)e

1  iwx
dx
2 

 
1 1
a  f ( x) e dx  b  g ( x )e
 iwx  iwx
dx
2  2 

 aF ( f )  bF ( g ) . 

Theorem 2 (Fourier transform of the derivative of f )

Let f be continuous on the x–axis and lim f ( x)  0 .
x 

Furthermore, let f  be absolutely integrable on the x–axis.

Then
F ( f )  iwF ( f ) .

Proof: From the definition of Fourier transform we have


1
F ( f )   f ( x)e
 iwx
dx .
2 

By using integration by parts we obtain

F ( f ) 
1 
 lim 
2  x  
f ( x ) e  iwx
 lim
x  
f ( x ) e  iwx
 iw


 f( x ) e  iwx
dx

.
 
But since lim f ( x)  0 , the expression in the brackets is zero. Thus
x 


1
F ( f )  iw  f ( x )e dx  iwF ( f )
 iwx
. 
2 

Note that from the above theorem it follows

F ( f  )  iwF ( f )  (iw) 2 F ( f )   w2 F ( f ) .
And in general

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 189

 Applied Mathematics III

F ( f ( n ) )  (iw) n F ( f ) , n Z  .

Activity 3.4.4
1. Find the Fourier cosine and sine transforms of

cos x, 0  x  b
a) f ( x)   b) f ( x)  e  x , x  .
0, xb

c) f ( x)  xe a x , a  0 d) f ( x)  xe  x cos x.

1, 0  x  1  1  cos w 
e) f ( x)   , show that  Sinwdw 
0, x  1 w 4
0

2. Find the Fourier sine transform of

 x , 1  x  1
a) f ( x)  xe ax b) f ( x)  
0, otherwise

sin x, | x |  
c) f ( x)  
0, | x |   .
3. Find the Fourier transform of the following functions.

1 if a  x  b e x if x  0
a) f ( x)   b) f ( x)    x
0 otherwise e if x  0
e x if  a  x  b  x if 0  x  a
c) f ( x)   d) f ( x)  
0 otherwise 0 otherwise

e  kx if x  0, k  0  x if  1  x  1
e) f ( x)   f) f ( x)  
0 if x  0 0 otherwise
sin x,  a  x  a
f ( x)    x, | x |  1
g) 0, | x |  a, a constant h) f ( x)  
0, | x |  1

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 190

 Applied Mathematics III

3.5 Unit Summary

 A function f is said to be periodic if it is defined for all real number x and if there is

some positive number p (called period of f ) such that

f ( x  p)  f ( x) for all x .
 The trigonometric series

a0   a
n 1
n cos nx  bn sin nx

with coefficients given by

1
a0   f ( x)dx ,
2 

1
an 
  f ( x) cos nxdx,

n  1, 2,  ,

and

1
bn 
  f ( x) sin nxdx,

n  1, 2,  ,

is called the Fourier series of f (x) .

 The Fourier series of a period function f of period P  2L is given by


n n
f ( x)  a0   (a
n 1
n cos
L
x  bn sin
L
x)

with the Fourier coefficients given by

1 L 1 L n
2 L L
a0  f ( x)dx , an   f ( x) cos xdx n,  1, 2, 3,  ,
L L L
and
1 L n
bn   f ( x) sin xdx, n  1, 2, 3,  .
L L L
 The Fourier integral of f (x) is give by


f ( x)   A( ) cos x  B( ) sin  xd ,

0

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 191

 Applied Mathematics III

where
 
1 1
A( ) 
 

f (v) cos vdv, B( ) 
 
f (v) sin  vdv

with  taking all values.

 The Fourier cosine transform of f is given by

2
Fc ( f ) 
  f ( x) cos wxdx .
0

 The Fourier sine transform of f is given by

2
Fs ( f ) 
  f ( x) sin wxdx .
0

 The Fourier transform of f is given by


1
F(f )   f ( x )e
iwx
dx .
2 

3.6 Review Problems

1. Find the Fourier series of the following periodic functions
a) f ( x)  x,    x   b) f ( x)    x, 0  x  2

 x if  2  x  0
c) f ( x)  sin x ,    x   d) f ( x)  
 x if 0  x  2
e) f ( x)    2 x ,    x   f) f ( x)  x 2 ,    x 
2. Find the complex Fourier series of
a) f ( x)  2 x, 0  x  3, f ( x  3 )  f ( x) ,

 0, 0  x  1
b) f ( x)   , f ( x  4 )  f ( x)
 1, 1  x  4
3. Find the Fourier integral representation of the function
 1, | x |  a
f ( x)   a0
 0, | x |  a ,
4. Find the Fourier cosine integral of

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 192

 Applied Mathematics III

 x 2 if 0  x  1
f ( x)  
0 if x  1
5. Find the Fourier sine integral of

e  x if 0  x  1
f ( x)  
0 if x  1
6. Find the complex Fourier integral (if it exists) for

 x , 2  x  2
f ( x)  
 0, | x |  2
 x if 1 x  a
6. Find the Fourier cosine transform of f ( x)  
0 otherwise
 x  1 if 0  x 1
7. Find the Fourier sine transform of f ( x)  
0 otherwise
8. Find the Fourier transform of

 x 2 x if x  0 kx if a  x  b
a) f ( x)   b) f ( x)  
0 if x  0 0 otherwise

 xe  x if x  0  x 2 if 0  x  1
c) f ( x)   d) f ( x)  
0 if x  0 0 otherwise

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 193

 Applied Mathematics III

UNIT FOUR

VECTOR DIFFERENTIAL CALCULUS

Unit introduction:
In this unit we study the calculus of vector fields. (These are functions that assign vectors to
points in space.) In particular we define line integrals (which can be used to find the work done
by a force field in moving an object along a curve). Then we define surface integrals (which can
be used to find the rate of fluid flow across a surface). The connections between these new types
of integrals and the single, double, and triple integrals that we have already met are given by the
higher-dimensional versions of the Fundamental Theorem of Calculus: Green’s Theorem,
Stokes’ Theorem, and the Divergence Theorem.
Unit Objectives:
At the end of this unit students will be able to:
 Define scalar and vector fields.
 Describe Vector Point functions graphically
 Define Line integrals, surface integrals
 Differentiate and integrate vector-valued functions. For a position vector function of
time, interpret these as velocity and acceleration.
 Evaluate line and surface integrals. Identify when a line integral is independent of
path and use the Fundamental Theorem of Line Integrals to solve applied problems
 Find the curl and divergence of a vector field, the work done on an object moving in a
vector field, and the flux of a field through a surface. Use these ideas to solve applied
problems
 Identify and use Green’s Theorem, the Divergence (Gauss’s) Theorem and Stokes’
Theorem.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 194

 Applied Mathematics III

4.1 Vector Fields

In this section we study calculus of a type of functions called a Vector Fields, which assigns
vectors to points in space.

Definition 2.15 A vector field F consists of two parts: a collection D of points in 2-

space or 3-space, called the domain, and a rule, which assigns to each point (x,y)
or (x,y,z) in D one and only one vector F(x,y) or F (x, y, z). In other words, a
vector field is a vector valued function of two or three variables.

A vector field F is graphically represented by drawing the vector F (x, y, z) as an arrow

emanating from (x, y, z).
z y

y x
x
F (x, y, z) =
F (x, y, z) = x i  y j

Example 1: The gravitational force F (x, y, z) exerted by a point mass m at the origin on a unit
mass located at point (x, y, z)  (0, 0, 0) is given by:
Gm
F ( x, y, z )  u ( x, y, z )
x  y2  z2
2

where G is a gravitational constant, u ( x, y, z ) is the unit vector emanating from (x, y, z) and
directed towards the origin. Hence the vector field is called the gravitational field of the point
mass.

Note that: u ( x, y, z ) has the same direction as  x i  y j  z k .

 xi  y j  z k
Then u ( x, y, z ) = .
x  y z
2 2 2

Hence, F (x, y, z) = 
Gm
x i  y j  z k .
x 
3
2
 y 2
z2 2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 195

 Applied Mathematics III

If a point (x, y, z) in space is represented by the vector r , then the gravitational field can be
written as:
Gm
F (x, y, z) =  r , where r = x i  y j  z k .
3
r

Example 2: Imagine a fluid flowing steadily along a pipe and let V(x,y,z) be the velocity vector
at a point (x,y,z). Then V assigns a vector to each point (x, y, z) in a certain domain E (the
interior of the pipe) and so V is a vector field on R 3 called a velocity field. The speed at any
given point is indicated by the length of the arrow.
A vector field F can be expressed in terms of its components, say M, N and P as follows:

F (x, y, z) = M ( x, y, z ) i  N ( x, y, z ) j  P ( x, y, z ) k
In short we can write

F (x, y, z) = M i  N j  P k .
Note that: M, N and P are scalar fields.

Let F (x, y, z) = M i  N j  P k be a vector field, we say F is continuous at (x, y, z) if and only

if M, N and P are continuous at (x, y, z).
The Gradient Fields
Definition: A scalar field is a scalar valued function of two or more variables.
Suppose f is a scalar differentiable function of three variables. Then the gradient of f, is a vector
field, denoted by grad f or  f and is given by:
f f f
grad f (x, y, z) =  f ( x, y, z ) = ( x, y , z ) i + ( x, y , z ) j + ( x, y , z ) k .
x y z
This formula defines a vector field in 3-space called the gradient field of f. Similarly, the
gradient of a function of two variables defines a gradient field in 2-space. At each point in a
gradient field where the gradient is nonzero, the vector points in the direction in which the rate of
increase of f is maximum.
Definition: If a vector field F is equal to the gradient of some differentiable function f of several
variables, then F is called a conservative vector field, and f is a potential function for F.
Example 1 Show that the gravitational field F of a point mass is a conservative vector field.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 196

 Applied Mathematics III

Gm
Solution F (x, y, z) =  r , where r = x i  y j  z k .
3
r

Then F (x, y, z) = M i  N j  P k for some scalar fields M, N and P.

We need to show that there is a differentiable function f of several variables such that
F =  f.
Gm Gm Gm
Now M =  x, N =  y and P =  z
3 3 3
r r r

x Gm
Then  M dx =  G m  dx + k (y, z) = + k (y, z)
 
3
x2  y2  z 2
x2  y 2  z 2 2

y Gm
 N dy =  G m  dy + ℓ(x, z) = + ℓ(x, z)
 
3
x  y2  z2
2
x2  y 2  z 2 2

z Gm
and  P dz =  G m  dz + q (x, y) = + q (x, y)
 
3
x2  y2  z 2
x2  y 2  z 2 2

Gm
Now let f (x, y, z) = , then F = grad f.
x  y2  z2
2

Therefore, F is a conservative vector field.

4.2 The Divergence and Curl of a Vector Field
There are two types of derivatives of a vector field, one that is a real valued function and the
other one is a vector valued function
The Divergence of a Vector Field

M N P
Definition 2.16 Let F = M i  N j  P k be a vector field such that , and
x y z
exists. Then the divergence of F, denoted div F or   F is the function defined by
div F (x, y, z) =   F ( x, y, z )
M N P
= ( x, y, z )  ( x, y, z )  ( x, y, z )
x y z

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 197

 Applied Mathematics III

Divergence is used to measure the density of a vector field in order to see how quickly matter is
moving into and out of a region.
Example 1: Find the divergence of the vector field F, where

F ( x, y, z)  ( y  z ) i  ( x  z) j  ( x  y) k
 ( y  z )  ( x  z )  ( x  y)
Solution: div F (x, y, z) =   = 0.
x y z
Therefore, div F = 0.
Note that: If div F = 0, then F is said to be divergence free or solenoidal (incompressible).

Example 2: Find the div F, if F ( x, y, z )  xe y i  ye x j  sin yz k .

 ( x e y )  ( y e x )  (sin yz )
Solution div F (x, y, z) =   .
x y z

= e y  e x  y cos yz

Therefore, div F = e x  e y  y cos yz .

The Curl of a Vector Field
The curl is used on a vector and the result of calculating the curl is also a vector. Calculating the
curl is very similar to calculating the cross product of two vectors and is denoted   F .

Definition 2.17 Let F = M i  N j  P k be a vector field such that the first partial
derivatives of M, N and P all exist. Then the curl of F, which is denoted curl F or
  F is the function defined by
curl F (x, y, z) =   F ( x, y, z)

 P N   M P   N M 
=    i     j    k
 y z   z x   x y 

The Curl of F is symbolically expressed as:

i j k
  
Curl F =
x y z
M N P

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 198

 Applied Mathematics III

Curl is used to determine the amount of rotation of a vector field. If the curl of a vector field is
zero, then it is said to be ‘irrotational’, i.e. it does not rotate.

Example 1 Find curl F if F ( x, y, z )  ( y  z ) i  ( x  z ) j  ( x  y) k

Solution M = y + z, N = x + z and P = x + y.

P N M P N M
Then = = 1, = = 1 and = = 1.
y z z x x y

Therefore, curl F = 0.

Note that: If curl F = 0, then F is said to be irrotational.

Example 2 Find curl F if F ( x, y, z )  cos x i  sin y j  e xy k .

Solution M = cos x, N = siny and P = e xy .

P P N M N M
Then = x e xy , = y e xy and = = = = 0.
y x z z x y

Therefore, curl F = x e xy i  y e xy j .

Quick check Exercises : Find the divergence and curl of the following vector fields:

a) F  x 2 y iˆ  z 3 yjˆ  xkˆ

b) F  y 2 z iˆ  x3 z 2 ˆj  3 ykˆ

c) F  3 y cos 2 x iˆ  3x sin 2 zjˆ  cos ykˆ

d) F  ze xy iˆ  z 2e x ˆj  xe3 y kˆ

The Laplacian of a vector field

Let f be a scalar field, then

2 f 2 f 2 f
  ( f ) = div (grad f) =  
x 2 y 2 z 2

The right side of this formula is the Laplacian of f usually denoted by  2 f . A function that
satisfies the equation
2 f = 0

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 199

 Applied Mathematics III

which is known as the Laplace’s equation is said to be harmonic.

Let f, M and N be functions of two variables, and let F (x, y) = M ( x, y) i  N ( x, y) j , then

f f  N M 
grad f (x, y) = i  j , curl F (x, y) =    k
x y  x y 
M N 2 f 2 f
div F (x, y) =  and  2 f (x, y) = 
x y x 2 y 2

Suppose F = M i  N j  P k is a vector field such that M, N and P have continuous partial

derivatives and if there is a function f such that F = grad f, then curl F = curl (grad f) = 0, But
curl F = 0 is equivalent to:
P N M P N M
 ,  and  (*)
y z z x x y

Note that: (*) holds for a vector field F = M i  N j  P k need not imply that F is conservative.

Theorem 2. 6 Let F = M i  N j  P k be a vector field. If there is a function f having a

continuous mixed partial derivatives whose gradient is F, then
P N M P N M
 ,  and 
y z z x x y

If the domain of F is  3 and if (*) holds, then there is a function f such that F = grad f.

In case a vector field F is given by

F (x, y) = M ( x, y) i  N ( x, y) j
the conditions in (*) reduce to
N M

x y
and the corresponding statements in the theorem holds for such vector fields.

Example 1 Let F ( x, y, z )  yz i  xz j  xy k

and G ( x, y, z )  ( x 2  y 2 ) i  ( y 2  z 2 ) j  ( x 2  z 2 ) k .
Show that F is the gradient of some function but G is not the gradient of any function.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 200

 Applied Mathematics III

Solution: For F we have

P N M P N M
x ,  y and  z .
y z z x x y

Since the domain of F is  3 , F is the gradient of some function f.

For G we have
P N
 0 and  2 z , so that the first equation in (*) is not satisfied.
y z
Therefore, G is not the gradient of any function.
x y
Example 2 Let F ( x, y)  y 2 e xy i  (1  xy ) e xy j and G ( x, y, z )  i  j .
y x
Show that F is the gradient of some function but G is not the gradient of any function.
Solution For F we have
M N
 (2  xy ) y e xy 
y x

Since the domain of F is  2 , F is the gradient of some function f.

For G we have
M x N y
  2 and  2.
y y x x
Therefore, G is not the gradient of any function.
Note that: (*) is a necessary condition for a vector field to be a conservative field.
If (*) does not hold, then there is no scalar field f such that F = grad f.
Vector Identities
The notation  that we saw in the div F and curl F is said to be the del operator and
  
 i  j  k
x y z
Note that: The derivative operations appearing in the del operator act only on functions
appearing to the right of the del operator.
Let F and G be vector fields having continuous partial derivatives and let f and g be scalar fields.
Then
i) div (curl F ) = 0 and curl (grad f) = 0
ii) div (f F) = f div F + ( grad f )  F and curl (f F) = f (curl F) + ( grad f )  F
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 201
 Applied Mathematics III

iii) div ( F  G ) = (curl F )  G – F  (curl G) and div ( grad f  grad g ) = 0

4.3 Line and surface Integrals
The concept of the integral of a function y=f(x) on an interval [a, b] can be extended to the
integral of a function z=f(x, y) along a curve in the xy plane or the integral of a function w=f(x,

y, z) on a curve in space. These integrals are called line integrals and are denoted by C
f ( x, y)ds

or C
f ( x, y, z )ds . They were invented in the early 19th century to solve problems involving fluid

flow, forces, electricity, and magnetism.

Definition(Line Integrals ):
Suppose a function f(x, y z) is defined at each point of the smooth curve c parametrized by x =
x(t), y = y(t), z =z (t), a ≤ x ≤ b. The line integral of f along C in space with respect to arc
length is defined by
n
C f ( x, y, z )ds  lim
t  0
 f ( x(ti* ), y (ti* ), z (t i* ))si
i 1

provided the limit exist.

The line integral of f along C in xy-planewith respect to arc length is defined by
n
C f ( x, y, z )ds  lim
t  0
 f ( x(ti* ), y (ti* ), z (t i* ))si
i 1

When the path of the integration is closed curve, this fact is indicated by  f x, y, z ds in place of
C
f ( x, y, z )ds .

Therefore, It can be shown that if f is continuous on C, then

C f ( x, y, z)ds a f ( x(t ), y(t ), z(t )) x(t )   y(t ) z (t ) dt

b 2 2 2
iii.

 f ( x, y)ds   f ( x(t ), y(t )) x(t )   y (t ) dt

2 2
Similarly,
C

Two Other Line Integrals:

Line integral with respect to x: C
f ( x, y, z )dx

Line integral with respect to y: C

f ( x, y, z )dy

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 202

 Applied Mathematics III

Theorem: The value of a line integral over a smooth curve C is independent of the parametric
representation of the curve C.
Example 1. Evaluate the following line integral  f ( x, y)ds,  f ( x, y)dx,  f ( x, y)dy
C C C

 x  4t  1
where f x, y   x 2  y 2 ;  C : , 1  t  1
 y  3t  1
Method 1. By parametrizing the variables.

 x  4t  1
f ( x, y )  x 2  y 2 ; C :  , 1  t  1
 y  3t  1

 f ( x, y)ds   4t  1  3t  1   

1 1
4  3 dt  5  25t 2  2t  2 dt
2 2 2 2

C 1 1

 25t 3 1
   25   25    50  310
5  t  2t   5   1  2      1  2    5  4  
2
 3   3   3   3  3
 1 

 f ( x, y)dx   25t   50   62  248

1
2
 2t  2 4dt  4  4   4  
C 1  3   3 3

   50   62 
1


C
f ( x, y )dy   25t 2  2t  2 3dt  3  4   3   62
1  3   3 

Method 2  e lim inating the parameter

 x  4t  1  3 x  12t  3 3x  7
 , 1  t  1   ,  1  t  1  3 x  4 y  7  y 
 y  3t  1  4 y  12t  4 4
 2  3x  7  2  2
3
 
3 3
5
 f ( x, y )ds    x  
5   4  
  1    dx 
4

64 5
25 x 2  42 x  49 dx
C 
3
5  25 x 3  5  3125  5 3968 5  62 310
  21x 2  49 x   225  189  147  3  525  245  64  3  3  3
64  3  5 64  
 2  3x  7  2 
3
1  25 x 3 
 
3 3
1 248
 f ( x, y )dx    x    dx       21x 2  49 x  
2
25 x 42 x 49 dx 
C 5 
  4   16 5 16  3  5 3
  3x  7   3
3
3 1  25 x 3 
3 2
3 248
 f ( x, y )dy    x 2  
5   4
 
  4
dx 
4
 
16  3
 21x 2  49 x   
 4 3
 62
C    5

 x 
 y 2 dx  2 xydy along the path consisting of the line
2
Example 2: Evaluate the line integral

segment from (0, 0) to (0,2) followed by the segment from (0, 2) to (1, 2)

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 203

 Applied Mathematics III

Solution:

y (1, 2)
C1
C2

C C 1C 2
C1 : x  0, dx  0   x 
 y 2 dx  2 xydy  0
2

C2
1
 x3 
 x   
1
11
C 2 : y  2, dy  0  2
 y dx  2 xydy  
2
x  4 dx  0    4 x   
2

C2 0  3 0 3

 x   11  11
Therefore, 2
 y 2 dx  2 xydy  0      
C  3 3

 x  2t
Example 3. Sketch the path and evaluate the line integral  (2 x  3 y)dx, C : , 0t 2
C  y  3t

Y (4, 6)

 x  2t 2

C ( 2 x  3 y ) dx , C :
 y  3t
, 0  t  2  C ( 2 x  3 y ) dx  t 0(4t  9t )(2dt )
2 2

   10tdt   5t 2
 20 answer :  20
t 0 0

Line Integrals of Vector Fields

Definition: Let F be a continuous vector field defined on a smooth curve given by a vector
function r t , a  t  b , . Then the line integral of F along C is
b

 F  dr   F r t   r t dt   F  Tds
'

C a C

where dr  dxi  dyj  dzk

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 204

 Applied Mathematics III

Therefore,
i. If F x, y, z   Pi  Qj  Rk then

 C
f ( x, y, z )ds =  Pdx  Qdy  Rdz  Pdx   Qdy   Rdz
C C C C

Similarly If F x, y   Pi  Qj then

 C
f ( x, y)ds =  Pdx  Qdy  Pdx   Qdy
C C C

Theorem: The Fundamental Theorem of Calculus for Line Integrals,


Given that F is a conservative field, is
   
 F  dr   f  dr  f (Q)  f (P)
where P is the starting point and Q is the ending point.
Definition: ( Surface Integrals)
Suppose that a surface S has a vector equation
r u, v   xu, v i  yu, v  j  zu, v k , u, v  D
The surface integral of f along S is defined by

 f x, y, z dS   f r u, v r

S D
u  rv dA

provided the limit exist.

Surface integrals of Vector Fields:
Definition: If F x, y, z   F1i  F2 j  F3 k is a continuous vector field defined on an oriented
surface with unit normal vector n, then the surface integral of over S is

 F  dS   F  ndS   F dydz  F dzdx  F dxdy 

S S S
1 2 3

Example 1: Evaluate  F  dS  
where F x, y, z   xi  z 2  zx j  xyk ,and S is the surface of the
S

triangle with Vertices 2,0,0, 0,2,0 and 0,0,4 .

Solution: Here S is the plane surface bounded 2 x  2 y  z  4 bounded by the lines
x  y  2, z  0,2 y  z  4 and 2 x  z  4, y  0 .
Now,
 
F  dS  [ xi  z 2  zx j  xyk ]  dxdyi  dzdxj  dxdzk 

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 205

 Applied Mathematics III


 xdydz  z 2  zx dzdx  xydxdy 
Hence

 F  dS   F  ndS   xdydz  z 
 zx dzdx  xydxdy 
2

S S S

1
4 2  4 z  2 4 2 x 2 2 y

 xdydz    z  22
 2
 zx dzdx    xydxdy 
0 0 0 0 0 0
3

Exercise: Show that evaluating also  F  dS   F  ndS also gives the same result.
S S

4.4 Green’s and Stoke’s Theorem:

Green’s Theorem, expresses a double integral over a region as a line integral along the boundary
of the region. This has many important consequences.
Theorem (Green’s theorem): Let C be a piecewise smooth simple closed curve that bounds the
region R in the plane. Suppose that the functions P(x,y) and Q(x,y) are continuous and have
continuous first-order partial derivatives on R. Then
 P Q 
 F  T ds   Pdx  Qdy   x  y  dA
C
C
R

Proof: Reading Assignment

Example 1. Use Green’s theorem to evaluate the integral  Pdx  Qdy
C
where P(x, y)=x-y and

Q(x, y) =y, C is the boundary of the region between the x-axis and the graph of y=sinx for
0 x 
Solution:

Sin x
1

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 206

 Applied Mathematics III

 sin x

 Pdx  Qdy   ( x  y)dx  ydy   (0  1)dA    dydx

C C R 0 0

  sin xdx   cos x 0  cos   cos 0  2


Example 2. Use Green’s Theorem to evaluate the integral  Pdx  Qdy if

C

P( x, y )  sin x  y and Q( x, y)  ln(1  y 2 ), C is the circle x 2  y 2  9.

 Pdx  Qdy  sin   

x  y dx  ln(1  y 2 ) dy   [0  1]dydx    dydx  9
C C R x 2  y 2 9

Corollary to Green’s Theorem: The area A of the region R bounded by the piecewise smooth
simple closed curve C is given by

1
2 C
A=  ydx  xdy    ydx   xdy
C C

Example 3. Use the corollary to Green’s theorem to find the area of the region between the x-
axis and one arch of the cycloid with parametric equations x=5(t-sint) and y=5(1-cost)

Cycloid
C2

C1

2 0
A    ydx    ydx    ydx    odx   5(1  cos t )(5)(1  cos t )
C C1 C2 0 2
2 2 2
1  cos 2t 
  
 25  1  cos t  dt  25  1  2 cos t  cos 2 t dt  25  1  2 cos t 
2
dt
0 
0 0
2
2
3 1  3
 25 t  2 sin t  sin 2t   25   2  75
2 4 0 2

Example 4. Use the corollary to Green’s theorem to find the area of the region between the
y  x 2 and y  x 3

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 207

 Applied Mathematics III

1 0 1 1
Area   xdy   x(3x )dx   x(2 x)dx   3x dx   2 x dx
2 3 2

C C1C 2 0 1 0 0
3 1
3x 4 2 x 3 2 1
    
4 3 0
4 3 12
Example 5. Find the area bounded by one loop of the curve x=sin2t, y=sint
Note:
x  sin 2t  x  2 sin t cos t  x 2  4 sin 2 t cos 2 t  x 2  4(sin 2 t )(1  sin 2 t )
 x 2  4 y 2 (1  y 2 )
For the y  int ercepts , x  0  4 y 2 (1  y 2 )  0  y  0 y  1 or y  1

Y
1

-1

  
2 cos 3 t
Area   xdy   2 sin t cos t cos tdt   2 cos t sin tdt  
2

C 0 0
3 0


2
3

cos 3   cos 3 0 
4
3

 
Example 6. Find the work done by the force F  5x 2 y 3 i  (7 x 3 y 2 ) j in moving a particle
once around the triangle with vertices (0, 0), (3, 0) and ((0, 6)

6 y

3 x

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 208

 Applied Mathematics III


Work   Pdx  Qdy   5 x 2 y 3dx  7 x 3 y 2 dy   21x 2 y 2  15 x 2 y 2 dA 
C C R
3  2 x 6 3 3
dx   2 x 2  2 x  6 dx 
2 x 6 972
  6 x y dydx   2 x y
2 2 2 3 3

0 0 0
0
0
6
Vector form of Green’s Theorem or divergence theorem in the plane.
Theorem:(Divergence Theorem):Let R be a bounded region with complete boundary C
oriented in positive sense, and suppose F is a vector field defined on R and its boundary. Then

 F  nds   divFdA
R

Proof:
dx dy dx dy
 dt i  dt j  dt i  dt j  dx dy
1) T   T  T  i j
v ds ds ds
dt
 dy dx
2) Let N  i j
ds ds
  dx dy dy dx   
3) T  N      0, therefore T  N and N is normal to C
ds ds ds ds
   dy dx 
4) F  N  Pi  Qj    i 
dy dx
j  P   Q
 ds ds  ds ds
 dy dx 
5)  F  nds    P   Q  ds   Pdy  Qdx
C
C
ds ds  C
Green 's Theorem
 P Q   P Q 
6)  F  nds    Qdx  Pdy    x   y dA    x  y dA
C C R R

  divF dA
R

Example 1. Verify the divergence theorem for R the interior of x 2  y 2  1 and F=2xi+3yj
P Q
1) divF    2  3  5   F  nds   divFdA   5dA  5
x y C R Circle

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 209

 Applied Mathematics III

 x  cos t
2) Let C : x 2  y 2  1 or  , 0  t  2 , P  2 x, Q  3 y
 y  sin t
2

 F  nds   Pdy  Qdx   2 cos t cos tdt  3sin t  ( sin tdt )

C C 0
2 2 2
 1  cos 2t 
  (2 cos t 3 sin t )dt   2  sin t dt    2  2 dt
2 2 2

0 0 0
2
5 1  5
  t  sin 2t    2  5
2 4 0 2

Example 2. Use the divergence theorem to evaluate the line integral  (4 x  y)dx  (3x  2 y)dy ;
C

C the complete boundary of the annulus 1  x 2  y 2  4 in the positive sense.

P
  4
 P 4 x y x P Q
  Q  divF   2
Q  3x  2 y  2 x y
y
2 2 2 2
P Q
 (4 x  y)dx  (3x  2 y)dy     2rdrd   r d   3d  6
2
 dA  4

C R
x y 0 1 0
1


Definition: Flux is the rate of flow of the fluid across C in the direction of vector n

Example 9. Use Green’s theorem to calculate the outward flux =  F  nds Where F=2xi+3yj
C

x2 y2
and C is the ellipse  1
9 4

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 210

 Applied Mathematics III

x2
2 1
  P Q
3 9
flux    F  nds   divFdA  
C R R

x y
dxdy  4
0
 2  3dxdy
0
3  /2
x2 x
 20 2 1  dx  40  1  sin 2  3 cos  d , sin  
0
9 0
3
 /2  /2  /2
1  cos 2 1 1 
 120  cos d  120
2
 d  120   sin 2   30
0 0
2 2 4 0
Stokes' Theorem:
Stokes' Theorem states that if  is an oriented surface with boundary curve C, and F is a vector

field differentiable throughout , then  F  T ds   curl F  n dS , where n (the unit normal to

C 

) and T (the unit tangent vector to C) are chosen so that n  T points inwards from C along .
4.5 Gauss Divergence Theorem
In this section we are going to relate surface integrals to triple integrals. We will do this with the
Divergence Theorem.
Theorem: Divergence Theorem
Let E be a simple solid region and S is the boundary surface of E with positive orientation. Let

be a vector field whose components have continuous first order partial derivatives. Then,


 F  ndS   div FdV

S D

Let’s see an example of how to use this theorem.

Example 1 Use the divergence theorem to evaluate  F  ndS where F x, y, z   xyi 
1 2
y j  zk
S
2

and the surface consists of the three surfaces,

z  4  3x 2  3 y 2 ,1  z  4 on the top, x 2  y 2  1,0  z  1 on the sides and z  0 on the bottom.
Solution: Let’s start this off with a sketch of the surface.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 211

 Applied Mathematics III

The region E for the triple integral is then the region enclosed by these surfaces. Note that
cylindrical coordinates would be a perfect coordinate system for this region. If we do that here
are the limits for the ranges
0  z  4  3r 2 , 0  r  1 and 0    
We’ll also need the divergence of the vector field so let’s get that.
divF  y  y  1  1
The integral is then,

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 212

 Applied Mathematics III

Group Activity
x2 y2
1.Compute the surface area of the ellipsoid whose equation is   z 2  1 . We may
4 9
parametrize this ellipse, as we have done in the past, using modified spherical coordinates
2. (a) Compute the surface area of the portion of the paraboloid z = 9  x2  y2 that lies above
the xy-plane.
3
(b) Evaluate  z

2
dS , where  is the surface whose area you found in part (a).

3.Evaluate the surface integral  F  ndS

S
for the vector field F and the oriented surface S

a) F x, y, z   xyi  yzj  zxk . S is the part of the paraboloid z  4  x 2  y 2 that lies above
the square with vertices [-1,1] , [-1,1] with positive orientation.

b) F x, y, z    yi  xj  3zk , where S is the hemisphere z  16  x 2  y 2 with upward

orientation.

Let S be the conical surface z  x 2  y 2 when z  2 .

c) A fluid has density of 15 and velocity v  xi  yj  k . Find the rate of flow downward
through S.
d) Use the divergence theorem (Gauss’ Thm) to find  F  ndS
S
given F(x,y,z)=x3i + 2xz2j +

3y2zk where S is the surface of the solid bounded by the paraboloid z  4  x 2  y 2 and
the xy plane.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 213

 Applied Mathematics III

4.6 Unit Summary

 A vector field is a vector valued function of two or three variables.
 A vector field F can be expressed in terms of its components, say M, N and P as follows:

F (x, y, z) = M ( x, y, z ) i  N ( x, y, z ) j  P ( x, y, z ) k

 The divergence of F, denoted div F or   F is the function defined by

div F (x, y, z) =   F ( x, y, z )
M N P
= ( x, y, z )  ( x, y, z )  ( x, y, z )
x y z
 Then the curl of F, which is denoted curl F or   F is the function defined by

curl F (x, y, z) =   F ( x, y, z)

 P N   M P   N M 
=    i     j     k
 y z   z x   x y 
 Then the line integral of F along C is
b

 F  dr   F r t   r t dt   F  Tds
'

C a C

where dr  dxi  dyj  dzk

 The surface integral of over S is

 F  dS   F  ndS   F dydz  F dzdx  F dxdy 

S S S
1 2 3

 Theorem: Gauss Divergence Theorem

Let E be a simple solid region and S is the boundary surface of E with positive orientation. Let

be a vector field whose components have continuous first order partial derivatives. Then,


 F  ndS   div FdV

S D

Review Problems
1.Compute divergence, curl, grad
i. Compute gradients of
a) exp(x)cos(y) b)
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 214
 Applied Mathematics III

ii. Compute divergence and curl of .

iii. Compute the curl of ,

2. Derive the expression for curl

Derive the expression for the curl of a vector function A  As sˆ  A ˆ  Az zˆ in cylindrical

coordinates. Not that you can write down sˆ  ˆ , ˆ  ẑ and zˆ  sˆ without any computations. You
may do this two ways, either (a) repeat the derivation the way we did it for   A or (b) read
through the notes about the general form of the derivative operators in curvilinear coordinates

and compute   A h
by identifying the functions s , h and hz for cylindrical coordinates.

3. Div, grad, curl on scalar
i. Given the scalar function T(x,y,z) (e.g. the temperature at any point in the room)
   
Which of the three operations (div, grad, or curl) can be sensibly operated on T?
For each which can:
a) give a formula for the result
b) explain in words how you would interpret the result.
c) is the result a vector or scalar?
ii. Given an arbitrary vector function V(x,y,z) (e.g. the velocity of a flowing liquid). Which of
the three operations (div, grad, or curl) can be sensibly operated on V? .For each which can:
a) give a formula for the result
b) explain in words how you would interpret the result.
c) is the result a vector or scalar?
4.Divergence and curl
r
Consider an electric field E = c (Please note the numerator is not rˆ : this is NOT the usual E
r2
r
field from a point charge at the origin, which would give c' , right?!)
r3


a) - Calculate the divergence and the curl of this E field.- Explicitly test your answer for the
divergence by using the divergence theorem. (Is there a delta function at the origin like there was

for a point charge field, or not?)
b) What are the units of c? What charge distribution would you need to produce an E field like
this? Describe it in words as well as formulas. (Is it physically realizable?)
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 215
 Applied Mathematics III

UNIT 5

5. COMPLEX ANALYSIS

Unit Introduction
Complex analysis extends the familiar 'real calculus' to complex calculus by introducing
complex numbers and functions. Problems which are difficult to solve in real calculus may be
solved by the complex analysis in a short period of time..The importance of complex analysis in
Applied mathematics has three main reasons.
1. Two dimensional potential problems can be modeled and solved by method of analytic
functions which we shall see in this unit. This reason is the real and imaginary parts of
analytic functions satisfy Laplace's equations in two variables.
2. Many difficult integrals (real or complex) that appear in applications can be solved by
complex integration.
3. Most functions in engineering mathematics are analytic functions ,and their study as
functions of complex variable leads to a deeper understanding of their properties and to
interrelations in complex that have no analog in real calculus.
Thus, this unit is devoted to the complex analysis and its applications.
5.1 Revision on complex numbers
The transition from real calculus to complex calculus starts with a discussion of complex
numbers and their geometric representation in the complex plane. But Students are familiarized
with most of the algebraic and geometric structure of complex numbers in their pre-university
education. To make the lesson ease and more understandable we shall revise the properties of
complex numbers along with some new concepts which are very important in the subsequent
sections of this chapter.
Definition 5.1. A complex number z is an expression which is written in the form z=x+yi where
x and y are real numbers called the real part, Re z, and imaginary part, Im z, of z respectively and
i=√−1. We can write z also as an ordered pair (x ,y) of real number x and y.
Example 1:For z=4-5i, Re z=4 and Im z=-5. For z=7i, Re z=0 and Im z=7
Definition 5.2
Two complex numbers are equal if and only if their real parts are equal and their imaginary parts
are equal.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 216

 Applied Mathematics III

Example 2 Let z1=4x-18i and z2=-36+3yi. if z1=z2 , then 4x=-36 and 3y=-18.Thus x=-9 and
y=-6.
5.1.1 Operations on complex numbers
Definition 5.3
If z1 and z2 are two complex numbers such that z1=x1+y1i and z2=x2+y2i, then the sum,
difference and product and division of the two complex numbers are defined as
i. z1+ z2=( x1+x2)+( y1+y2)i
ii. z1- z2=( x1-x2)+( y1-y2)i
iii. z1 z2=(x1x2- y1y2)+( x1y2+ y1x2)i.
𝑧1 x +y i (x +y i)(x−yi) (x1 x2 +y1 y2 𝑥2 𝑦1 −𝑦2 𝑥1
iv. = x1 +y1 i = (x1 +y1 i)(x−yi) = +
𝑧2 2 2 2 2 𝑥22 +𝑦22 𝑥22 +𝑦22

Complex or z plane
We can represent the complex number x+ yi=(x ,y) by a position vector in the xy plane whose
tail is at the origin and its head is at the point (x, y).When the xy plane is used for displaying
complex number ,it is called the complex plane ,or simply ,the z plane(Fig 5.1). Real numbers or
points corresponding to z=x =x+0i=(x, y) are represented by points on the x-axis; hence the x-
axis is called the Real axis. Purely imaginary numbers or points corresponding to z=iy=0+0i =
(0, y) are represented by points on the y-axis, and hence we call the y-axis the Imaginary axis.

Fig 5.1 The complex plane

Definition 5.4
The complex conjugate of a complex number z=x+ yi is defined as the complex number x-iy and
is denoted by 𝑧̅ : that is,
𝑧̅ = 𝑥 − 𝑖𝑦.
Example 1
a) If z=4-5i,then 𝑧̅ = 4 − (−5𝑖) = 4 + 5𝑖

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 217

 Applied Mathematics III

b) If z=6=6+0i,then 𝑧̅ = 6 − 0𝑖 = 6
Theorem 5.1
Suppose that z, z1 and z2 are arbitrary complex numbers. Then
i. 𝑧̅̅ = 𝑧 iii. 𝑧1 𝑧2 ̅1 𝑧̅2
̅̅̅̅̅̅=𝑧 v. Re(z)=
𝑧+𝑧̅
2
̅̅̅̅̅̅̅̅̅ (̅̅̅̅
ii. 𝑧1 + 𝑧2 = 𝑧̅1 + 𝑧̅2 𝑧1 ̅̅̅
z1
iv. ) = ̅̅̅ if 𝑧2 ≠ 0. vi. Im(z)=
𝑧−𝑧̅
𝑧2 z2
2𝑖

Proof. Left as an exercise.

Properties ii and iii can be extended to any finite number of terms. That is
̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
𝑧1 + 𝑧2 + 𝑧3 + − − +𝑧𝑛 = 𝑧̅1 + 𝑧̅2 + 𝑧̅3 + − − − + ̅̅̅
𝑧𝑛 and
𝑧1 𝑧2 𝑧3 − − − 𝑧𝑛 = 𝑧̅1 𝑧̅2 𝑧̅3 − − − ̅𝑧̅̅𝑛̅.
̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
Definition 5.4
The absolute value (or modulus) of a complex umber z=x+iy ,denoted by |𝑧|, is defined as
|𝑧| = √𝑥 2 + 𝑦 2 . The number |𝑧| is the distance between the origin and the point z=(x, y).
Example 2
a) If z=3-4i, then |𝑧| = √32 + (−4)2 = 5.
b) If z=9i, then |𝑧| = √02 + 92 = 9
The difference 𝑧1 − 𝑧2 represents the displacement from 𝑧2 to 𝑧1 , so the distance between 𝑧1 and
𝑧2 is given by |𝑧1 − 𝑧2 | and is drfined as
|𝑧1 − 𝑧2 | = √(𝑥1 − 𝑥2 )2 + (𝑦2 − 𝑦2 )2 if 𝑧1 = 𝑥1 + 𝑦1 𝑖 = 𝑥1 + 𝑦1 𝑖 and 𝑧2 = 𝑥2 + 𝑦2 𝑖
Example 3.Find the distance between the points 𝑧1 = 4 + 3𝑖 and 𝑧2 = 5 − 𝑖 is
|𝑧1 − 𝑧2 | = √(4 − 5)2 + (3 − (−1))2 = √17
Here are further properties of conjugates and modulus.
For any complex numbers z , 𝑧1 and 𝑧2
1. z𝑧̅ =|𝑧|2 = 𝑥 2 + 𝑦 2 if z = 𝑥 + 𝑖𝑦
2. |𝑧̅| = |𝑧|
3. |𝑧1 𝑧2 | = |𝑧1 ||𝑧2 |
|𝑧 | |𝑧 |
4. ||𝑧1 || = |𝑧1 | provided that 𝑧2 ≠ 0.
2 2

5. |𝑧1 + 𝑧2 | ≤ |𝑧1 | + |𝑧2 | (Triangular inequality)

6. |𝑧1 − 𝑧2 | ≥ |𝑧1 | − |𝑧2 |

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 218

 Applied Mathematics III

Note : The Triangular inequality can be extended to a finite sum as follows:

|𝑧1 + 𝑧2 + 𝑧3 + − − − + 𝑧𝑛 | ≤ |𝑧1 | + |𝑧2 | + − − − |𝑧𝑛 | for any integer n.
5.1.2 Polar coordinates of a complex number
Let r and 𝜃 be polar coordinates of the point (x,y) that corresponds to a nonzero complex number
z=x+ iy. Since x=r cos 𝜃 and y= r sin 𝜃,then the number z can be reprened in polar form as
𝑦
z=r(cos 𝜃 + i s 𝑖𝑛 𝜃) where r=|𝑧| = √𝑥 2 + 𝑦 2 and 𝜃 = tan−1 𝑥 (Fig 5.2).The real number 𝜃 is

called an argument of z and is denoted by arg z. The unique value of arg z lying in the range
−𝜋 < 𝜃 ≤ 𝜃 is called the principal argument of z and denoted by Arg z.If Arg z= 𝜃, then
arg z=Arg z + 2n𝝅, n∈ 𝒁 When z is a negative real number , Arg z has a value 𝜋 not − 𝜋.

Fig 5.2
Example1 . Express each of the following numbers in polar form.
a) 1 + 𝑖 b) −1 − 𝑖
𝜋
Solutions. a) r = √1 + 1 = √2. Since z is in first quadrant ,Arg z=tan−1 1 = .Therefore
4
𝜋 𝜋
z=1 + 𝑖=√2 (cos 4 + sin 4 )

This is only one of an infinite number of possibilities for the polar form of 1 +𝑖 (for n=0) :
𝜋 𝜋
1 + 𝑖=√2 (cos( 4 + 2𝑛𝜋) + sin( 4 + 2𝑛𝜋) (n=0, ±1, ±2, − − − )

3𝜋
b) r=√1 + 1 = √2 .Since the number is in third quadrant , Arg z=tan−1 1 = − .Then
4
3𝜋 3𝜋
-1-i = √2 (cos(− 4
) + sin(− 4
).

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 219

 Applied Mathematics III

3𝜋 3𝜋
Note that -1-i = √2 (cos(− + 2𝑛𝜋) + sin(− + 2𝑛𝜋 ) (n=0, ±1, ±2, − − − ).
4 4

The symbol 𝑒 𝑖𝜃 𝑜𝑟 exp(𝑖𝜃) is defined by means of Euler formula as 𝑒 𝑖𝜃 = cos 𝜃 + 𝑖 sin 𝜃 where
𝜃 is to be measured in radian measure.
It enables us to write the polar form 𝑧 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃 )more compactly in exponential form
as 𝑧 = 𝑟𝑒 𝑖𝜃 .
4𝜋
Example 2.For a complex number z=-4+4√3 , we have = 8 𝑎𝑛𝑑 𝐴𝑟𝑔 𝜃 = . Then the
3
4𝜋 𝑖
exponential form of the complex number is 𝑧 = 8𝑒 3 .
Remark : If 𝑧1 = 𝑟1 (cos 𝜃1 + 𝑖 sin 𝜃1 ) and 𝑧2 = 𝑟2 (cos 𝜃2 + 𝑖 sin 𝜃2 ), then
𝑟1 = 𝑟2 𝑎𝑛𝑑 𝜃1 = 𝜃2 + 2𝑛𝜋,n∈ 𝑧.
The polar representation of a complex number gives a simple method to multiply and divide
complex numbers.
THEOREM 5.2
If 𝑧1 = 𝑟1 (cos 𝜃1 + 𝑖 sin 𝜃1 ) and 𝑧2 = 𝑟2 (cos 𝜃2 + 𝑖 sin 𝜃2 ), then
i. 𝑧1 𝑧2= 𝑟1 𝑟2 (cos(𝜃1 + 𝜃2 ) + 𝑖 sin(𝜃1 + 𝜃2 )
𝑧1 𝑟
ii. = 𝑟1 (cos(𝜃1 − 𝜃2 ) + 𝑖 sin(𝜃1 − 𝜃2 )
𝑧2 2

If z=r(cos 𝜃 + i s 𝑖𝑛 𝜃),then by the above Theorem zz=z2=r2(cos 2𝜃 + 𝑖 sin 2𝜃 and

1 1
= 𝑧 −1 = 𝑟 (cos( −𝜃) + 𝑖 sin(−𝜃).
𝑧

From this we can generalize ,by mathematical induction ,that 𝑧 𝑛 = 𝑟 𝑛 (cos 𝑛𝜃 + 𝑖 sin 𝑛𝜃) for
any integer n .For |𝑧| = 𝑟 = 1 this formula becomes
(𝐜𝐨𝐬 𝜽 + 𝒊 𝐬𝐢𝐧 𝜽)𝒏 = 𝐜𝐨𝐬 𝒏𝜽 + 𝒊 𝐬𝐢𝐧 𝜽 which is called De Moivre's formula,in honor of
French Mathematician Abrham De moivre (1667-1754).
Example . Evaluate (−√3 − 𝑖)30 .
1 5𝜋
Solution . Since r=2 and Arg z=tan−1 =− , then
√3 6

−5𝜋 −5𝜋
(−√3 − 𝑖)30 = 230 (cos(30 ∗ ) + 𝑖 sin (30 ∗ )
6 6
= 230 (cos −25𝜋𝑖 + 𝑖 sin −25𝜋𝑖 = −230
Roots of complex numbers

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 220

 Applied Mathematics III

If 𝑧 = 𝑤 𝑛 (𝑛 = 0,1, ,3 − − −),then to each value of w there corresponds one value of z.

Conversely, to a given 𝑧 ≠ 0 there correspond precisely n distinct values of w. Each of these
values is called an nth root of z, and we write
𝑛
w= √𝑧.
𝑛
The n distinct values of √𝑧 are
𝑛 𝜃+2𝑘𝜋 𝜃+2𝑘𝜋
√𝑟 (cos( 𝑛
)+ i sin(
𝑛
))
𝑛
where k = 0,1,2,---,n-1. These n values lie on a circle of radius √𝑟 with center at the origin and
constitute the vertices of a regular polygon of n sides.
Proof . Left as an exercise.
4
Example 3. Find all values of √1 .
Solution. Since z=1, we have 𝑟 = 1 𝑎𝑛𝑑 𝐴𝑟𝑧 = 0. Then
4 2𝑘𝜋 2𝑘𝜋
√1 = cos( 𝑛
)+ i sin (
𝑛
) (𝑘 = 0,1,2 𝑎𝑛𝑑 3.

= ±1, ±𝑖
These values lie the circle of radius 1 and center at the origin.
1
Example 4. Find all values of (−8i)3 .
1 𝜋 𝜋
𝜋 3 − +2𝑘𝜋 − +2𝑘𝜋
Solution . r = 8 and Arg z= − 2  (−8𝑖)3 = √8((cos( 2
)+ i sin( 2
)) (k =0,1,2)
3 3
1
Therefore, (−8𝑖)3 are √3 − 𝑖 , 2𝑖 𝑎𝑛𝑑 − √3 − 𝑖 for k=0,1 and 2 respectively.
Remark if w1 is any nth root of an arbitrary complex number 𝑧 ≠ 0 and w is the value
𝑛
corresponding to k=1, then the n values of √𝑧 are
w1 , w1w , w1w2---, w1wn-1
Quick Exercises .
1. Find all roots of the following complex numbers
3 5 4
a) √−𝑖 b) √−3 c) √1 − 𝑖
Definition 5.8
Let z be a complex number and k be a fixed number. An equation of the form |𝑧 − 𝑘| = 𝜌 is a
circle of radius a and center 𝜌.It is the set of all z whose distance |𝑧 − 𝑘| from the center 𝑘
equals 𝜌 whose figure is given below(Fig 5.3).

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 221

 Applied Mathematics III

Fig 5.3
Example 1.|𝒛 − 𝟒| = 𝟐 is an equation of a circle with radius 2 and center at 4.
Example 2. |𝑧 + 2𝑖| = 1 is an equation of a circle with center at -2i and radius 1.
Example 3. |𝑧 + 1 − 3𝑖| = 5 is an equation of a circle with center at -1-2i = (-1,2) and radius 5.
An open circular disk which is given by |𝑧 − 𝑘| < 𝜌 is the set of all points z lying inside but not
on a circle with center at k.
A closed circular disk which is given by |𝑧 − 𝑘| ≤ 𝜌 is the set of all its interior points plus the
circle itself. An open circular disk |𝑧 − 𝑘| < 𝜌 is also called a neighbourhood of k or a 𝜌-
neighbourhood of k.
Examples 4.The solution set of the inequalities |𝑧| < 3, |𝑧 − 𝑖| < 5 and |𝑧 + 2 + 3𝑖| < 8 are
neighbourhods of the points 0,i and -2-3i with radii 3,5 and 8, respectively
Any set containing a 𝜌-neighbourhood of k is also called neighborhood of k.
The set of all z whose distance |𝑧 − 𝑘| from 𝑘 is greater than 𝜌1 ,but less than 𝜌1 is called an
open annulus(or circular ring)(Fig 5.4)and is given by the inequality 𝜌1 < |𝑧 − 𝑘| < 𝜌2
.Similarly the closed annulus 𝜌1 ≤ |𝑧 − 𝑘| ≤ 𝜌2 (Fig 5.4)includes the two circles.

Fig 5.4 Open annulus in the complex plane

The set of all points z=x+ i y such that y>0 is called an (open) upper half plane. Similarly, y<0
the lower half plane, x>0 the right half-plane and x<0 the left half plane.
A set S is called open if every point of S has a neighborhood consisting entirely of points that
belongs to S. For example, the points in the interior of a circle or a square and the right half-
plane Re z=𝑥 > 0 form an open set.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 222

 Applied Mathematics III

An open set S is connected if each pair of points z1 and z2 in it can be joined by a polygonal line,
consisting of a finite number of line segments joined end to end, that lies entirely in S. The open
set |z| < 1 is connected. The annulus 1 < |z| < 2 is, of course, open and it is also connected. A
nonempty open set that is connected is called a domain. thus any neighborhood is a domain. An
open square with a diagonal removed is not a domain since it is not connected.
The complement of a set S in the complex plane is the set of all points of the complex plane that
do not belong to S. A set S is called closed if its complement is open. For example, the points on
and inside the unit circle form a closed set (“closed unit disk”) since its complement is open.
A boundary point of a set S is a point every neighborhood of which contains both points that
belong to S and points that do not belong to S. For example, the boundary points of an annulus
are the points on the two bounding circles. Clearly, if a set S is open, then no boundary point
belongs to S; if S is closed, then every boundary point belongs to S. The set of all boundary
points of a set S is called the boundary of S.
A region is a set consisting of a domain plus, perhaps, some or all of its boundary points.
Quick Exercises.
Sketch the following sets and determine and determine which are domains.
1. |𝑧 − 2 + 𝑖| ≤ 1 4. Im z = 1
𝜋
2. |3𝑧 + 4| > 5 5. 0 ≤ arg 𝑧 ≤ 4
3. Im z > 2
5.2 COMPLEX ANALYTIC FUNCTIONS
From the real calculus, a real function f defined on a set S of real numbers (usually an interval)
is a rule that assigns to every x in S a real number f(x), called the value of f at x.
Now in complex, S is a set of complex numbers and a function f defined on S is a rule that
assigns to every z in S a complex number w, called the value of f at z and we use the notation

w=f(z).
Here z varies in S and is called a complex variable. The set S is called the domain of definition
of f or, simply, the domain of f. The set of all values of a function f is called the range of f.
Example1: w=2z4+5z-4 is a complex function defined for all z; that is, its domain S is the whole
complex plane.
z
Example2:f(z)=z2 +4 is a complex valued function defined for all z except at z = ± i.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 223

 Applied Mathematics III

In general if u(x ,y) and v(x ,y) are real valued function of x & y defined on a region s of the
complex plane then
w=f(z)=u(x ,y)+iv(x ,y)
is a complex valued function defined on S.
Conversely each complex valued function w=f(z) can be written as
w=f(z)=u(x,y)+iv(x,y)
where u and v are the real valued functions and called the real part and the imaginary part of
the function f(z) respectively.
E x a m p l e 1 Find the real and imaginary part of the following function and evaluate the value
of f at the indicated point.
a) W=z2+1 at z=2-3i
Solution. Let z=𝑥 + 𝑖𝑦.Then w=( 𝑥 + 𝑖𝑦)2 +1=x2-y2+1+2xyi so that u=Re f(z)= x2-y2+1 , v=Im
f(z)=2xy and f(2-3i)=(2-3i)2+1=-4-12i.From this we can see that u(2,-3)=4 and v(2,-3)=-12
b) w=f(z)=4z-iz at z=2-i
Solution. f(z)=4(𝑥 + 𝑖𝑦)-i(𝑥 + 𝑖𝑦)=4x+y+(4y-x)i yields u(x,y)=4x+y and v(x,y)=4y-x
Also f(2-i)=4(2-i)-i(2-i)=8-4i-2i-1=7-6i
Remark. If the polar coordinates r and θ, instead of x and y, are used, then u + iv = f (𝑟𝑒 𝑖𝜃 )
where w = u + iv and z = reiθ In that case, we may write
f (z) = u(r, θ) + iv(r, θ).
Example 2. If f (z) = z2, then
f (𝑥 + 𝑖𝑦) = (𝑥 + 𝑖𝑦)2 = x2 − y2 + i2xy.
Hence
u(x, y) = x2− y2 and v(x, y) = 2xy.
When polar coordinates are used,
f(𝑟𝑒 𝑖𝜃 ) = 𝑟 2 𝑒 2𝑖𝜃 = 𝑟 2 (cos2 𝜃 + 𝑖 sin2 𝜃) so that u(x ,y)= 𝑟 2 cos 2𝜃 and v(x ,y)= 𝑟 2 sin 2𝜃

Group Activity
1. For each of the functions below, describe the domain.
z−4 z
a) f(z)= z2 −5z+6 ; c) f(z)=z+z̅
1 1
b) f(z)= Arg ( z) d) f(z)=. 1−|z|2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 224

 Applied Mathematics III

2. Find the real and imaginary part of the following function and evaluate the value of f at the
indicated points.
a) f(z)=4z2-5z+3 at z=4-5i c) f(z)=𝑧̅ 𝑅𝑒(𝑧) − 4𝑧 + 2𝐼𝑚(𝑧) at 𝑧 = −𝑖
2z 1
b) 𝑓(𝑧) = 3−z at z= 3 -2i d) f(z)=𝑧̅ 2 − (−4 − 2𝑖)𝑧 at z=1+i

3. Write the following function in the form f(z)= u(r, θ)+iv(r, θ).
1
a) f(z)= z+z (z≠ 0) b) f(z)=𝒛𝟑 + 𝒛−𝟑

4. Let f(z)=f(𝑥 + 𝑖𝑦)=𝑒 𝑥 cos 𝑦 + 𝑖𝑒 𝑥 sin 𝑦.Find

a) f(0) b) f(𝜋3 𝑖) c) 𝜋
f(3 + 2 𝑖)

5. Let f(z)=𝑧 21 − 5𝑧 7 +9𝑧 4 .Use polar coordinates to find

a) f(1-i) b) f(√3 + 𝑖)
5.2.1 Limit and Continuity
Definition 5.9: A function w=f(z) is said to have a limit l as z tends to z0 if given any ε > 0
∃ δ > 0 ∋ |f(z) − l| < ε whenever |z − z0 | < δ for all z ≠ z0 (Fig 5.5).In this case we write
limz→z0 f(z) = l.

Fig 5.5.
Geometrically it means that given any open disk with center l and radius ε there exist an open
disc with center z0 and radius δ for every point z0 in the disc |z − z0 | < δ
Formally, this definition is similar to that in calculus, but there is a big difference. In the real
case, x can approach an x0 only along the real line, whereas in the complex calculus z may
approach z0 from any direction in the complex plane.
Remark: If limz→z0 f(z) exists then it is unique.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 225

 Applied Mathematics III

Now let us state the following theorems on the sum, product and quotient of two complex
functions whose proof is analogous to the proof of real functions.
Theorem 5.3. Let f and g be two functions such that limz→z0 f(z)=l and limz→z0 g(z) = m.then
I. limz→z0 f(z) ± g(z) = l ± m f(z) l
III. limz→z0 g(z) = m provided m≠ 0
II. limz→z0 f(z) g(z) = lm
It is easy to see from definition (5.9) of limit that
limz→z0 c=c and limz→z0 z = z0
where z0 and c are any complex numbers; and, by mathematical induction, it follows that
limz→z0 z n = z0n (n = 1, 2, . . .)
So, the limit of a polynomial
P(z) = a0 + a1z + a2z2 + ·· ·+anzn
as z approaches a point z0 is the value of the polynomial at that point: that is
lim p(z) = p(z0 )
z→z0
p(z) p(z )
If p(z) and q(z) are polynomials q(z0)≠ 0 then limz→z0 q(z) = q(z0). If both p(z)=0 and q(z)=0,
0

then p and q can be factored as p(z)=(z-z0)p1(z) and q(z)= (z-z0)q1(z).If q1(z0) ≠ 0, then
𝑝(𝑧) 𝑝1 (𝑧0 )
lim =
𝑧→𝑧0 𝑞(𝑧) 𝑞1 (𝑧0 )
Example 1.Evaluate the following limits
a) limz→i 2z 3 − 4z + 1 b) limz→1+2i
z2 +6z−3
c) limz→i
z2 +1
3z−2 z2 −3iz−2

Solutions. a)Since 2z 3 − 4z + 1 is polynomial,

lim 2z 3 − 4z + 1 = 2i3 − 4i + 1 = −6i + 1
z→i

b) Since z 2 + 6z − 3 and 3z − 2 are both polynomials,

z2 +6z−3 (1+2i)2 +6(1+2i)−3 16i+3
limz→1+2i = = 3(1+2i)
3z−2 3(1+2i)

z2 +1 (z−i)(z+i) (z+i)
c) limz→i = limz→i = limz→i = −2
z2 −3iz−2 (z−i)(z−2i) (z−2i)

Quick Exercise.
Evaluate the following limits
1.limz→2−3i 2z 2 − 5z + 3 2.limz→−2i
z3 −3z2 +1
3.limz→1+i
z2 −2i
z4 +z−2 z2 −2z+2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 226

 Applied Mathematics III

CONTINUITY
Definition 5.10 A function f is continuous at a point z0 if all three of the following conditions
are satisfied:
1.limz→z0 exists
(2) f (z0) is defined
(3) limz→z0= f (z0)
A function of a complex variable is said to be continuous in a region R if it is continuous at each
point in R.
Theorem 5.4 . Suppose that f and g are continuous at z0. Then the the sum f+g, the difference f-
𝑓
g, the product fg, the quotient provided that g(z0)≠ 0 and the composition fog, where
𝑔

(fog)(z)=f(g(z) provided f is continuous in a neighborhood of g(z0) are also continuous at z0.

Example 1. The polynomial function defined by

P(z) = a0 + a1z + a2z2 + ·· ·+an zn
is continuous at each point z0 in the complex plane.
𝑧−3𝑖
Example 2, The function f(z)=𝑧 2 +9 is not continous at z=±3𝑖 since f(±3𝑖) is not defined.

5.2.2 Derivative
Definition 5.11 The derivative of a complex function f at a point 𝑧0 is written f ′ (z0 ) and is
defined by
𝑓(𝑧0 +∆𝑧)−𝑓(𝑧0 )
𝑓 ′ (𝑧0 ) = lim∆𝑧→0 (1)
∆z
provided this limit exists. Then f is said to be differentiable at z0 . If we write ∆z = z − z0 ,
we have z = z0 +∆z and (1) takes the form
𝑓(𝑧)−𝑓(𝑧0 )
𝑓 ′ (𝑧0 ) = lim𝑧→𝑧0 (2)
𝑧−𝑧0
𝑑𝑤 ∆𝑤
If we let w=f(z) and ∆𝑤 = 𝑓(𝑧) − 𝑓(𝑧0 ), then we can use the Leibniz notation = lim∆𝑧→0
𝑑𝑧 ∆𝑧

Differentiability at z0 means that, along whatever path z approaches , the quotient in (2) always
approaches a certain value and all these values are equal.
Example3

The function f(z)= z 2 is differentiable for all z and has the derivative 𝑓 ′ (z) = 2z because

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 227

 Applied Mathematics III

(z + ∆z)2 − ∆z z 2 + 2z∆z − (∆z)2 − ∆z

lim = lim = lim 2z + ∆z = 2z
∆z→0 ∆z ∆z→0 ∆z ∆z→0

Example 4. Show that f(z)= 𝑧̅ is nowhere differentiable.

Solution. Let ∆𝑧 = ∆𝑥 + 𝑖∆𝑦 and we get
f(z+∆z)−f(z) ̅̅̅̅̅̅̅−𝑧̅
z+∆z ̅∆𝑧
̅̅̅ ∆𝑥−𝑖∆𝑦
= = ∆𝑧 = ∆𝑥+𝑖∆𝑦 ( ****)
∆z ∆z

If ∆𝑦 = 0,this is 1.If ∆𝑥 = 0,this is -1.Thus (****) approaches 1 along path I in Fig and -1
along path II. Hence , by definition, the limit of **** as ∆z approaches to zero does not exist at
any z. Therefore, f(z)= 𝑧̅ is not differentiable at z, and since z is arbitrary, f(z) is nowhere
differentiable.

Fig 5.6
Quick Exercise
1. Show that 𝑓 ′ (𝑧) = 4𝑧 3 if f(z)=𝑧 4 .
2. Show that f(z)=|𝑧|2 is differentiable only at z=0.
3. Show that 𝑓 ′ (𝑧) does not exist at any when
a) f(z)= Re z b) f(z)=Im z
It is also easy to show that, by definition of limit of complex function,
𝐝 𝐝 𝐝
𝐜=𝟎, 𝐳 = 𝟏 𝐚𝐧𝐝 (𝐜𝐟(𝐳) = 𝐜𝐟 ′ (𝐳)
𝐝𝐳 𝐝𝐳 𝐝𝐳
Also, if n is a positive integer,
d n
z = nz n−1
dz
This formula remains valid when n is a negative integer, provided that z ≠ 0.
Theorem: Differentiation Formula
If the derivatives of two functions f and g exist at a point z, then
d
i. (f(z) + g(z)) = f ′ (z) + g ′ (z)
dz
𝑑
ii. (𝑓(𝑧)𝑔(𝑧)) = 𝑓 ′ (𝑧) 𝑔(𝑧) + 𝑓 (𝑧) 𝑔′ (𝑧)
𝑑𝑧

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 228

 Applied Mathematics III

𝐝 𝐟(𝐳) 𝐟 ′ (𝐳)𝐠(𝐳)−𝐟 (𝐳) 𝐠 ′ (𝐳)

iii. ( )= provided g(z )≠ 0
𝐝𝐳 𝐠(𝐳) 𝐠(𝐳)𝟐

Proof . Left as an exercise

There is also a chain rule for differentiating composite functions. Suppose that f has a derivative
at z0 and that g has a derivative at the point f (z0 ). Then the function F(z) = g[f (z)] has a
derivative at z0 ,and
F ′ (z0 ) = g ′ [f(z0 )]f ′ (z0 )
If we write w = f (z) and W = g(w), so that W = F(z), the chain rule becomes
dW dW dw
=
dz dw dz
EXAMPLES: Find the value of the derivative of
1. f(z)=2𝑧 5 − 4𝑧 3 + 3𝑧 at z=i 2z3 +1
3. f(z)=2z4 −3z at z= -2i
2. f(z)=(3z 2 -2i)(2z+3i) at 1+i
4. f(z) = (3z 4 + i)8
Solutions.
1. 𝑓 ′ (𝑧) =10𝑧 4 − 12𝑧 2 + 3 and 𝑓 ′ (𝑖) =10𝑖 4 − 12𝑖 2 + 3 =10+12+3=25
2. Using product rule , 𝑓 ′ (𝑧) = (6𝑧)(2𝑧 + 3𝑖) + (3z 2 − 2i)2 = 18z 2 + 18zi − 4i and thus
𝑓 ′ (1 + 3𝑖) =18(1 + 𝑖)2 + 18(1 + 𝑖)𝑖 − 4𝑖 = 58𝑖 − 18
6𝑧 2 (2z4 −3z)−(2z3 +1)(8z−3)
3. Using quotient rule, 𝑓 ′ (𝑧) = and thus
(2z4 −3z)2
−24(32+6𝑖)−16𝑖+1)(−16𝑖−3
𝑓 ′ (−2𝑖) = (32+6i)2

4. To find the derivative of (3z 4 + i)8, write w = 3z 4 + i and W=w 8 Then

d
(2z 2 + i)8=8w 7 12z=96z 3 (3z 4 + i)7
dz

Theorem. If f(z) is differentiable at z0, then it is continuous at z0.

Proof. Left as an exercise.
The converse of the above theorem is not necessarily true. For example, f(z)=𝑧̅ is continous
everywhere but differntiable nowhere.
Analytic Functions
Complex analysis is concerned with the theory and application of “analytic functions,” that is,
functions that are differentiable in neighborhoods of points, which are pillars to the complex
analysis structure whose definition is given as follows.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 229

 Applied Mathematics III

Definition 5.12 :A function f is said to be analytic in a domain D if f (z) is defined and

differentiable at all points of D. The function f (z) is said to be analytic at a point in D if f (z) is
analytic in a neighborhood of z0 .
Hence analyticity of f (z) at z0 means that f (z) has a derivative at every point in some
neighborhood of z0 (including z0 itself since, by definition, z0 is a point of all its
neighborhoods). We have a special term if f is analytic on the whole complex plane.
Definition 5.13 If f is analytic on the whole complex plane , then f is said to be entire.
A more modern term for analytic in D is holomorphic or regular in D.
Example 1 : The nonnegative integer powers 1, z, z 2 , z 3 , ….are analytic in the entire complex
plane, and so are polynomials, that is, functions of the form
𝑓(𝑧) = 𝑐0 + 𝑐1 𝑧 + 𝑐2 𝑧 2 + ⋯ + 𝑐𝑛 𝑧 𝑛
where c0. c1. c3 … cn are complex constants. Therefore, every polynomial is an entire function,
The quotient of two polynomials g(z)and h(z),
g(z)
f(z)=h(z)

is called a rational function. This f is analytic except at the points where h(z)= 0:
Here we assume that common factors of g and h have been canceled.
Example 2.f (z) = 1/z is analytic at each nonzero point in the finite plane.
Example 3.The function f (z) = |z|2 is not analytic at any point since its derivative exists only at
z = 0 and not throughout any neighborhood which we have seen in the previous section.
Definition 5.14: A point 𝑧0 is said to be singular or singularity of f if the function f fails to be
analytic at 𝑧0 .
𝑧−5 𝑧−5
Example 4.f(z)=𝑧 2 −3𝑧+2 ==(𝑧−2)(𝑧−1) is not analytic at z=1 and z=2 and hence z=1 and z=2 are

the singularities of f.
Example 5.f(z)=|𝑧|2 has no singular point since f is nowhere analytic.
Theorem (L'HOPITAL rule)
Assume that f and g are analytic at 𝑧0 . If f(𝑧0 ) = 0 , g(𝑧0 ) = 0 and 𝑔′ (𝑧0 ) ≠ 0, then
𝑓(𝑧 ) 𝑓 ′ (𝑧 )
lim𝑧→𝑧0 𝑔(𝑧0 )= lim𝑧→𝑧0 𝑔′ (𝑧0 )
0 0

Example 6 . Evaluate the following limits.

𝑧 6 +1 𝑧 2 −𝑖𝑧−1−𝑖
1. lim𝑧→𝑖 𝑧 2 +1. 2. lim𝑧→1+𝑖 𝑧 2 −2𝑧+2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 230

 Applied Mathematics III

Solution
Let f(z)= 𝑧 6 + 1 and g(z)= 𝑧 2 + 1 .Thus 𝑓 ′ (𝑧) = 6𝑧 5 and 𝑔′ (𝑧) = 2𝑧.Since f(i)= g(i)=0 and
𝑔′ (𝑖) = 2𝑖 ≠ 0, then by the L'HOPITAL rule,
𝑧 6 +1 6𝑧 5 −6𝑖
lim𝑧→𝑖 𝑧 2 +1= lim𝑧→𝑖 = = −3
2𝑧 2𝑖

Quick exercises
1. Find all the singularities of the following functions
𝑧 𝑧+6 𝑧+6
a) f(z)=𝑧 3 −8 b) f(z)=𝑧 2 (𝑧 2 +9) c) f(z)=𝑧 3 −𝑧 2 +4𝑧−4

2. Use L'HOPITAL rule to find the following limits

𝒛𝟒 +𝟏 𝑧 2 −𝑖𝑧−1−𝑖 𝑧 6 −64
a) 𝐥𝐢𝐦𝒛→𝟏+𝒊 b) lim𝑧→1+𝑖 c) lim𝑧→1+𝑖√3
𝒛𝟐 −𝟐𝒛+𝟐 𝑧 2 −2𝑧+2 𝑧 3 +8

5.2.3 Cauchy–Riemann Equations

Many times we counter complex functions written in the form of f(z)=f(x+ iy)=u(x ,y)+iv(x
,y)=(x3-3xy)+i(3x2y-y3).Is there some criterion that we can use to determine whether f is
differentiable and ,if so find the value of f(z)?
The answer to this questions is yes, thanks to the independent discovery of two important
equations by the French mathematician Augistin-Louis Cauchy and the German mathematician
Georg Friedrich Bernhard Riemann which provide a criterion (a test) for the analyticity of a
complex function w=f(z)= u(x, y)+iv(x, y)
Theorem 5.5 Cauchy–Riemann Equations
Let f be defined and continuous in some neighborhood of a point and differentiable at z itself.
Then, at that point, the first-order partial derivatives of u and v exist and satisfy the Cauchy–
Riemann equations
ux = vy , uy = −vx (1)
Hence, if f is analytic in a domain D, those partial derivatives exist and satisfy (1) at all points of
D.
Proof : Since 𝑓 ′ (𝑧) exists,
𝑓(𝑧 + ∆𝑧) − 𝑓(𝑧)
𝑓 ′ (𝑧) = lim (2)
∆𝑧→0 ∆𝑧
Let ∆𝑧 = ∆𝑥 + 𝑖∆𝑦. 𝑇ℎ𝑒𝑛 𝑧 + ∆𝑧 = 𝑧 + ∆𝑥 + 𝑖(𝑦 + ∆𝑦), and in terms of u and v the derivative
in (2) becomes

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 231

 Applied Mathematics III

[𝑢(𝑥+∆𝑥,𝑦+∆𝑦)+𝑖𝑣((𝑥+∆𝑥,𝑦+∆𝑦)]−[𝑢(𝑥,𝑦)−𝑖𝑣(𝑥,𝑦)]
𝑓 ′ (𝑧) = lim∆𝑧→0 (3)
∆𝑥+𝑖∆𝑦

We first choose path I in Fig 5.6. Thus we let first ∆𝑦 → 0 and then ∆𝑥 → 0. After ∆𝑦 = 0is
zero, ∆𝑧 = ∆𝑥 . Then (3) becomes, if we first write the two u-terms and then the two v-terms,

𝑢(𝑥+∆𝑥,𝑦)−𝑢(𝑥,𝑦) 𝑣(𝑥+∆𝑥,𝑦)−𝑣(𝑥,𝑦)
𝑓 ′ (𝑧) = lim∆𝑥→0 + i lim∆𝑥→0
∆𝑥 ∆𝑥

Fig 5.6. Paths in (2)

Since𝑓 ′ (𝑧) exists, the two real limits on the right exist. By definition, they are the partial
derivatives of u and v with respect to x. Hence the derivative of f (z) can be written
𝑓 ′ (𝑧) = 𝑢𝑥 + 𝑖𝑣𝑥 (4).
Similarly, if we choose path II in Fig 5.6, we let ∆𝑥 → 0 first and then∆𝑦 → 0 . After ∆𝑥 is
zero, ∆𝑧 = 𝑖∆𝑦 , so that from (3) we now obtain
𝑢(𝑥,𝑦+∆𝑦)−𝑢(𝑥,𝑦) 𝑣(𝑥,𝑦+∆𝑦)−𝑣(𝑥,𝑦)
𝑓 ′ (𝑧) = lim∆𝑦→0 + i lim∆𝑦→0
𝑖∆𝑦 𝑖∆𝑦

Since 𝑓 ′ (𝑧) exists, the limits on the right exist and give the partial derivatives of u and v with
1
respect to y; noting that 𝑖 = −𝑖. we thus obtain

𝑓 ′ (𝑧) = −𝑖𝑢𝑦 + 𝑣𝑦 . (5)

The existence of the derivative 𝑓 ′ (𝑧) thus implies the existence of the four partial derivatives in
(4) and (5). By equating the real parts 𝑢𝑥 and 𝑣𝑦 in (4) and (5) we obtain the first Cauchy–
Riemann equation (1). Equating the imaginary parts gives the other. This proves the first
statement of the theorem and implies the second because of the definition of analyticity.
Example 1. In the previous discussion, we showed that the function f (z) = 𝑧 2 = x2 – y2 + i2xy
is analytic for all z and that 𝑓 ′ (z) = 2z. To verify that the Cauchy–Riemann equations are
satisfied everywhere, write
u(x, y) = x2 – y2 and v(x, y) = 2xy.
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 232
 Applied Mathematics III

Thus
ux = 2x =vy , uy = −2y =−vx ,
EXAMPLE 2. Is 𝑧̅analytic?
Solution. Since 𝑧̅=𝑥 − 𝑖𝑦,we have u=x and v=-y .Then
𝑢𝑥 = 1 , 𝑢𝑦 = 0, 𝑣𝑥 = 0 & 𝑣𝑦 = −1 𝑤ℎ𝑖𝑐ℎ 𝑠ℎ𝑜𝑤𝑠 𝑡ℎ𝑎𝑡 𝑢𝑦 = −𝑣𝑥 𝑏𝑢𝑡 𝑢𝑥 ≠ 𝑣𝑦.
∴f(z)=𝑧̅ 𝑖𝑠 𝑛𝑜𝑡 𝑎𝑛𝑎𝑙𝑦𝑡𝑖𝑐
The Cauchy–Riemann equations are fundamental because they are not only necessary but also
sufficient for a function to be analytic. More precisely, the following theorem holds.
Theorem 5.6 Cauchy–Riemann Equations
If two real-valued continuous functions u(x, y)and v(x, y) of two real variables x and y have
continuous first partial derivatives that satisfy the Cauchy–Riemann equations in some domain
D, then the complex function f(z)=u(x, y)+iv(x, y) is analytic in D and 𝑓 ′ (z) = 𝑢𝑥 + 𝑖𝑣𝑥 in D.
Using Theorems 5.5 and 5.6 we can now easily find out whether or not a given complex
function is analytic or not.
Example 3. Is f(z)=𝑒 −2𝑥 (𝑐𝑜𝑠 2𝑦 − 𝑖𝑠𝑖𝑛 2𝑦) anyltic ?
Solution. We have u=e−2x cos 2y and v=− e−2x sin 2y and by differentiation
𝑢𝑥 = −2 e−2x cos 2y 𝑣𝑥 = 2 e−2x sin 2y
𝑢𝑦 = −2 e−2x sin 2y 𝑣𝑦 = −2 e−2x cos 2y
Since the Cauchy Riemann equations are satisfied, that is 𝑢𝑥 = 𝑣𝑦 𝑎𝑛𝑑 𝑢𝑦 = −𝑣𝑥 , we
conclude that f(z) is analytic for all z and
𝑓 ′ (z) = 𝑢𝑥 + 𝑖𝑣𝑥 =−2 e−2x cos 2y + 2 e−2x sin 2y
Quick exercises :Are the following functions analytic ?If so, find 𝑓 ′ (z).
1. f(z) = cos x cosh y − i sin x sinh y.
2. f (z) = 2x + ix𝑦 2
3. 𝑓(𝑧) = 𝑥 3 − 3𝑥𝑦 2 + 𝑖(3𝑥 2 𝑦 − 𝑦 3 )
4. 𝑓(𝑧) = 𝑒 −𝑦 cos 𝑥 − +𝑖𝑒 −𝑦 sin 𝑥
Remark. If we use the polar form z=r(cos𝜃 + 𝑖𝑠𝑖𝑛𝜃)and set f(z)=𝑢(𝑟, 𝜃) + 𝑖𝑣(𝑟, 𝜃), then the
Cauchy–Riemann equations are
1 1
ur = vθ , vr = − uθ (r > 0).
r r

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 233

 Applied Mathematics III

1
and f ′ (z) = f ′ (reiθ ) = r e−iθ (vθ − uθ ).

Example 1.
1 1 1 1
𝑓(𝑧) = = 𝑖𝜃
= 𝑒 −𝑖𝜃 = (𝑐𝑜𝑠𝜃 − 𝑠𝑖𝑛𝜃) (𝑧 ≠ 0)
𝑧 𝑟𝑒 𝑟 𝑟
1 1
from which u(r,𝜃) = 𝑐𝑜𝑠𝜃 𝑎𝑛𝑑 𝑣(𝑟, 𝜃) = − 𝑟 𝑠𝑖𝑛𝜃 and
𝑟

𝑢𝑟 = cos 𝜃 , 𝑣𝑟 = − sin 𝜃 ,
1 1
𝑢𝜃 = − 𝑟 𝑠𝑖𝑛𝜃, 𝑣𝜃 = − 𝑟 𝑐𝑜𝑠𝜃 .
𝟏 𝟏
Since the Cauchy-Riemann equations 𝐮𝐫 = 𝐫 𝐯𝛉 𝐯𝐫 = − 𝐫 𝐮𝛉 are satisfied, then f is analytic at

any nonzero point z and

𝑐𝑜𝑠𝜗 𝑠𝑖𝑛𝜃 𝑒 −𝑖𝜃 1 1
𝑓 ′ (𝑧) = 𝑒 −𝑖𝜃 (− 2
+ 2
) = −𝑒 −𝑖𝜃
2
=− 𝑖𝜃 2
=− 2
𝑟 𝑟 𝑟 (𝑟𝑒 ) 𝑧
The following two theorems are the consequences of the Cauchy-Riemann equations.
Theorem 5.7. If is analytic in a domain D and |𝑓(𝑧)| = 𝑘 = 𝑐𝑜𝑛𝑠𝑡 in D, then f(z)=const in D.
Theorem 5.8. Let f be analytic in the domain D .If 𝑓 ′ (𝑧) = 0 for all z in D, then f is constant in
D.
Proof. By the Cauchy-Riemann equations 𝑓 ′ (𝑧) =𝑢𝑥 + 𝑖𝑣𝑥 = 𝑣𝑦 − 𝑖𝑢𝑦 for all z ∈ 𝐷.By the
hypothesis 𝑓 ′ (𝑧) = 0 for all z in D the functions 𝑢𝑥 , 𝑢𝑦, 𝑣𝑥 , 𝑢𝑥 and 𝑣𝑦 are identically zero .By the
theorem5.7 both u and v are constant functions and hence f(z) is constant.
̅̅̅̅̅̅ analytic in a region D show that f(z) is constant in D.
Example 1.If f (z) and 𝑓(𝑧)
̅̅̅̅̅̅= u(x ,y) - iv(x ,y).Since f(z) is analytic in D,
Solution. Let f(z)=u(x ,y) + iv(x ,y).Then 𝑓(𝑧)
̅̅̅̅̅̅ is analytic in D, 𝑢𝑥 = −𝑣𝑦 and 𝑢𝑦 = 𝑣𝑥 (ii).
𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥 (𝑖) and since 𝑓(𝑧)
Adding (i) & (ii) we get 𝑢𝑥 = 0 & 𝑢𝑦 = 0. Hence f(z)= 𝑢𝑥 + 𝑖 𝑣𝑥 = 0.
Hence f(z) is constant.
GROUP ACTIVITY
1. Use the Cauchy-Riemann equations to determine the following functions are analytic and
find 𝑓 ′ (𝑧) at those points of z where they exist.
a) f(z)=𝑧 6 𝑦+𝑖𝑥
d) f(z)=𝑥 2 +𝑦 2
1
b) f(z)=𝑧−𝑧 5 e) f(z)=Arg 2𝜋𝑧
c) f(z)=Re (𝑧 2 ) − 𝑖 𝐼𝑚 (𝑧 2 ) 𝑖
f) f(z)=𝑧 8

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 234

 Applied Mathematics III

g) f(z)=𝑥 3 + 𝑖(1 − 𝑦)3 i) f(z)=𝑒 𝑥 𝑒 −𝑖𝑦

h) f(z)=⌈𝑧 − (2 + 𝑖)⌉2 j) f(z)=
2. Show that when f(z)= 𝑥 3 + (1 − 𝑦)3, 𝑓 ′ (𝑧) = 3𝑥 2 only when z=i
3. Show that the following functions are nowhere analytic
a) f(z)= 2xy+i (𝑥 2 − 𝑦 2 )
b) f(z)=𝑒 𝑦 𝑒 𝑖𝑥
4. Find the constants a and b so that f(z)=𝑥 2 + 𝑎𝑦 2 − 2𝑥𝑦 + 𝑖(𝑏𝑥 2 − 𝑦 2 + 2𝑥𝑦) is analytic
for every z?
5. Find the constants a , b and c so that f(z)=𝑎𝑥 2 − 𝑏𝑦 2 + 𝑖 𝑐𝑥𝑦 is analytic for all z?
6. Prove that an analytic function whose real part is constant is itself constant.
5.2.4 Laplace’s Equation. Harmonic Functions
The great importance of complex analysis in engineering mathematics results mainly from the
fact that both the real part and the imaginary part of an analytic function satisfy Laplace’s
equation, the most important PDE of physics. It occurs in gravitation, electrostatics, fluid flow,
heat conduction, and other applications .
If a function of two real variables f(x ,y) satisfies Laplace equation
𝜕 2 𝑓(𝑥, 𝑦) 𝜕 2 𝑓(𝑥, 𝑦)
+ =0
𝜕𝑥 2 𝜕𝑦 2
Then we say that f(x ,y) is an harmonic function.
Harmonic functions are in applied mathematics, engineering and mathematical physics
T H E O R E M 5.9 Laplace's Equation
If f is analytic in a domain D, then both the real part and imaginary parts of f are harmonic.
Proof. Let f(z)=u(x ,y)+v(x ,y) be analytic so that 𝑢𝑥 = 𝑣𝑦 𝑎𝑛𝑑 𝑢𝑦 = −𝑣𝑥 .
Differentiating the first with respect to x and the second with respect to y gives
𝑢𝑥𝑥 = 𝑣𝑥𝑦 and 𝑢𝑦𝑦 = −𝑣𝑥𝑦=− 𝑢𝑥𝑥
And so 𝑢𝑥𝑥 + 𝑢𝑦𝑦 = 0.
By a similar reasoning
𝑣𝑥𝑥 + 𝑣𝑦𝑦 = 0
∴ u and v are harmonic.
The functions u and v are called conjugate functions. The meaning of the word conjugates here

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 235

 Applied Mathematics III

is, of course, different from that, where 𝑧̅ is defined.

Example 1. Verify that u= 𝑦 3 − 3𝑥 2 𝑦 is Harmonic and find the Harmonic conjugate v of u and
the corresponding analytic function f(z)=u +iv.
Solution . To verify that u is Harmonic ,we evaluate the second partial derivatives and note that
𝑢𝑥𝑥 + 𝑢𝑦𝑦 = −6𝑦 + 6𝑦 = 0,
so u satisfies the Laplace's equation and then u is Harmonic.
From the Cauchy-Riemann equation
𝑢𝑥 = 𝑣𝑦 = −6𝑥𝑦,
We get 𝑣𝑦 = −6𝑥𝑦.
To construct v, we integrate 𝑣𝑦 with respect to y holding x fixed and we get
𝑣 = −3𝑥𝑦 2 + ℎ(𝑥) where h(x) is an arbitrary function of x. Integrating v with respect x and
using the Cauchy-Riemann equation 𝑢𝑦 = −𝑣𝑥 , we get
3𝑦 2 − 3𝑥 2 = −(−3𝑦 2 + ℎ′ (𝑥)) or ℎ′ (𝑥) = 3𝑥 2 .
Thus h(x)=𝑥 3 + 𝐶 where C is an arbitrary real number.
Therefore, v=−3𝑥𝑦 2 + 𝑥 3 + 𝐶 is a Harmonic conjugate of u and the corresponding analytic
function is
f(z)= 𝑦 3 − 3𝑥 2 𝑦 +i(−3𝑥𝑦 2 + 𝑥 3 + 𝐶).
Example 2. Verify that u=𝑥 3 𝑦 − 𝑦 3 𝑥 is harmonic in the whole complex plane and find a
harmonic conjugate function v of u and the corresponding analytic function f?
Solution. Because 𝑢𝑥 = 3𝑥 2 𝑦 − 𝑦 3 𝑠𝑜 𝑢𝑥𝑥 = 6𝑥𝑦 𝑎𝑛𝑑 𝑢𝑦 = 𝑥 3 − 3𝑦 2 𝑥 𝑠𝑜 𝑢𝑦𝑦 = −6𝑥𝑦
Therefore 𝑢𝑥𝑥 + 𝑢𝑦𝑦 = 0.This means that u(x, y)= 𝑥 3 𝑦 − 𝑦 3 𝑥 is harmonic.
Now, if f(z)=u(x ,y) + iv(x ,y) is analytic then u(x ,y) and v(x ,y) satisfy the C-R equations.
That is
𝑢𝑥 = 3𝑥 2 𝑦 − 𝑦 3 = 𝑣𝑦
and
𝑢𝑦 = 𝑥 3 − 3𝑦 2 𝑥 = −𝑣𝑥
Integrating 𝑢𝑦= 3𝑥 2 𝑦 − 𝑦 3 with respect to y gives
3 1
v(x ,y)=2 𝑥 2 𝑦 2 − 4 𝑦 4 + ℎ(𝑥)

Now, differentiating this with respect to x gives

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 236

 Applied Mathematics III

𝑣𝑥 = 3𝑥𝑦 2 + +ℎ′ (x)

Because
3 1
v(x,y)=2 𝑥 2 𝑦 2 − 4 𝑦 4 + ℎ(𝑥) and so 𝑣𝑥 = 3𝑥𝑦 2 + +ℎ′ (x) and this is equal to -𝑢𝑦.

Now -𝑢𝑦 = −𝑥 3 + 3𝑦 2 𝑥 and so

𝑥4
ℎ′ (x)= −𝑥 3 𝑔𝑖𝑣𝑖𝑛𝑔 ℎ(𝑥) = − +𝐶
4
3 1 𝑥4
Therefore v(x,y)=2 𝑥 2 𝑦 2 − 4 𝑦 4 + − +𝐶
4

Because comparing 𝑣𝑥 = 3𝑥𝑦 2 + +ℎ′ (x) and -𝑢𝑦 = −𝑥 3 + 3𝑦 2 𝑥 where -𝑣𝑥 =-𝑢𝑦 then it is seen
that ℎ′ (x) =−𝑥 3 .
𝑥4 3 1 𝑥4
Therefore h(x)= − + 𝐶 giving v(x,y)=2 𝑥 2 𝑦 2 − 4 𝑦 4 + − + 𝐶 and the corresponding
4 4

analytic function is
3 1 𝑥4
f(z)= 𝑥 3 𝑦 − 𝑦 3 𝑥 + 𝑖 (2 𝑥 2 𝑦 2 − 4 𝑦 4 + − + 𝐶).
4

Quick Exercises.
Are the following functions harmonic? If your answer is yes, find a corresponding analytic
function f(z)=u(x ,y)+iv(x ,y)
1. u=𝑥 3 − 3𝑥𝑦 2 2. u=𝑒 𝑥 sin 2𝑦
Group Exercise
1. Determine whether the following functions(1-11) are Harmonic or not? If so, find a
corresponding analytic function f(z)=u +iv.
1. u=𝑥 3 − 3𝑥 2 7. v=(2x+1)y
𝑥
2. u=𝑥 2 +𝑦 2 8. u=x cos 3x cosh 3y + y sin 3x sinh 3y
9. 𝑢 = 𝑥 3 − 3𝑥𝑦 2 + 2𝑥 + 𝑦
3. v=𝑒 𝑥 sin 2𝑦
10. u=𝑒 𝑥 (𝑥 cos 𝑦 + 𝑦 𝑠𝑖𝑛𝑦)
4. u= sinx coshy
−(𝑥2 −𝑦2 )
5. u=𝑒 −𝑥 cos 𝑦 11. 𝑒 2 cos 𝑥𝑦
6. u=𝑥 2 − 2𝑥 − 𝑦 2
2. Find the values of the constants a, b and c that will guarantee that the function f(z)=𝑎𝑥 2 +
𝑏𝑥𝑦 + 𝑐𝑦 2 is Harmonic?
3. Determine the values of a and b so that the following functions are Harmonic and find a
Harmonic conjugate

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 237

 Applied Mathematics III

a) u=𝑒 𝜋𝑥 𝑐𝑜𝑠 𝑎𝑦 b) u=a𝑥 3 + 𝑏𝑥𝑦

4. Show that if u is Harmonic and v is a Harmonic conjugate of u, then u is a Harmonic
conjugate of -v.
5.2.5 Basic elementary functions
In this section we discuss the basic elementary complex functions which are indispensable
throughout applications, and some of them have interesting properties not shared by their real
counterparts.
The Exponential Function

Definition 5.15 The exponential function of complex analysis is defined for all z by the function
f(z)=𝑒 𝑧 = 𝑒 𝑥+𝑖𝑦 = 𝑒 𝑥 (𝑐𝑜𝑠𝑦 + 𝑖𝑠𝑖𝑛𝑦) (*)
when y is taken to be in radian measure. For convenience we use the notation exp 𝑧 instead of
𝑒 𝑧.
Example1: Show that f(z)=𝑒 𝑧 is analytic for all z and 𝑓 ′ (𝑧) = 𝑒 𝑧 using Cachy-Riemann
equations.
Solution. From f(z)= 𝑒 𝑥 (𝑐𝑜𝑠𝑦 + 𝑖 𝑠𝑖𝑛𝑦) we have
u(x, y)= 𝑒 𝑥 𝑐𝑜𝑠𝑦 ,v(x, y)= 𝑒 𝑥 𝑠𝑖𝑛𝑦; and by differentiation
𝑢𝑥 = 𝑒 𝑥 𝑐𝑜𝑠𝑦 , 𝑣𝑥 = 𝑒 𝑥 𝑠𝑖𝑛𝑦
𝑢𝑦 = −𝑒 𝑥 𝑠𝑖𝑛𝑦, 𝑣𝑦 = 𝑒 𝑥 𝑐𝑜𝑠𝑦.
Since 𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥 then f(z)= 𝑒 𝑧 is analytic for all z by the Cachy-Rieman equations
and 𝑓 ′ (𝑧) = 𝑢𝑥 + 𝑖𝑣𝑥 = 𝑒 𝑥 𝑐𝑜𝑠𝑦 + 𝑒 𝑥 𝑠𝑖𝑛𝑦
= 𝑒 𝑧 ∀𝑧.
Therefore 𝑒 𝑧 is an entire function
If z is a pure imaginary number ,that is z= iy, then 𝑒 𝑖𝑦 = 𝑐𝑜𝑠𝑦 + 𝑖 𝑠𝑖𝑛𝑦 which is called a Euler
formula.
Example 1. 𝑒 2𝜋𝑖 = 𝑐𝑜𝑠2𝜋 + 𝑖𝑠𝑖𝑛2𝜋 = 1 + 0 = 0
Recall that the polar form of a complex number z is z=r(cos 𝜃 + 𝑖 sin 𝜃) and can be written more
compactly as an exponential form z=r𝑒 𝑖𝜃 .
Example 2 . Write the exponential form of z=1+i.
𝜋
𝜋
Solution. r= √2 and 𝜃 = 4
and z=√2 𝑒 𝑖 4 .

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 238

 Applied Mathematics III

We can easily verify that

1. 𝑒 𝑧1 +𝑧2 = 𝑒 𝑧1 𝑒 𝑧2 𝑒 𝑧1
2. 𝑒 𝑧1 −𝑧2 =𝑒 𝑧2

Remarks.
1. 𝑒 𝑧 is a periodic function with a period of 2𝜋𝑖 since f(z+2𝜋𝑖)= 𝑒 𝑧+2𝜋𝑖 = 𝑒 𝑧 =f(z).This
property is not shared by the real fuctio f(x)=x.
2. In the real case, f(x)=ex is never negative whereas in the complex case ez can be
negative.
Example 3. Solve ez = -1

Solution . ez= ex( cos y + i sin y)= -1+ 0i and 𝑠𝑖𝑛𝑐𝑒 |𝑒 𝑧 | = 𝑒 𝑥 = 1 𝑡ℎ𝑒𝑛 𝑥 = 0. As ex=1,
ex cos y= -1 and ex sin y=0 ,then x=0 and y=(2𝑛 + 1)𝜋, 𝑛 = 0, ±1, ±2, − − −.
Therefore z=(2𝑛 + 1)𝜋, 𝑛 = 0, ±1, ±2, − − −.
Example .𝑒 𝑧 = 1 + 𝑖
𝜋𝑖
𝜋
Solution . 𝑒 𝑥+𝑖𝑦 =𝑒 𝑥 𝑒 𝑦𝑖 = √2𝑒 4 implies that 𝑒 𝑥 = √2 and y= 4 + 2𝑛𝜋 𝑛 = 0, ±1, ±2, − − −.

Since ln ex =x , it follows that

𝜋
x=ln √2 and y= y= + 2𝑛𝜋 𝑛 = 0, ±1, ±2, − − −.
4
𝜋
∴ 𝑧 = ln √2 + 4 + 2𝑛𝜋 𝑛 = 0, ±1, ±2, − − −

Quick Exercises.
1. Write an exponential form of
a) z=1-i b) z=3-4i
2. Solve 𝑒 𝑧 = 2 + 3𝑖
Group Exercise.
1. Show that
a) exp(4+𝜋𝑖)=-𝑒 4 b) exp(𝜋𝑖 + 𝑧) = −𝑒𝑥𝑝𝑧
2. Find the real and imaginary part of
𝜋𝑧
a) exp(− ) c) exp(𝑧 2 )
3
1 d) exp(𝑧 3 )
b) exp(𝑧)

3. Find all values of z such that

a) 𝑒 𝑧 = −2 b) 𝑒 𝑧 = 1 + √3𝑖 c) 𝑒 2𝑧−1 = 1

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 239

 Applied Mathematics III

̅̅̅̅̅̅̅̅̅̅ = 𝑒𝑥𝑝(𝑖𝑧
4. Show that 𝑒𝑥𝑝(𝑖𝑧) ̅ ) if and only if z=n𝜋 (𝑛 = 0, ±1, ±2,---.
5. Show that if ez is real ,then Im z= n𝜋 (𝑛 = 0, ±1, ±2,---.
1
6. Write Re(𝑒 𝑧 ) interms of x and y.
7. Show that f(z)=exp𝑧̅ is not analytic anywhere ,using the Cauchy-Riemann equations.
Trigonometric function.

From Euler's formula 𝑒 𝑖𝑧 = cos 𝑧 + 𝑖 sin 𝑧 and 𝑒 −𝑖𝑧 = cos 𝑧 − 𝑖 sin 𝑧 for all z. Adding and
subtracting these equations we obtain the following definition.
Definition 5.16. The sine and cosine functions of a complex variable z is defined by as
1 1
sin z=2𝑖 (𝑒 𝑖𝑧 − 𝑒 −𝑖𝑧 ) and cos z=2 (𝑒 𝑖𝑧 + 𝑒 −𝑖𝑧 )

The other trigonometric functions tan z, cot z , sec z and csc z can be defined in terms of sin z
and cos z as of the real case. and moreover
Example 1. Show that sin z=sin x coshy +icos xsinh y .
1
Solution . sin z=2𝑖 (𝑒 𝑖𝑧 − 𝑒 −𝑖𝑧 )= sin z
1
=2𝑖 (𝑒 𝑖(𝑥+𝑖𝑦) − 𝑒 −𝑖(𝑥+𝑖𝑦) )
1
= (𝑒 −𝑦+𝑖𝑥 − 𝑒 𝑦−𝑖𝑥 )
2𝑖
1
= (𝑒 −𝑦 (cos 𝑥 + 𝑖𝑠𝑖𝑛 𝑥) − 𝑒 𝑦 (cos 𝑥 − 𝑖𝑠𝑖𝑛 𝑥)
2𝑖
𝑒 𝑦 +𝑒 −𝑦 𝑒 𝑦 −𝑒 −𝑦
= sin x( ) + 𝑖 cos 𝑥( )
2 2

=sinx cosh y + i cos x sinh y

𝑒 𝑦 +𝑒 −𝑦 𝑒 𝑦 −𝑒 −𝑦
where cosh y = and sinh y= .
2 2

Note that Re(sin z)= sin x cosh y and Im(sin z)= cos x sinh y.
Similarly, we can show that cos z= cosx coh y - isin x sinh y.
Here are some properties of the trigonometric functions.
 sin (z1+z2)= sin z1cos z2+cos z1sin z2 ∀𝑧  sin2z +cos2z=1 ∀𝑧.
 cos (z1+z2)= cos z1cos z2+sin z1sin z2 ∀𝑧  sin 2z=2sin z cos z ∀𝑧
 sin -z= -sin z ∀𝑧.  cos 2z=cos2z -sin2z ∀𝑧
 cos -z=cos z ∀𝑧.
Definition 5.17 A solution to the equation f(z)=0 is called a zero of the function f.
Example 2. Solve cos z=0.
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 240
 Applied Mathematics III

Solution. cos z= cosx coh y - i sin x sinh y=0 if and only if

cosx coh y=0 and sin x sinh y=0.
𝜋
Since cosh y is never zero, cosx coh y=0 implies that cos x=0 from which x= 2 (2𝑛 + 1) for any
𝜋
integer n. Substituting the values z=x+iy= 2 (2𝑛 + 1) + 𝑖𝑦 in sin x sinh y=0 we obtain
𝜋
0= sin (2 (2𝑛 + 1) sinh y=(−1)𝑛 sinh 𝑦,
𝜋
which implies that y=0.Therfore z= 2 (2𝑛 + 1) for any integer n.

Similarly, we can show that sin z=0 iff z=n 𝜋 for any integer n
Example 3 . Solve sin z=2
Solution . sin z=sin x cosh y +icos x sinh y=2+0i which implies sin x cosh y=2 and
𝜋
cos x sinh y=0.If cosx sinh y=0,then x= 2 +n𝜋 for any integer z or y=0. If y=0,the first equation
𝜋
sin x cosh y= sin x cosh 0=sin x=2 which is impossible because sin x ≤ 1 for all x. If x= 2 +n𝜋

for any integer z, then sin x=±1 and the first equation becomes
cosh y=±2. Since cosh y≥ 1,then cosh y=2 from which y=cosh-1 2.
𝜋
Therefore, z= 2 +n𝜋 + 𝑖 cosh-1 2

Quick exercises. Show that

1. |cos 𝑧|2 = 𝑐𝑜𝑠 2 𝑥 + 𝑠𝑖𝑛ℎ2 𝑦
2. |sin 𝑧|2 = 𝑠𝑖𝑛2 𝑥 + 𝑠𝑖𝑛ℎ2 𝑦
sin z and cos z are entire functions since they are the linear combination of the entire functions
𝑒 𝑖𝑧 and 𝑒 −𝑖𝑧 and we can easily prove that
𝑑 𝑑
sin 𝑧 = cos 𝑧 𝑎𝑛𝑑 cos 𝑧 = −sin 𝑧
𝑑𝑧 𝑑𝑧
Quick exercise. Show that
𝑑 𝑑
1. tan 𝑧 = 𝑠𝑒𝑐 2 (𝑧) 3. sec 𝑧 = 𝑠𝑒𝑐𝑧 𝑡𝑎𝑛𝑧
𝑑𝑧 𝑑𝑧
𝑑 𝑑
2. cot 𝑧 = −𝑐𝑠𝑐 2 (𝑧) 4. csc 𝑧 = − csc 𝑧 cot 𝑧
𝑑𝑧 𝑑𝑧

HYPERBOLIC FUNCTIONS

Definition 5.18 .The hyperbolic sine hyperbolic cosine functions of a complex variable z are
defined as
𝑒 𝑧 −𝑒 −𝑧 𝑒 𝑧 + 𝑒 −𝑧
sinh z = and cosh z = .
2 2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 241

 Applied Mathematics III

The other hyperbolic functions tanh z, sech z, csch z and coth z can be defined in terms of sinh z
and cosh z as of the real case.
Since ez and e-z are entire functions ,then sinh z and cosh z are entire functions. Furthermore,
𝑑 𝑑
sinh 𝑧 = cosh 𝑧 𝑎𝑛𝑑 cosh 𝑧 = sinh 𝑧
𝑑𝑧 𝑑𝑧
The derivatives of the remaining hyperbolic functions follow the same rule as in real calculus:
𝑑 𝑑
tanh 𝑧 = 𝑠𝑒𝑐ℎ2 𝑧 and coth 𝑧 = −𝑐𝑠𝑐ℎ2 𝑧
𝑑𝑧 𝑑𝑧
𝑑 𝑑
sech 𝑧 = − sech 𝑧 tanh 𝑧 and 𝑑𝑧 csch 𝑧 = − csch 𝑧 coth 𝑧
𝑑𝑧

The Complex trigonometric and hyperbolic functions are related.

Examples . Show that
1. cosh iz =cos z 2. sinh iz = i sin z
𝑒 𝑖𝑧 + 𝑒 −𝑧 𝑖 𝑒 𝑖𝑧 − 𝑒 −𝑖𝑧
Solutions. By definition cosh iz= = cos 𝑧 and sinh iz = =sin z.
2 2

Here are some important identities involving hyperbolic functions.

1. cosh z= coshx cosy+isinh x sin y 5. cosh2z - sinh2z=1
2. sinh z= sinhx cosy+icosh x sin y 6. sinh(z1+z2)=sinhz1cosh z2+cosh z1sinh z2
3. sin(iz)=isinh z 7. cosh(z1+z2)=coshz1coshz2+sinh z1sinh z2
4. cos(iz)=cosh z

Exercises
1. Prove that Im cos z and Re sin z are Harmonic.
2. Write the following in terms of u+ iv
a) sin 2𝜋𝑖 d) cosh(2+3i)
𝜋
b) cos i e) sin(𝜋 +
c) sin(1-3i)
3. Show that sinz is odd and cosz is even function.
4. Solve the following equations.
a) sin z=5 c) sinh z=0 e) sinh z=i
b) cosh z=0 d) cosh z=-1 1
f) cosh z=2

Logarithm and General power

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 242

 Applied Mathematics III

Finally we try to introduce a complex logarithm which is somewhat complicated compared to the
real logarithm.
Definition 5.19. The natural logarithm of z=x+iy denoted by lnz (or sometimes by log z) and is
defined as the inverse of the exponential function, that is, w=lnz is defined for z≠ 0 by the
equation ew=z.
If we write z=rei𝜃 , r > 0, then lnz=lnr+i𝜃 ,r=|𝑧| > 0, 𝜃 = arg 𝑧.Since arg 𝑧 is integer multiples
of 2𝜋,then ln z is a multiple valued function with infinitely many values.
The principal value of ln z denoted by Ln z is defined by
Ln z=ln|𝑧|+i Argz where −𝜋 < 𝜃 ≤ 𝜋.
The function Ln z is single-valued function as Argz is unique.
If 𝜃 is the particular value of arg z such that −𝜋 < 𝜃 ≤ 𝜋, then
ln z= ln|𝑧| + 𝑖(𝜃 + 2𝑛𝜋), 𝑛 𝑖𝑠 𝑎𝑛𝑦 𝑖𝑛𝑡𝑒𝑔𝑒𝑟.If z is positive real ,then Arg z=0 and Ln z becomes
identical with the real natural logarithm of z. If z is negative real,then
Arg z= 𝜋 and Ln z= ln|𝑧|+i 𝜋
Examples
1. ln1=0,±2𝜋𝑖, ±4𝜋𝑖, − − − 𝐿𝑛1 = 0
2. ln-1=±𝜋𝑖, 3𝜋𝑖, ±𝜋𝑖 𝐿n-1= 𝜋𝑖
3. ln-4=1.386294±(2𝑛 + 1)𝜋𝑖 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑖𝑛𝑡𝑒𝑔𝑒𝑟 𝑛 Ln-4= ln-4=1.386294+𝜋𝑖
𝜋𝑖 3𝜋 5𝜋𝑖 𝜋𝑖
4. lni= 2 , , ,--- Lni= 2
2 2

Quick execises: Evaluate the following.

1. ln4i and Ln4i 2. ln-4i and Ln-4i 3. ln3-4i and Ln3-4i
Now let's proof that at any point except for the origin and points on the negative real axis the
function ln z is analytic.
Let lnz=lnr+i(𝜃 + 2𝑛𝜋) so that u(r,𝜃)=lnr and v(r,𝜃)= 𝜃 + 2𝑛𝜋 for any integer n.
1
𝑢𝑟 = , 𝑢𝜃 = 0
𝑟
𝑣𝑟 = 0 , 𝑣𝜃 = 1.
Since the first order partial derivatives of u and v are ctinous and satisfies the Riemann-
𝟏 𝟏
equations 𝐮𝐫 = 𝐫 𝐯𝛉 𝐯𝐫 = − 𝐫 𝐮𝛉 ,lnz is analytic for all points other than the origin and the
negative real axis and

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 243

 Applied Mathematics III

𝐫
𝐟 ′ (𝐳) = 𝐳(𝑢𝑟 + 𝑖𝑣𝑟 )(𝑤ℎ𝑦)
𝐫 1 1
=𝐳( + 𝑖0) =
𝑟 𝑧

𝑑 1
∴ 𝑙𝑛𝑧 =
𝑑𝑧 𝑧
General power

Definition 5.19. The General powers of a complex number z=x+iy are defined by the formula
𝑧 𝑐 = 𝑒 𝑐𝑙𝑛𝑧 =𝑒 𝑐𝑙𝑛|𝑧|+𝑖(𝜃+2𝑛𝜋) where where c is complex,z≠ 0 and any integer n.
𝑧 𝑐 is multivalued function(why).The particular function
𝑧 𝑐 = 𝑒 𝑐𝐿𝑛𝑧
is called the principal value of 𝑧 𝑐 .
𝜋 𝜋 𝜋
Example 1: 𝑖 𝑖 = 𝑒 𝑖𝑙𝑛𝑖 = 𝑒 𝑖(2 𝑖±2𝑛𝜋𝑖) 𝑒 − 2 ±2𝑛𝜋) and the principal value (n=0) is 𝑒 − 2
1 1 𝜋 1 𝜋 𝜋𝑖 𝑛𝜋
1⁄ (𝑙𝑛|𝑖|)+𝑖( +2𝑛𝜋)= ( 𝑖( +2𝑛𝜋)=
Example 2: 𝑖 2 = 𝑒 (2𝑙𝑛𝑖) = 𝑒 (2 2 𝑒 2 2 𝑒4𝑒 for k∈ 𝑍.Since 𝑒 𝑛𝜋 = ±1,
1⁄ 𝜋𝑖 𝑛𝜋 1+𝑖
𝑖 2 = ±1𝑒 4 𝑒 = ±1 ( ).
√2
1+𝑖
The principal value is when n=0.
√2

Quick exercise:
Find the principal value of
1. (1 − 𝑖)1+𝑖 2. (1 + 𝑖)2−𝑖
The inverse trigonometric and Hyperbolic functions.

Definition 5.20. The inverse sine w=arcsin 𝑧 is the relation that sinw=z.
Example 1. Show that arcsin 𝑧 =-i ln(𝑖𝑧 ± √1 − 𝑧 2 ))
𝑒 𝑖𝑤 −𝑒 −𝑖𝑤
Solution . w=arcsin 𝑧 implies that z=sin w= from which
2𝑖

𝑒 2𝑖𝑤 − 2𝑖𝑧𝑒 𝑖𝑤 − 1 = 0.It is a quadratic equation in 𝑒 𝑖𝑤 .If we let m=𝑒 𝑖𝑤 ,then

2𝑖𝑧±2√1−𝑧 2
m2-2izm-1=0.Solving this quadratic equation we get m= = 𝑖𝑧 ± √1 − 𝑧 2 .Since
2

m=𝑒 𝑖𝑤 = 𝑖𝑧 ± √1 − 𝑧 2 .,then w=−𝑖ln(√𝑖𝑧 ± √1 − 𝑧 2 ).

A similar representation of arcos z, etc shows that
𝑖 𝑖+𝑧
1. arccos 𝑧 = −𝑖ln(√𝑧 ± √1 − 𝑧 2 ) 2. arctan 𝑧 =2 ln(𝑖−𝑧)

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 244

 Applied Mathematics III

1 𝑖+𝑧
3. arccosh 𝑧 = ln(√𝑧 ± √1 − 𝑧 2 ) 5. arctanh 𝑧 =2 ln(𝑖−𝑧)

4. arcsinh 𝑧 = ln(√𝑧 + √1 + 𝑧 2 )
5.3 Complex Integration
The previous two sections establish the basic or fundamental for the study of complex analysis
covered complex numbers, limits and differentiation and introduce the most important concept of
analyticity. Now we shall consider the next part of complex analysis, complex integration
First, we shall show how to evaluate integrals of a complex functions by defining the integral of
a complex valued function of a real variable
Definition 5.21. Let f(t)=u(t) + iv(t) where u and v are real-valued function of the real variable t
for a≤ 𝑡 ≤ 𝑏.Then
𝑏 𝑏 𝑏

∫ 𝑓(𝑡)𝑑𝑡 = ∫ 𝑢(𝑡)𝑑𝑡 + 𝑖 ∫ 𝑣(𝑡)𝑑𝑡

𝑎 𝑎 𝑎

Examples . Evaluate the following integrals

2 1
1. ∫1 (5𝑡 4 − 4𝑖𝑡 + 1) 𝑑𝑡 2. ∫0 (𝑡 + 1)3 𝑑𝑡

Solution.
1. Let u= 5𝑡 4 + 1 and v= -4t so that
2 2 2
4 4
∫ (5𝑡 − 4𝑖𝑡 + 1) 𝑑𝑡 = ∫ (5𝑡 + 1)𝑑𝑡 + 𝑖 ∫ −4𝑡𝑑𝑡
1 1 1

=𝑡 5 + 𝑡]12 − 𝑖𝑡 2 [12 =(34-2)-i(4-1)=32-3i

2. Let f (t)= (𝑡 + 1)3 =𝑡 3 + 3𝑡 + 𝑖(3𝑡 2 + 1).Now u(t)= 𝑡 3 + 3𝑡 and v(t)= 3𝑡 2 + 1 and
1 1
𝑡 4 3𝑡 2 1 7
3
∫ 𝑡 + 3𝑡 𝑑𝑡 = + ]0 = 𝑎𝑛𝑑 ∫ 3𝑡 2 + 1 𝑑𝑡 = 𝑡 3 + 𝑡[10 = 2
4 2 4
0 0
1 1 1
7
𝐻𝑒𝑛𝑐𝑒 ∫(𝑡 + 1)3 𝑑𝑡 = ∫ 𝑡 3 + 3𝑡 𝑑𝑡 + 𝑖 ∫ 3𝑡 2 + 1 𝑑𝑡 = + 2𝑖
4
0 0 0

Quick exercises
Evaluate the following integrals
𝝅 𝟐 𝜋
1. ∫𝟎𝟒 𝒕𝒆𝒊𝒕 𝒅𝒕 2. ∫−𝟏(𝟒𝒕 + 𝟐𝒊)𝟐 𝒅𝒕 3.∫0 𝑒 (𝑡+𝑖𝑡) 𝑑𝑡

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 245

 Applied Mathematics III

The following are some properties of complex integrals which can easily be verified.
Let f(t)=u(t)+iv(t) and g(t)=r(t)+is(t) be two continuous functions on[a,b].Then
𝑏 𝑏 𝑏
1. ∫𝑎 (𝑓(𝑡) + 𝑔(𝑡))𝑑𝑡 = ∫𝑎 𝑓(𝑡)𝑑𝑡 + ∫𝑎 𝑔(𝑡)𝑑𝑡
𝑏 𝑐 𝑏
2. ∫𝑎 𝑓(𝑡)𝑑𝑡 = ∫𝑎 𝑓(𝑡)𝑑𝑡 + ∫𝑐 𝑓(𝑡)𝑑𝑡 where a≤ 𝑐 ≤ 𝑏
𝑏 𝑏
3. ∫𝑎 𝑐𝑓(𝑡)𝑑𝑡 = 𝑐 ∫𝑎 𝑓(𝑡)𝑑𝑡 where c is a complex consant.
𝑏 𝑎
4. ∫𝑎 𝑓(𝑡)𝑑𝑡 = − ∫𝑏 𝑓(𝑡)𝑑𝑡
Now ,we shall see the integral of complex 𝑣𝑎𝑙𝑢𝑒𝑑 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑠 which are defined on curves in the
complex plane called contour integral or line integral.
5.3.1 Contour integral: Line integral in the complex plane
If z moves along the curve C in the z-plane(complex plane) and at each position z has associated
with it a function of z ,i.e. f(z),then summing up f(z) for all such points between A & B(Fig 5.7)
means that we are evaluating a line integral in the z-plane between A(z=z1) & B(z=z2) along the

curve C, i.e. we are evaluating c

f ( z )dz where C is the particular path joining A to B.

Fig 5.7
The evaluation of line integral in the complex plane is known as contour integral or complex

integral and is denoted by 

c
f ( z )dz or C
f ( z )dz if C is a closed path(one whose terminal

point coincides with initial point).Here the integrand f(z) is integrated over a given curve C,
called the path of integration. We may represent a curve C by a parametric representation
z(t)=x(t)+iy(t) ,a≤ 𝑡 ≤ 𝑏
The sense of increasing t is called the positive sense on C. We assume C to be a piecewise
smooth curve, that is they consist of finitely many smooth curves joined end to end.
Basic properties of line integrals

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 246

 Applied Mathematics III

1.  (c f ( z)  c g ( z))dz  c 
C
1 2 1
C
f ( z )dz  c2  g ( z )dz (Linearity)
C

2. 
C
f ( z )dz   f ( z )dz (Sense reversal)
C

3. 
C
f ( z )dz   f ( z )dz   f ( z )dz where C=𝐶1 ∪ 𝐶2
C1 C2
( Partition of path)

Complex integral is rich in methods for evaluating integrals. We shall discuss the first two in this
sections and other follows in the section 5.4.
Before we state the first theorem let's define a terminology called simple connected domain
which is very important in evaluating integrals of complex functions.
Definition 5.22. A domain D is called simply connected if every simple closed curve(closed
curve without self intersections) encloses only points of D.
For example , a circular disk is simply connected ,whereas an annulus is not simply
connected(why?).
First method
Integration by the use of the path
Now we shall see the second method of evaluating complex integral that is not restricted to
analytic function but applied to any continuous complex function.
Theorem 5.10 Let C be a piecewise differentiable curve given by the equation z=z(t),where
𝑎 ≤ 𝑡 ≤ 𝑏.Let f(z) be a continuous complex valued function defined in a region containing
C.Then
b
 C
f ( z )dz   f ( z (t )) z ' (t )dt
a

Proof. Exercise
Steps in Applying Theorem 5.10
(A) Represent the path C in the form z(t) (a≤ 𝑡 ≤ 𝑏)
𝑑𝑧
(B) Calculate the derivative 𝑑𝑡

(C) Substitute z(t) for every z in f(z) (hence x(t) for x and y(t) for y).
(D) Integrate f(z(t)𝑧 ′ (𝑡) over t from a to b
Note that the orientation of the curve is counterclockwise if it is not specified.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 247

 Applied Mathematics III

1
Example 1.Using theorem Prove that C z p
dz =2𝜋𝑖 where C is the circle of radius r and center

p, i.e,|𝑧 − 𝑝| = 𝑟
Solution . The parametric representation of C is given by z(t)=𝑝 + 𝑟𝑖𝑒 𝑖𝑡 and 0≤ 𝑡 ≤ 2𝜋.
it
1 2 ire dt 2
 dz =    idt  it 0  2i
2
Since dz=ir𝑒 𝑖𝑡 𝑑𝑡,we have
C z p 0 re it 0

1
In particular, if p=0, 
C z
dz  2i

Example 2.Evaluate  zdz where C is the right hand half of a circle of radius 2 center at the
C

origin.
−𝜋 𝜋
SOLUTION. The parametric representation of C is given by z(t)=2𝑒 𝑖𝑡 ( 2 ≤ 𝜗 ≤ 2)

From z=-2𝑖 𝑡𝑜 𝑧 = 2𝑖 so that

2

𝑧̅ = ̅̅̅̅̅
2𝑒 𝑖𝑡 = 2𝑒 −𝑖𝑡 and dz=2𝑖𝑒 𝑖𝑡 dt
𝜋 𝜋
Therefore  zdz =∫
C
2
−𝜋
2
2𝑒 −𝑖𝑡 𝑖𝑡
2𝑖𝑒 𝑑𝑡=4∫ 𝑑𝑡 = 4𝜋𝑖
2
−𝜋
2
−𝜋 𝜋
It is also possible to write z(t)=2𝑒 𝑖𝑡 =2(cos 𝑡 + isin 𝑡) ≤𝜗≤ and 𝑧̅ =2(cos 𝑡 − isin 𝑡)
2 2

EXAMPLE 3. Evaluate  C
f ( z )dz where f(z)=y-x-3𝑥 2 i C is the line segment from z=0 to

z=1+i.
Solution. The equation of the line segment C determined by the points z=0 and z=1+i is given
by y=x (check it).The parametric equation of C can be given as x= t & y= t where 0≤ 𝑡 ≤ 1 so
that
z(t)=x(t)+i y(t)= t+it and dz=(1+i)dt. Now f(t)= t-t-3𝑡 2 = −3𝑖𝑡 2 and 1+i

1 𝑡3 1
C
f ( z )dz =∫0 −3𝑖𝑡 2 (1 + 𝑖)𝑑𝑡 = −3 𝑖(1 + 𝑖) [ =
3 0
1−𝑖

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 248

 Applied Mathematics III

Theorem 5.9 (Independent of path)

Let f(z) be an analytic in a simply connected domain D .Then there exists an indefinite integral
of f(z) in the domain D, that is, an analytic function F(z) such that 𝐹 ′ (z)=f(z) in D,and for all
paths joining two points 𝑧0 and 𝑧1 in D we have
𝑧1
∫ 𝑓(𝑧)𝑑𝑧 = 𝐹(𝑧1 ) − 𝐹(𝑧0 )
𝑧0

Note that we can write 𝑧0 and 𝑧1 instead of C ,since we can get the same result for all those C
from 𝑧0 to 𝑧1 .
If f(z) is entire , we can take for D the complex plane (which is certainly simply connected)
Quick exercises
1. Evaluate  x  2 yi dz
C
where C is the parabola y=𝑥 2 from z=1+2i to z=2+4i

1 1
2. .Evaluate ∫−1 𝑧 𝑑𝑧 along the curve C

i. the upper half of the unit circle from 1 to -1.

ii. the lower half of the unit circle from 1 to -1.
3. 
C
ydz
where C is the union of the line segment joining 0 to I then to i+2 .
1 0 𝑖𝑓 𝑛 ≠ 1
4. Show that  C (z  k) n
dz  {
2𝜋𝑖 𝑖𝑓 𝑛 = 1
where C is the unit circle with center k

and radius r and n is an integer.

Bounds for Integrals. ML-Inequality

We turn now to an inequality involving contour integrals that is extremely important in various
applications.
Definition 5.23. Let C be a piecewise differentiable curve given by the equation z=z(t) where
a≤ 𝑡 ≤ 𝑏. Then the length L of C is defined by
𝑏
L=∫𝑎 |𝑧 ′ (𝑡)| 𝑑𝑡
EXAMPLE 1. Find the length L of the circle C with center z0 and radius r.
Solution. The parametric equation of C is given by he equation z(t)= z0+ r𝑒 𝑖𝑡 where
0≤ 𝑡 ≤ 2𝜋 and 𝑧 ′ (𝑡)= ir𝑒 𝑖𝑡 .

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 249

 Applied Mathematics III

∴ 𝐿 = ∫ |𝑧 ′ (𝑡)| 𝑑𝑡 =
0

THEOREM 5.11.(Absolute value inequality)

If f(t)=u(t)+iv(t) is a continuous function of 𝑡ℎ𝑒 𝑟𝑒𝑎𝑙 𝑝𝑎𝑟𝑎𝑚𝑒𝑡𝑒𝑟 𝑡 ,then
𝑏 𝑏

|∫ 𝑓(𝑡)𝑑𝑡| ≤ ∫|𝑓(𝑡)|𝑑𝑡
𝑎 𝑎

Theorem 5.12 (ML inequality)

If f(z)=𝑢(𝑥, 𝑦) + 𝑙𝑣)𝑥, 𝑦) is continuous on the contour(simple closed path) C,then

|  f ( z )dz | ≤ 𝑀𝐿
C

Where L is the length of the contour C & M is the upper bound of the modulus |𝑓(𝑧)| ,that
is |𝑓(𝑧)| ≤ 𝑀
𝑏 𝑏
Proof. |  f ( z )dz | = |∫𝑎 𝑓(𝑧(𝑡)𝑧 ′ (𝑡)𝑑𝑡| ≤ ∫𝑎 |𝑓(𝑧(𝑡)𝑧 ′ (𝑡)|dz (why?)
C

𝑏 𝑏
≤ ∫𝑎 𝑀|𝑧 ′ (𝑡)| 𝑑𝑧 = 𝑀 ∫𝑎 |𝑧 ′ (𝑡)|𝑑𝑧 = 𝑀𝐿 (𝑤ℎ𝑦)
EXAMPLE 2. Let C be the arc of the circle |𝑧|=2 from z=2 to z=2i that lies in the first
z4 6𝜋
quadrant. Show that |  dz | ≤
C z 3 1 7

Solution. |𝑧 + 4| ≤ |𝑧| + 4 = 2 + 4 = 6 (𝑤ℎ𝑦) 𝑎𝑛𝑑 |𝑧 3 − 1| ≥ |𝑧|3 − 1 = 23 − 1 = 7 (𝑤ℎ𝑦)

𝑍+4 |𝑍+4| 6
.When z lies on C ,|𝑍 3−1| ≤ |𝑍 3−1| ≤ 7 = 𝑀 and since the length of the curce C is 𝐿 = 𝜋(𝑤ℎ𝑦)

then
z4 6𝜋
| dz | ≤ 𝑀𝐿 =
C z 1
3
7
Quick exercise
dz
1. Without evaluating the integral ,show that |  dz | ≤ 4√2 where C is the line segment
C z4
joining z=i to z=1

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 250

 Applied Mathematics III

5.3.2 Cauchy’s Integral Theorem

Cauchy’s integral theorem is the most important theorem in this chapter and fundamental
throughout complex analysis.Before we state the Theorem let us discuss some points about
simple connectedness which we have seen in the previous discussion.
1. A simple closed path is a closed path that does not intersect or touch itself as shown in Fig.
345. For example, a circle is simple, but a curve shaped like an 8 is not simple.

Fig 5.8 . Closed paths

2. A simply connected domain D in the complex plane is a domain such that every simple
closed path in D encloses only points of D.
Examples: The interior of a circle (“open disk”), ellipse, or any simple closed curve.
A domain that is not simply connected is called multiply connected.
Examples: An annulus , a disk without the center, for example, 0 < |𝑧| < 1.

Fig 5.9.Simply and multiply connected domain.

If w=f(z) where w= 𝑢 + 𝑖𝑣 and z= 𝑥 + 𝑖𝑦 ,then dz= 𝑑𝑧 + 𝑖𝑑𝑦 and

∫ 𝑓(𝑧)𝑑𝑧 = ∫(𝑢 + 𝑖𝑣)(𝑑𝑥 + 𝑖𝑑𝑦) = ∫ 𝑢𝑑𝑥 − 𝑣𝑑𝑦 + 𝑖 ∫ 𝑣𝑑𝑥 + 𝑢𝑑𝑦

If C is a simple closed curve as the path of integration ,then

C
f ( z )dz   (udx  vdy)  i  vdx  udy
C C

Applying Green’s theorem to each of the two integrals on the right hand side in turn, we have

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 251

 Applied Mathematics III

 (udx  vdy =  (v

C
D
x  u y )dxdy where D is the region enclosed by the curve. Also if f(z) is

analytic at every point within and on C, then Cauchy-Riemann equations give

𝑢𝑦 = −𝑣𝑥 and therefore −𝑣𝑥 − 𝑢𝑦 = 0

∴ ∫ 𝑢𝑑𝑥 − 𝑣𝑑𝑦 = 0 (𝑖)

Similarly with the second integral, we have

∫ 𝑣𝑑𝑥 + 𝑢𝑑𝑦 = 0 (ii)
Combining the results (i) & (ii) we have the following theorem called Cauchy integral theorem
Theorem 5.13 . Cauchy’s Integral Theorem
If f is analytic in a simply connected domain D, then for every simple closed path C in D,

C
f ( z )dz  0

 sin zdz  0, e 0 ,  (4 z 3  3z  1)  0 for any closed path, since

4z
Examples 1.
C C C

these functions are analytic for all z(entire functions).

Example 2. Points outside the Contour Where f (x) is Not Analytic
1
 sec zdz  0 ,
C z 2
9
dz

1 ±𝜋 ±3𝜋 ±5𝜋
where C is the unit circle, secz = 𝑐𝑜𝑠𝑧 is not analytic at z= , , ,….but all these points lie
2 2 2

outside C; none lies on C or inside C. Similarly for the second integral, whose integrand is not
analytic at z=±3𝑖.
2 z sin z
Example 3. If C denotes any closed contour lying in the open disk |𝑧| < 4 , then C z 2  25
dz =0.

This is because the disk is a simply connected domain and the two singularities z = ±5i of the
integrand are exterior to the disk.
Remark. Analyticity is sufficient but not necessary condition for Cauchy theorem to be true.
1 1
Example 4. 
C ( z  1) 2
 0 where C is a circle of radius 2 and center 1.But f(z)=(𝑧−1)2 is not

analytic at z=1.
Quick exercise.
Apply the Cauchy–Integral theorem to show that

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 252

 Applied Mathematics III


f ( z )dz  0 
C
when the contour C is the unit circle |z| = 1, in counterclockwise direction, and when
𝑧 1
1. 𝑓(𝑧) = 𝑧 2 +2𝑧+2 2. 𝑓(𝑧) = 𝑠𝑒𝑐ℎ 𝑧 3. f(z)=tan(4 𝑧)

The following theorem is an adaptation of the Cauchy Integral theorem to multiply connected
domains.
5.3.3 Cauchy’s Integral Formula
Cauchy’s integral theorem leads to Cauchy’s integral formula which is very important in
i. evaluating integrals as we shall see in this section
ii. proving that analytic functions have derivatives of all orders and
iii. Showing that all analytic functions have a Taylor series representation.
Theorem 5.15. Cauchy’s Integral Formula
Let f(z)be analytic in a simply connected domain D. Then for any point z 0 in D and any simple
closed path C in D that encloses z0 (Fig.5.13),
f ( z)
C z  z0
 2if ( z 0 ) ( 1) (Cauchy integral formula)

the integration being taken counterclockwise. Alternatively (for representing f(z0) by a contour
integral, divide (1) by 2𝜋𝑖),
𝟏 f ( z)
f(z0)=𝟐𝝅𝒊  (Cauchy’s integral formula).
C z  z0

Fig 5.13 . Cauchy's integral theorem

Example 1.
cos z

C z 
 2i cos z[ z   2i

for any contour enclosing z0=π (since cos z is entire), and zero for any contour for which z0=π
lies outside (by Cauchy’s integral theorem).
Example 2.
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 253
 Applied Mathematics III

1 2
z 7
z  21
2
1 190
C 3z  i dz  
3  2i z 2  7[ i  
i 3 z  27
z 3
3
𝑖
When z0=3 lies inside C.

zdz
Example 3. Evaluate 
C (9  z 2 )( z  i )
Let C be the positively oriented circle|𝑧| = 2.since the

𝑧
function f(z)=9−𝑧 2 is analytic within and on C and since the point z0= −i is interior to C, the

integral formula tells us that

z
zdz 9  z2 i 
C (9  z 2 )( z  i) = C 9  (i) dz  2i( 10 )  5
z 3  1dz
Example 4. Evaluate C z 2  4 where C is a circle of radius 2 and center 1,that is, |𝑧 − 1| = 2.
Solution. Here the circle C encloses the singular point 2 but not -2 and
𝑧 3 +1 𝑧 3 +1⁄ 3
= 𝑧+2
so that f(z) = 𝑧 + 1⁄𝑧 + 2 which is analytic at 2, then by the Cauchy integral
𝑧 2 −4 𝑧−2

theorem

z 3  1dz 𝑧 3 + 1⁄ 9𝜋𝑖
C z 2  4 =2𝜋𝑖 ( 𝑧 + 2 ) [ 𝑧=2 = 2

Quick exercise:
𝑍2
1. Integrate by Cauchy’s Integral formula counterclockwise around the circle.
𝑍 2 −1
𝜋
𝑖. |𝑧 + 1| =1 𝑖𝑖. |𝑧 + 𝑖| = 1.4 𝑖𝑖𝑖. |𝑧 − 1 − 𝑖| = 2

2. Integrate the following counterclockwise around the given curve C

z2
i) C z2
dz C : Z  1  2

dz
ii )  2 C : 4 x 2  ( y  2) 2  4
C z 4

cosh 2 ( z  i )
iii )  dz C the boundary of the square with vertices  2,4i
C z  i
z
iv )  2 dz , C : z  1  4 (Use partial fraction )
C z 4

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 254

 Applied Mathematics III

Group Activity
Evaluate the Integrals 1-15 using either the first or second method.
1
 Re z dz , C the parabola y 1  2 ( x  1) from 1  i to 3  3i.
2
1.
C

z
2.  e 4 dz , C the shortest path from 4i to 8i
C

3.  cos 2 z dz , C the semicircle

C
z   , x  0 from  i to i.

 i
 sec
2
4. z dz , C any path from to .
C
4 4

 Im z
2
5. dz , counterclo ckwise around the triangle with vertices 0,1, i.
C

 (z  z
1
6. ) dz , C the uni tcircle , counterclo ckwise
C

1 6
7.  ( z  2i  ( z  2i)
C
2
)dz , C the circle z  2i  4, clockwise.


8.  z exp z dz where C is the line segment from  1  2 i to 2  i
C

9.  z cos z dz where C is the line segment from 0 to i

C

1 z
10.  dz where C is the line segment from 1to i
C
z

dz
11.  where C is the line segment from 2 to 2  i
C
z z
2

12.  sin 2 z dz where C is the line segment from 0 to i.

C

13.  Re z 2 dz , clockwise aroud the boundary of the square with vertices 0, i ,1  i,1
C

z dz , where C is given by C : z (t )  t  i t 2 , 0  t  1
2
14. Evaluate
C

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 255

 Applied Mathematics III

5.3.4 Derivative of analytic functions

In real calculus, a real function is differentiable once does not necessarily imply that it is twice
differentiable nor that any of its higher derivatives exist. Unlike real functions, complex analytic
functions have derivatives of all derivatives which is resulted from Cauchy integral formula.
Theorem 5.16 Derivatives of an Analytic Function
If f(z)is analytic in a domain D, then it has derivatives of all orders in D, which are then also
analytic functions in D. The values of these derivatives at a point z0 in D are given by the
formulas
n! f ( z) f(z) 2if (n) ( z 0 )
2i C ( z  z 0 ) n1 C (z - z 0 ) n1
f ( n) ( z0 )    (n  1,2,3....);
n!
here C is any simple closed path in D that encloses z0 and whose full interior belongs to D; and
we integrate counterclockwise around C (Fig.5.14).

Fig.5.14

Example 1. For any contour enclosing the point 2𝜋𝑖 (counterclockwise)

ez
C ( z  2i) 2 dz  2i(e )  2ie 2i  2i
z '
z  2i

EXAM PLE2.For any contour enclosing the point -2i we obtain by counterclockwise integration
z 3  2 z 2  1 2i 3
C ( z  2i) 3  2! (z  2 z  1)  12i  4
2 ''
z  2 i

Example3.Evaluate
cos z
C z ( z 2  9)
2
dz where C is a circle of radius 3 and center 1 counterclo ckwise , that is |𝑧 − 1|=3.

cos z
Solution. The function is not analytic at z=0 and 𝑧 = ±3𝑖 and the contour C encloses
z ( z 2  9)
2

𝑐𝑜𝑠𝑧
the point 0 but not ±3𝑖.If we let f(z)=𝑧 2 +9 , then by the Cauchy Integral domain

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 256

 Applied Mathematics III

cos z 𝑐𝑜𝑠𝑧 ′ − sin 𝑧(𝑧 2 +9)−2𝑧𝑐𝑜𝑠 𝑧

C z 2 ( z 2  9)dz =2𝜋𝑖(𝑧 2+9) z  0 =2𝜋𝑖 ( (𝑧 2 +9)2
) z 0 =0

Example 4.Evaluate
ze z
C 4 z  i 3 dz where C is z  1
1 z
ze
ze z ze z 1 z
Solution.  dz =  dz  
64 dz so that f(z)= ze and
C 
4 z  i     
3 3 3
C
 C
 64
 4( z  i   z  i
 4   4 

1 z 𝜋𝑖
𝑓 ′′ (𝑧) = e (2  z ) . The singular point z = −. is also inside C.
64 4

ze z 𝜋 1 i i 1 2 2 i
Therefore C 4 z  i 3 dz =2𝜋𝑖(𝑓 ′′ (− 4 ) = 64 e 4 (2  4 )  64 ( 2  2 )(2  4 ) .
Quick exercise . Evaluate then following integrals
sin z
1. C z4
where C is a unit circle counterclockwise

1
2. C i
dz , where C is a unit circle ,
( z  ) 2 ( z  2i) 2
3
z 3  sin z
3. C ( z  i) 3 dz , where C is the boundary of the square whose vertices are ±2, ±2𝑖
counterclockwise.
2
ez
4. C dz ,C:|𝑧 − 3𝑖| = 2 , clockwise
z ( z  2i ) 2
We turn now to the most powerful consequences of the Cauchy integral formula .
Theorem 5.17. CAUCHY’S INEQUALITY
Let 𝑧0 be a fixed complex number. If f is analytic within and on a circle |𝑧 − 𝑎| = 𝑅 in
counterclockwise direction , then
n!M
f ( n) ( z0 ) 
Rn

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 257

 Applied Mathematics III

Proof. Applying the ML-inequality with |𝑓(𝑧)| ≤ 𝑀 on C and the length of the given circle is
2πR ,we obtain from derivative of analytic function
𝑛! 𝑀. 2𝜋𝑅 𝑛! 𝑀
|𝑓 (𝑛) (𝑧0 | ≤ = 𝑛
2𝜋 𝑅 𝑛+1 𝑅
Now let us proof the famous theorem on entire function called Liouville’s theorem using the
above Cauchy’s Inequality theorem.
T HEOREM 5.18. Liouville’s Theorem
If an entire function is bounded in absolute value in the whole complex plane, then this function
must be a constant.
Proof. By assumption, |𝑓(𝑧)| is bounded, say |𝑓(𝑧)| ≤ 𝑘, for all z. Using Cauchy’s inequality,
𝑘
we see that |𝑓 ′ (𝑧0 )| ≤ 𝑅 . Since f(z) is entire, this holds for every r, so that we can take r as large

as we please and conclude that 𝑓 ′ (𝑧0 ) = 0 .Since 𝑧0 is arbitrary, 𝑓 ′ (𝑧) = 𝑢𝑥 + 𝑖𝑣𝑥 = 0 for all z
(why?). Thus u= const, v= const and f= 𝑢 + 𝑖𝑣 = const for all z. This completes the proof.
The following theorem is also another very interesting consequence of Cauchy Integral formula
T H E O R E M 5.19 Morera’s Theorem (Converse of Cauchy’s Integral Theorem)
If is continuous in a simply connected domain D and if

C
f ( z )dz  0

for every closed path in D, then f(z) is analytic in D.

Group Activity

tan z

1. 2 dz, C the boundary of the square whose sides lie along the line lines x  2 and y  2
C ( z  x0 ) 2
and -2< x0 < 2.
cosh z
2. C z4
dz, C : z  1

exp( z 2 )
3. C ( z  1) 4 dz , C : z  3  2
1
4. C ( z  1) 2
2
dz where C:|𝑧 − 3| = 3 counterclokwise

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 258

 Applied Mathematics III

4z 3  6
5.  C z( z  1  i) 2
dz, C consists of z  3 counterclo ckwise and z  1 clockwise.

1
 C z ( z 2  9)
2
dz, where C is

3
6. (i ) z  (ii ) z  1
2
3
(iii ) z  2i  (iv ) z  2i  3 (v) z  3
2
5.4 Complex series,Taylor and Laurent series, residues and poles
In the next section we shall see a new approach of evaluating complex integral called Residue
integral. But this method needs a thorough understanding of Taylor and Laurent series.
Therefore, we shall discuss these series in such a way that they play a leading role in evaluating
complex integral using residues.
Definition 5.24. If a complex function f is analytic at z=0, then the Maclaurin series expansion
of f is
∞
′ (0)
𝑧 2 𝑓 ′′ (0) 𝑧 2 𝑓 ′′ (0) 𝑓 (𝑛) (0)𝑧 𝑛
𝑓(0) + 𝑧𝑓 + + +⋯= ∑
2! 2! 𝑛!
𝑛=0

Everything that has been said about the Maclaurin series expansion of an expression involving a
real variable x can equally be said about an expression involving a complex variable z.
𝑧3 𝑧5 𝑧 2𝑛+1
Example 1. The Maclaurin series expansion of sinz=z− 3! + − ⋯ ((−1)𝑛 (2𝑛+1)!
5!

Because
f(z)=sinz and f(0)=0
𝑓 ′ (𝑧) = 𝑐𝑜𝑠𝑧 and 𝑓 ′ (0) = 1
𝑓′′ (𝑧) = −𝑠𝑖𝑛𝑧 and 𝑓 ′′ (0) = 0
𝑓 ′′′ (𝑧) = −𝑐𝑜𝑠𝑧 and 𝑓 ′′′ (0) = −1
.
.
.
Therefore, the Maclaurin series expansion of f is

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 259

 Applied Mathematics III

′ (0)
𝑧 2 𝑓 ′′ (0) 𝑧 2 𝑓 ′′ (0) 𝑧3 𝑧5 (−1)𝑛 𝑧 2𝑛+1
𝑓(0) + 𝑧𝑓 + + +⋯=z− + −⋯= ∑
2! 2! 3! 5! (2𝑛 + 1)!
𝑛=0

Furthermore, applying the Ratio test the series is valid for all finite values because
𝑎𝑛+1 (𝑧)
lim𝑧  0 | |= 0 < 1 (verify it)
𝑎𝑛 (𝑧)

∴ The expansion is valid for all z by the Ratio test .

Example 2. Evaluate the Maclaurin series expansion of f(z)=ln(1+z)
Solution.
f(z)=ln(1+z) & f(0)=0
1
f’(z)=1+𝑧 & f’(0)=1
1
f’’(z)=− (1+𝑧)2 & f’’(0)= -1
2
f(3)(z)=(1+𝑧)3 & f(3)(z)=2

.
.
.
𝑓 (𝑛) (𝑧) = (−1)𝑛+1 𝑛! (1 + 𝑧)−𝑛 and so 𝑓 (𝑛) (0) = (−1)𝑛+1 𝑛!
Therefore, the required Maclaurin series expansion is
∞
𝑧2 𝑧3 (−1)𝑛+1 𝑧 𝑛
𝑧− + −⋯= ∑
2 3 𝑛
𝑛=1
𝑎𝑛+1 (𝑧)
since lim𝑧  0 | |=|𝑧|, then the series is valid for |𝑧| < 1 by the ratio test
𝑎𝑛 (𝑧)

Quick exercise. Show that

𝑧𝑛 1
1. 𝑒 𝑧 = ∑∞
𝑛=0 𝑛! For all z. 3.1−𝑧 = ∑∞
𝑧=0 𝑧
𝑛

𝑧 2𝑛 1
2.cosz= ∑∞ 𝑛 4. 1+𝑧 = ∑∞ 𝑛 𝑛
𝑧=0(−1) 𝑧
𝑛=0(−1) (2𝑛)!

The followings are important Maclaurin series that can be verified .

𝑧 2𝑛 𝑧2 𝑧4
1. Coshz=∑∞
𝑛=0 (2𝑛)! = 1 + + + ⋯ for all z
2! 4!

𝑧 2𝑛+1 𝑧3 𝑧5
2. Sinhz =∑∞
𝑛=0 (2𝑛+1)! = 𝑧 + + 5! +…. For all z.
3!

𝑧 2𝑛+1 𝑧3 𝑧5
3. arc tanz =∑∞ 𝑛
𝑛=0(−1) 2𝑛+1 = 𝑧 − + − ⋯ |𝑧| < 1
3 5

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 260

 Applied Mathematics III

4. Binomial series:
∞
1 −𝑚 𝑛
= (1 + 𝑧)−𝑚 = ∑ ( )𝑧
(1 + 𝑧) 𝑚 𝑛
𝑧=0

𝑚(𝑚 + 1) 2 𝑚(𝑚 + 1)(𝑚 + 2)

= 1 − 𝑚𝑧 + 𝑧 − +⋯
2! 3!
Radius of convergence
We have seen that the Maclaurin series expansion of f(z)=ln(1+z) is valid for |𝑧| < 1.This
inequality defines the interior of a circle of radius 1 and center at the origin, namely
z=𝑒 𝑖𝜃 (𝐹𝑖𝑔.5.15)

Fig 5.15
This means that the expansion is valid for all z within this circle. The radius of the circle within
which a series expansion is valid is called the radius of convergence and the circle is called the
circle of convergence.
Taylor series
Theorem 5.20. Let f(z) be analytic inside and on a simple closed curve C.Then the Taylor
expansion of f(z) about the point 𝑧0 which is interior to C is given by
𝑓 (𝑛) (𝑧0) 𝑓 ′ (𝑧0 )(𝑧−𝑧0 ) 𝑓 ′′ (𝑧0 )(𝑧−𝑧0 )2 𝑓 ′′′ (𝑧0 )(𝑧−𝑧0 )3
f(z)=∑∞
𝑛=0 (𝑧 − 𝑧0 )𝑛 =𝑓(𝑧0 ) + + + + ⋯+
𝑛! 1! 2! 3!
𝑓 (𝑛) (𝑧0)
(𝑧 − 𝑧0 )𝑛 …
𝑛!

where 𝑧0 is the center of the circle of convergence and the center of convergence is given by
|𝑧 − 𝑧0 | = 𝑅 where R is the radius of convergence.
Remark. The Maclaurin series is a special case of Taylor series about 𝑧0 =0.
1
Example 1.The Taylor series of (z) = 𝑧 at z=1 is given by

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 261

 Applied Mathematics III

1 𝑓 ′ (1) 𝑓 ′ (1)(𝑧−1) 𝑓 ′′ (1)(𝑧−1)2 𝑓 ′′′ (1)(𝑧−1)3

= f(1) + + + + +⋯
𝑧 1! 1! 2! 3!

Now
1
f(z) = 𝑧  f(1)=1
1
𝑓 ′ (𝑧) = − 𝑧 2  𝑓 ′ (1) = −1
2
𝑓 ′′ (𝑧) = 𝑧 3  𝑓 ′′ (1) = 2
6
𝑓 ′′′ (𝑧) = − 𝑧 4  𝑓 ′ (1) = −6

.
.
.
1
Therefore the Taylor expansion of about z=1is
𝑧
∞
1
= 1 − (𝑧 − 1) + (𝑧 − 1)2 − (𝑧 − 1)3 + ⋯ = ∑(−1)𝑛 (𝑧 − 1)𝑛
𝑧
𝑛=0
𝑎𝑛+1
and lim𝑛→∞ | | = |𝑧 − 1| so that its circle of convergenc is |𝑧 − 1| < 1.
𝑎𝑛

We can also do this problem in the following way.

1 1 1
= 1+(𝑧−1) = ∑∞ 𝑛 𝑛
𝑛=0(−1) (𝑧 − 1) since = ∑∞ 𝑛 𝑛
𝑧=0(−1) 𝑧
𝑧 1+𝑧
𝑧−2
Example 2. Express f(z)= 𝑧+2 as a Taylor series about the point z=2 and determine its regions of

convergence.
𝑧−2 𝑧−2 𝑧−2 𝑧−2 1 𝑧−2 (𝑧−2)𝑛
Solution. f(z)= 𝑧+2 = 4+(𝑧−2) = 𝑧−2 = 𝑧−2 = ∑∞
𝑛=0(−1)
𝑛
(𝑤ℎ𝑦? )
4(1+ ) 4 1+ 4 4𝑛
4 4

(𝑧−2)𝑛+1
=∑∞
𝑛=0(−1)
𝑛
4𝑛+1
𝑧−2
By the ratio test the region of convergence is given by | | < 1 which is the circular disc
4

|𝑧 − 2| < 4.
Example 3. Find the Taylor series of f(z)=𝑧 4 𝑒 3𝑧 at z=0 and determine its region of convergence.
𝑧𝑛 (3𝑧)𝑛 3𝑛 𝑧 𝑛
Solution. Since 𝑒 𝑧 = ∑∞
𝑛=0 𝑛! , 𝑒
3𝑧
= ∑∞
𝑛=0 = ∑∞
𝑛=0 .Therefore
𝑛! 𝑛!
3𝑛 𝑧 𝑛+4
𝑧 4 𝑒 3𝑧 = ∑∞
𝑛=0 and by the ratio test the series converges for all z,i.e |𝑧| < ∞.
𝑛!

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 262

 Applied Mathematics III

Quick exercise. Find the Taylor series of the following functions and determine its region of
convergence.
𝑧
1. f(z)= 𝑧+5 about the point z =1
𝜋
2. f(z)=cosz with center z= 3

3. f(z)=sinh(2z-i) with center z= 𝑖⁄2

Laurent Series
If a function f fails to be analytic at a point 𝑧0, one can’t apply the Taylor theorem at that point.
It is often possibly ,however to find a new series representation of f involving both positive and
negative powers of (z-z0)(or constant) called Laurent series. Laurent series is the generalization
of Taylor series.
Definition 5.25. A Laurent series is a series of positive and negative integers powers of (z-z0)
by which we can represent a given function f(z) in an annulus (a circular ring with center z0) in
which f(z) is analytic, f(z) may have singularities outside the ring as well as in its hole.
The series (or finite sum of the negative powers is called the principal part(or singular part) of
the Laurent series. The principal part is used
 for the classification of singularities
 in powerful integration method(Residue integration) that we shall discuss in the next
section.
Theorem 5.21. (Laurent’s Theorem )
Let f(z) be analytic in a domain containing two concentric circles and with center z0 and the
annulus between them ( Fig.5.16). Then f(z) can be represented by the Laurent series
𝒃
𝒇(𝒛) = ∑∞ ∞ 𝒏
𝒏=𝟎 𝒂𝒏 + ∑𝒏=𝟏 {𝒛−𝒛 𝒏 =
𝟎)

𝒃𝟏 𝒃𝟐
𝒂𝟎 + 𝒂𝟏 (𝒛 − 𝒛𝟎 ) + 𝒂𝟐 (𝒛 − 𝒛𝟎 )𝟐 + … . + + +⋯ (𝟏)
𝒛−𝒛𝟎 (𝒛−𝒛𝟎 )𝟐

consisting of nonnegative and negative powers. The coefficients of this Laurent series are given
by the integrals

1 f (z  ) 1
𝑎𝑛 = 2𝜋𝑖 C 
z  z0
dz  n=0, ±1, ±2, …, bn 
2I 
C
f ( z  )( z  z 0 ) n1 f ( z  )dz  (2)

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 263

 Applied Mathematics III

taken counterclockwise around any simple closed path C that lies in the annulus and encircles the
inner circle, as in Fig. 5.16.
This series converges and represents f(z) in the enlarged open annulus obtained from the given
annulus by continuously increasing the outer circle C1 and decreasing C2 until each of the two
circles reaches a point where f(z) is singular.
In the important special case that z0 is the only singular point of f(z) inside C2 , this circle can be
shrunk to the point z0 giving convergence in a disk except at the center. In this case the series
(or finite sum) of the negative powers of (1) is called the principal part or singular part of f (z)
at z0 or of that Laurent series (1).

Fig 5.16 . Laurent theorem.

Examples :Find the Laurent series expansions of the following functions with the given centers.
𝑐𝑜𝑠𝑧 1
1. f(z)= ; z0=0 3.f(z)=𝑧(𝑧−1) ;z0=1
𝑧3
1
𝑒𝑧
2. f(z)=𝑧 3 𝑒 𝑧2 ; z0=0 4. 𝑓(𝑧) = (𝑧−1)2; z0=1
1
5. Develop 𝑓(𝑧) = 1−𝑧 a) in non-negative power of z b) in negative power of z.

Solutions.
𝑧 2𝑛
1. Since cos z= ∑∞ 𝑛
𝑛=0(−1) (2𝑛)! and f (z) is analytic at all points z≠ 0, we have

𝑐𝑜𝑠 𝑧 1 𝑧2 𝑧4 𝑧6 1 1 𝑧 𝑧3
= 𝑧 3 (1 − + − + ⋯)=𝑧 3 − 2𝑧 + 4 − +… (|𝑧| > 0)
𝑧3 2! 4! 6! 6!

Here the annulus of convergence is the whole complex plane without the origin and its principal
1 1
part is − 2𝑧
𝑧3
𝑧𝑛
2. Since 𝑒 𝑧 = ∑∞
𝑛=0 𝑛! for all z and f(z) is analytic at all points z≠ 0 we have

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 264

 Applied Mathematics III

1 1
2𝑛 1 1 1
f(z)=𝑧 3 𝑒 𝑧2 = 𝑧 3 ∑∞ 3
𝑛=0 𝑛! = 𝑧 (1 + 𝑧 2 +2 𝑧 4 + 6 𝑧 6 … for all z and its principal part is
𝑧

1 1 1
+ 6𝑧 3 + 24𝑧 5 + ⋯
2𝑧

3. Since f(z) is analytic at all points z≠ 0,1,then f is analytic in annulus 0 < |𝑧 − 1| < 1
and hence can be expressed as a Laurent series in powers of z-1 as
1 1 1 1
= 𝑧−1 (1+(𝑧−1)) = 𝑧−1 (1 − (𝑧 − 1) + (𝑧 − 1)2 − (𝑧 − 1)3 + ⋯
𝑧(𝑧−1)
1
=𝑧−1 − 1 + (𝑧 − 1) − (𝑧 − 1)2 + ⋯ since |𝑧 − 1| < 1
𝑒𝑧 𝑒 𝑢+1 𝑒𝑢 𝑒 𝑢2 𝑢3 𝑢4
4. Let z-1=u. Then 𝑓(𝑧) = (𝑧−1)2 = (𝑢)2
= 𝑒 𝑢2 = 𝑢2 (1 + 𝑢 + + + +⋯
2! 3! 4!

𝑒 (𝑧−1)2 (𝑧−1)3 (𝑧−1)4 𝑒 𝑒 𝑒 𝑒(𝑧−1)

= (𝑧−1)2 (1 + (𝑧 − 1) + + + + ⋯)= + (𝑧−1) + 2! +
2! 3! 4! (𝑧−1)2 3!

𝑒(𝑧−1)2
+ + ⋯ which is valid for 0 < |𝑧 − 1| < ∞.
4!
1
5.a) 1−𝑧 = ∑∞ 𝑛
𝑛=0 𝑧 which is valid for |𝑧| < 1 by the ratio test

1 −1 −1 1 1 1 1
𝑏) = 1 = ∑∞
𝑛=0 = − ∑∞
𝑛=0 𝑧 𝑛+1 = − 𝑧 − 𝑧 2 − ⋯ which is valid for |𝑧| > 1
1−𝑧 𝑧(1− ) 𝑧 𝑧𝑛
𝑧

Quick exercise.
1
1. Find the Laurent series that represents the function 𝑓(𝑧) = 𝑧 2 sin 𝑧 2 in the domain 0

< |z| < ∞.

3−2𝑧
2. Find the Laurent series representation of the function 𝑓(𝑧) = 𝑧 2−3𝑧+2 which is valid

for
i) |𝑧| < 1 ii) 1< |𝑧| < 2 iii) |𝑧| > 2
𝑧
3. Find the Laurent series of f(z)= (𝑧+1)(𝑧+2) with center z0= -1

Group Activity
1. Find the Laurent series representations of the following function that converges for 0<|𝑧| <
𝑅 and determine its region of convergence.
𝑠𝑖𝑛𝜋𝑧 1
a) f(z)= d) f(z)= 𝑧 3 𝑐𝑜𝑠ℎ 𝑧
𝑧2
1 𝑒𝑧
b) f(z)= 𝑧 2 −𝑧 3 e) f(z)= = 𝑧 2 −𝑧 3
𝑠𝑖𝑛ℎ3𝑧
c) f(z)= 𝑧2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 265

 Applied Mathematics III

2. Find the Laurent series representations of the following function that converges for 0<|𝑧 −
𝑧0 | < 𝑅 and determine its region of convergence.
𝑒𝑧 𝑒 2𝑧
a) f (z)=(𝑧−2)2 , 𝑧0 = 2 c) f(z)=(𝑧−1)2 , 𝑧0 = 1
cos 𝑧 1
b) f(z)=(𝑧−𝜋)2, 𝑧0 = 𝜋 d) f(z)=(𝑧 2 (𝑧−𝑖), 𝑧0 =i
𝑧−2
3. Find the Laurent series of (𝑧+2)(𝑧+5) that is valid for

a) 2< |𝑧| < 5 b) |𝑧| > 5 c) |𝑧| < 2

4. Show that for 0< |𝑧 − 1| < 2,
𝑧 (𝑧−1)𝑛 1
= −3 ∑∞
𝑛=0 − 2(𝑧−1)
(𝑧−1)(𝑧−3) 2𝑛+2
1
5. Write the two Laurent series in power of z that represent the function f(z)=𝑧(1+𝑧 2 ) in

the domains 0< |𝑧| < 1 and 1< |𝑧| < ∞

Singularities
Recall that a function f(z) is singular or has a singularity at a point z=z0 if f(z) is not analytic
(perhaps not even defined) at z=z0 but every neighborhood of z=z0 contains points at which f(z)
is analytic. We also say that is z=z0 singular point of f(z).
Definition 5.25: A singular point z=z0 is called isolated singular point of f(z) if it has a
neighborhood without further singularities of f(z) .Otherwise it is called non-isolated singularity.
𝜋 3𝜋 5𝜋
Example 1. f(z)= tan z has isolated singularities at ± 2 , ± ,± , 𝑒𝑡𝑐 .
2 2
1 sin1⁄
Example 2 . Let f(z)= tan 𝑧 = cos1⁄𝑧 .The singular points of f(z) are 0 and those points of z for
𝑧
1 1 𝜋 2
which cos 𝑧 = 0, 𝑡ℎ𝑎𝑡 𝑖𝑠 = (2𝑛 + 1) 2 𝑎𝑛𝑑 𝑡ℎ𝑒𝑛 𝑧 = (2𝑛+1)𝜋 , 𝑛 = 0, ±1, ±2, …
𝑧

But the singular point z = 0 is not isolated because every neighborhood of the origin contains
other singular points of the function .Therefore, z=0 is non-isolated singular point.
Laurent series expansion of the function f(z) can be used to classify the isolated singular points.
Let z=z0 be an isolated singular points of f(z).Then there exist a neighborhood of the point z 0
,0 < |𝑧 − 𝑧0 | < 𝑅 inside this f(z) is analytic.Hence in this region we can expand the function f(z)
as a Laurent series
𝑏
f(z)=∑∞ 𝑛 ∞ 𝑛
𝑛=0 𝑎𝑛 (𝑧 − 𝑧0 ) + ∑𝑛=1 (𝑧−𝑧 )𝑛
0

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 266

 Applied Mathematics III

valid in the neighborhood of the singular point z =𝑧0 except at z0 itself, that is ,in the region of
the form 0 < |𝑧 − 𝑧0 | < 𝑅.
Removable Singularity
Definition 5.26. If f (z) has an isolated singularity at z=z0 but lim𝑧→𝑧0 𝑓(𝑧) exists , then the point
z=z0 is called a removable singularity.
Note that in this case the principal part of the Laurent series is zero and lim𝑧→𝑧0 𝑓(𝑧) = 𝑎0

Example 1.
𝑠𝑖𝑛𝑧 1 𝑧3 𝑧5 𝑧2 𝑧4
Let f(z)= = 𝑧 (𝑧 − + −⋯= 1− + − ⋯ .Obviously , z=0 is an isolated
𝑧 3! 5! 3! 5!

singularity. Since the principal part is zero and lim𝑧→0 𝑓(𝑧) = 1 = 𝑎0 , then z=0 is a removable
singularity.
Alternatively(without using Laurent series expansion), since z=0 is the isolated singular point of
𝑠𝑖𝑛 𝑧
f and lim𝑧→0 = 1, then z=0 is the removable singularity.
𝑧
1−cos 𝑧
Example 2. The point z=0 is a removable singular point of the function f(z)=
𝑧
because
1 𝑧2 𝑧4 𝑧6 𝑧 𝑧3 𝑧5
f(z) = [1 − (1 − + − + − − −)] = − + − ----
𝑧 2! 4! 6! 2! 4! 6!
1−cos 𝑧
whose principal part is zero and moreover lim𝑧→0 = 0 by L'hopital rule.
𝑧
1−cos 𝑧
Alternatively, since z=0 is an isolated singular point of f and lim𝑧→0 exists (≠ 0), then z=0
𝑧

is the removable singularity.

POLE
Definition 5.27. If the principal part of the Laurent series of the function f(z) has only a finite
number of terms of the form,
𝒃𝟏 𝒃𝟐 𝒃𝟐
+
𝒛−𝒛𝟎 (𝒛−𝒛𝟎 ) 𝟐 + ⋯+ (𝒛−𝒛𝟎 )𝒏
where n is a finite integer and 𝑏𝑛+1 = 𝑏𝑛+2 = ⋯ = 0 then

z=z0 is called a pole of order. In particular, if n=1(Poles of the first order) , we call a simple
pole.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 267

 Applied Mathematics III

Alternatively, an isolated singular point z=z0 is said to be a pole if lim𝑧→𝑧0 |𝑓(𝑧)| = ∞ and
lim𝑧→𝑧0 (z − z0 )n f(z) exists and different from zero . The smallest values of n for which this
limit exists define the order of the pole.
Example 1. The function
2 𝑧−4
f(z) = 𝑧(𝑧+3)4 + (𝑧−5)2 (𝑧+6)3

has a simple pole at 0 , a pole of fourth order at -3, a pole second order at 5 and a pole of third
order at -6.In each case we can easily show that lim𝑧→𝑧0 (𝑧 − 𝑧0 )𝑛 𝑓(𝑧) ≠ 0.
Example 2. The function
sin 𝑧
f(z)= has an isolated singular point at z=0 and thus can be expressed by the Laurent series as
𝑧4
𝑠𝑖𝑛𝑧 1 𝑧3 𝑧5 1 1 𝑧 𝑧3 1 1
= 𝑧 4 (𝑧 − + − ⋯ = 𝑧 3 − 3!𝑧 + 5! − +⋯ whose principal part is − 3!𝑧.Therfore
𝑧4 3! 5! 7! 𝑧3

z=0 is a pole of third order.

In the above example 2,the isolated singular point z=0 is not a pole of order 4 because
sin 𝑧
lim𝑧→0 𝑧 4 ( )=0.
𝑧4
sin 𝑧 sin 𝑧
However , since lim𝑧→0 | | = ∞(𝑤ℎ𝑦) 𝑎𝑛𝑑 lim𝑧→0 𝑧 3 ( )=1≠ 0, then z=0 is a pole of
𝑧4 𝑧4

third order.
Essential singularity.
Definition 5.28. If the principal part of the Laurent expansion of f(z) has infinite number of
terms, then the point z=z0 is called an essential singular point of f(z).
1⁄
Example 1 . f(z)= 𝑒 𝑧 has an essential singularity at z=0 since the principal part of the Laurent
series expansion about z=0 has infinite number of terms. That is
∞ ∞
1 (1⁄𝑧)𝑛 1 1 1 1
𝑒 ⁄𝑧 =∑ = ∑ = 1 + + + +⋯
𝑛! 𝑛! 𝑧 𝑛 𝑧 2! 𝑧 2 3! 𝑧 𝑛
𝑛=0 𝑛=0

Quick exercises. Determine the location of the singularities of the the following function and for
pole state the order.
1−𝑐𝑜𝑠𝑧 1−𝑒𝑥𝑝2𝑧 1
1. 𝑓(𝑧) = 2. 𝑓(𝑧) = 3. 𝑓(𝑧) = sin 𝑧
𝑧2 𝑧5

Group Activity
1. Locate the singularities of the following functions and state the orders for the poles.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 268

 Applied Mathematics III

1 1⁄
1. (𝑧) = 𝑧 3 (𝑧 2 +9)2 4. 𝑓(𝑧) = 𝑧𝑒𝑥𝑝 𝑧

𝑧
2. 𝑓(𝑧) = 𝑡𝑎𝑛𝑧 5. 𝑓(𝑧) = sin 𝑧
𝑧2 cos 𝑧−cos 2𝑧
3. 𝑓(𝑧) = 𝑧−sin 𝑧 6. 𝑓(𝑧) = 𝑧4

5.5 Residue integration method

If a function f(z) is analytic inside and on a simple closed contour C, then by Cauchy integral

theorem, we have  f ( z )dz  0 . However, when the integral f(z) has one or more isolated
C

singular points inside a closed contour C, then the Cauchy integral theorem can’t be used and the

value of the complex integral C

f ( z )dz may not be zero. Each of the isolated singular points

inside C contribute to the value of the complex integral. These contributions are called residues.
Residues of analytic functions at an isolated singular point.

Definition 5.29. Let f(z) be analytic at all points z except at z=𝒛𝟎 .Let 𝒛𝟎 be an isolated singular
point. Then f(z) can be expanded as a Laurent series about z=𝒛𝟎 in the form
𝒃𝟏 𝒃
𝒇(𝒛) = 𝒂𝟎 + 𝒂𝟏 (𝒛 − 𝒛𝟎 ) + 𝒂𝟐 (𝒛 − 𝒛𝟎 )𝟐 + … . + + (𝒛−𝒛𝟐 )𝟐 + ⋯
𝒛−𝒛𝟎 𝟎

Which converges in some region R: 0 < |𝑧 − 𝑧0 | < 𝑟. The coefficient bn are defined by
1
bn 
2i C
f ( z  )( z  z 0 ) n1 f ( z  )dz 

Where n=1,2,3… and C is a simple closed curve in the annulus 0< |𝑧 − 𝑧0 | < 𝑟 .The
coefficient of (z-z0)-1 ,that is, b1 in the Laurent series expansion of f(z) is called Residue of f(z)
at z=z0 and written as

b1  Re s( f ( z ); z0 )

Therefore , if the residue b1 at z=z0 can be determined by some method then the value of the
contour integral is given by

 f ( z)dz  2ib . 1

sin 𝑧
Example 1. Integrate the function f(z)= counterclockwise around the unit circle C.
𝑧4

Solution. The Laurent series expansion of the series is

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 269

 Applied Mathematics III

1 1 1 𝑧3
− 3!𝑧 + 5! − + ⋯ which converges for |𝑧|>0 .From the series we can see that z=0 is a pole
𝑧3 7!
1 1
of order 3 and the residue b1=− 3! = − 6 .
𝜋
∴  C
f ( z )dz  2ib1 =− 3

Example 2.Integrate 𝑓(𝑧) = 𝑧 3 −𝑧 4 clockwise around the circle C: |𝑧|=1⁄2.

1

1 1
Solution. = z3 (1−z) has isolated singular points z=0 and z=1 of which z=1 lies outside C.
z3 −z4

So we need to find the residue of f(z) at 0. From its Laurent series

1 1 1 1
= 𝑧 3 + 𝑧 2 + 𝑧 + 1 + 𝑧 + ⋯ which converges in the annulus 0 < |𝑧| < 1 and from which
z3 −z4

the residue is 1.Since the integration is clockwise,

1

C z  z4
3
dz  2ib1  2

Method of finding the residues of f(z) at an isolated singular point.

1. Residue at removable singularity
If z=z0 is removable singular point, then there is no term in the principal part of the Laurent
series expansion of f(z).In this case b1=0 and

 f ( z)dz  2i(0)  0 .
𝑠𝑖𝑛3𝑧
Example . z=0 is a removable singular point of the function f(z)= because its Laurent series
𝑧
1 (3𝑧)3 (3𝑧)5 27𝑧 2 243𝑧 4
f(z)= 𝑧 (3𝑧 − + − ---) =3− + − --- has no principal part and thus
3! 5! 3! 5!

b1=0.
2. Residue at a simple pole
Let z0 be a simple pole of f(z).In this case
𝒃𝟏
𝒇(𝒛) = ∑∞ 𝒏
𝒏=𝟎 𝒂𝒏 (𝒛 − 𝒛𝟎 ) + .
𝒛−𝒛𝟎

Multiplying both sides by (z-z0) and take the limit as z→ 𝑧0 we get

b1=𝐥𝐢𝐦𝒛→𝒛𝟎 (𝒛 − 𝒛𝟎 )𝒇(𝒛).
4𝑧−3𝑖 4𝑧−3𝑖 4𝑧−3𝑖
Example 1.f(z)=𝑧 3 +4𝑧 = 𝑧(𝑧 2 +4) = 𝑧(𝑧+2𝑖)(𝑧−2𝑖) has simple poles 0 and ±2𝑖.Then
4𝑧−3𝑖 −3𝑖
Res(f(z);0)= 𝐥𝐢𝐦𝒛→𝟎 (𝒛) 𝑧(𝑧 2 +4) = 4

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 270

 Applied Mathematics III

4𝑧−3𝑖 5𝑖
Res(f(z);2i) 𝐥𝐢𝐦𝒛→𝟐𝒊 (𝒛 − 𝟐𝒊) =−8
𝑧(𝑧+2𝑖)(𝑧−2𝑖)
4𝑧−3𝑖 11𝑖
Res(f(z);-2i) 𝐥𝐢𝐦𝒛→−𝟐𝒊 (𝒛 + 𝟐𝒊) 𝑧(𝑧+2𝑖)(𝑧−2𝑖) = 8

A second formula for the residue at a simple pole is

p( z) p( z )
b1  Re sf ( z)  Re s  '
0

z  z0 q( z) q ( z ) 0
z  z0

𝑝(𝑧)
assuming that f(z)= 𝑞(𝑧) with p(z0)≠ 0 and f(z) has a simple pole at z0.

Proof.
𝑝(𝑧)
Res(f(z);z0)=lim𝑧→𝑧0 (𝑧 − 𝑧0 )(𝑓(𝑧) = lim𝑧→𝑧0 (𝑧 − 𝑧0 ) 𝑞(𝑧) =
𝑧−𝑧0 0 𝑧−𝑧 1
lim𝑧→𝑧0 𝑝(𝑧) lim𝑧→𝑧0 = lim𝑧→𝑧0 𝑝(𝑧) lim𝑧→𝑧0 𝑞(𝑧)−𝑞(𝑧 )
= 𝑝(𝑧0 ) 𝑞′ (𝑧 ) (why?)
𝑞(𝑧) 0 0

4𝑧−3𝑖
Example 2. From the above example f(z)=𝑧 3 +4𝑧 has simple poles at z=0 and z=±2𝑖 and its

corresponding residues are

4𝑧−3𝑖 4𝑧−3𝑖 𝟑𝒊
Res(f(z);0)= 𝐥𝐢𝐦𝒛→𝟎 𝑧 3 +4𝑧 = 𝐥𝐢𝐦𝒛→𝟎 3𝑧 2 +4 =− 𝟒
4𝑧−3𝑖 5𝑖
Res(f(z);2i) 𝐥𝐢𝐦𝒛→𝟐𝒊 3𝑧 2+4 = − 8
4𝑧−3𝑖 11𝑖
Res(f(z);-2i) 𝐥𝐢𝐦𝒛→−𝟐𝒊 =
3𝑧 2 +4 8

3. Residue of a pole of order n.

Let z=z0 be a pole of order m.In this case we write f(z) as a Laurent sries
∞
𝐛𝟏 𝐛𝟐 𝐛𝐦
𝐟(𝐳) = ∑ 𝐚𝐧 (𝐳 − 𝐳𝟎 )𝐧 + + 𝟐
+⋯+
𝐳 − 𝐳𝟎 (𝐳 − 𝐳𝟎 ) (𝐳 − 𝐳𝟎 )𝐦
𝐧=𝟎

Multiplying both sides by (𝒛 − 𝒛𝟎 )𝒏 we get ,

(z − z0 )m f(z)= ∑∞
𝒏=𝟎 𝒂𝒏 (𝒛 − 𝒛𝟎 )
𝒏+𝒎
+ 𝒃𝟏 (𝒛 − 𝒛𝟎 )𝒎−𝟏 + 𝒃𝟐 (𝒛 − 𝒛𝟎 )𝒎−𝟐 + ⋯
then differentiating m-1 times and taking the limit as z→ 𝑧0 we obtain
𝟏 𝒅𝒎−𝟏
𝒃𝟏 = (𝒎−𝟏)! 𝐥𝐢𝐦𝒛→𝒛𝟎 ((𝒛 − 𝒛𝟎 )𝒎 𝒇(𝒛).
𝒅𝒛𝒎−𝟏
𝑑𝑧
Example 1 .The function 𝑓(𝑧) = 𝑧(𝑧−2)3 has a pole of order three at z=2 (verify!) and its

residues there is

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 271

 Applied Mathematics III

1 𝒅𝟐 𝟏 𝟏 𝟐 𝟏
Res(f(z);2)=b1= (3−1)! 𝐥𝐢𝐦𝒛→𝟐 𝒅𝒛𝟐 ((𝒛 − 𝟐)𝟐 𝒛(𝒛−𝟐)𝟐 = 𝟔 𝐥𝐢𝐦 𝒛𝟑 = 𝟐𝟒
𝒛→𝟐

4. Residue at an essential singularity.

Let z=z0 be an essential singular point f(z).In this case the only way of computing the residue at
z=z0 is expanding the Laurent series of f(z) about the point z=z0.
1
Example 1.f(z)=z 𝑒 𝑧 has an essential singularity at z=0 since z=0 is not analytic at z=0 and
its Laurent series
1
1 1 1 1 1
z 𝑒 𝑧 = 𝑧(1 + + + + ⋯ =𝑧 + 1 + + + ⋯ has infinite terms in the principal
𝑧 2!𝑧 2 3!𝑧 3 2!𝑧 3!𝑧 2
1
part.From the Laurent series Res(f(z);0)= b1=2 .
1
1
  ze z dz  2i( 2 )  i .
Quick exercise.
Find all the singularities in the finite plane and the corresponding residues.
𝑠𝑖𝑛4𝑧 𝑧4 1
1. 2. 3. 𝑧 2 cos 𝑧
𝑧8 𝑧 2 −𝑖𝑧+2

Group Activity
Find all the singularities in the finite plane and the corresponding residues of the functions 1-8.
1 4. cot 𝜋𝑧 1⁄
1. sin
𝑧
8. 𝑒 1−𝑧
cos 𝑧
sinh 𝑧 5.
2. 𝑧4
𝑧4
𝑧
1 6. ( )3
3. 2𝑧+1
1−𝑒 𝑧 exp 𝑧
7.
𝑧 2 +𝜋2
𝜋cot(𝜋𝑧) 𝜋2
9. Show that the residue of f(z)= at z=0 is -
𝑧2 3
We have seen how to find the integral of f(z) which is analytic in the contour C at single
singularity .What if the contour C contains a finite number of singularities? Here is the answer.
Theorem 5.30 The residue Theorem
Let f(z) be analytic inside a simple closed path C and on C, except for finitely many singular
points 𝑧1 , 𝑧2 , 𝑧3 ,…, 𝑧𝑘 inside C. Then the integral of f(z) taken counterclockwise around C
equals 2πi times the sum of the residues of f(z) at 𝑧1 , 𝑧2 , 𝑧3 ,…, 𝑧𝑘 :
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 272
 Applied Mathematics III

 f ( z )dz  2i  Re s( f ( z ); z j )
C
j 1

Proof. Since 𝑧1 , 𝑧2 , 𝑧3 ,…, 𝑧𝑘 are isolated singularities ,then there is a set of non-overlapping
circles Ck with center 𝑧𝑘 ,k=1,2,3,…,n(for example,Fig.5.17 where k=3) such that each circle Ck
enclosed only one singular point zk and lies inside C. By Cauchy integral theorem for multiply
connected domain we write,

 f ( z )dz =  f ( z )dz +  f ( z )dz +….+  f ( z )dz (*)

C c1 c2 ck

Where all the curves C1,C2,C3,…,Ck traversed in the anticlockwise direction.

We know ck
f ( z )dz =2𝜋𝑖Res(f(z); 𝑧𝑘 ).Substituting this into (*) we have

 f ( z )dz  2i  Re s( f ( z ); z j )
C
j 1

Fig 5.17 Laurent theorem

3z  2
Example 1. Evaluate 
C z ( z  3)
dz where C is a circle |𝑧 − 1| =3 in counterclockwise

Solution. The integrand has two simple poles at z=0 and z=3, both of which are interior to C.
Also
3𝑧+2 2 3𝑧+2 8
Res(f(z),z=0)=lim𝑧→0 𝑧 𝑧(𝑧−3) = − 3 𝑎𝑛𝑑 Res(f(z),z=3)=lim𝑧→3 ( 𝑧 − 3)(𝑧(𝑧−3) = 3 ,then by the

Residue theorem
3z  2 2 8

C z ( z  3)
dz =2𝜋𝑖(− 3 + 3) = 4𝜋𝑖

e z 1
Example 2. Evaluate  dz where
C z ( z  1)( z  i ) 2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 273

 Applied Mathematics III

1 C2: |𝑧| = 2
i. C1: |𝑧| = 2 ii.

Solutions. z=0,1,2 are isolated singularities of f(z).Then z=0 is a removable singularity, z=1 is a
simple pole and z=0 is a pole of order 2.
i. z=0 is the only singularity which lies in C1 and Res(f(z);z=0)=0 since z=0 is a
removable singular point and thus
e z 1
C z( z  1)( z  i) 2 dz  2i(0)  0
ii. All the three singular points lie in C2 and
𝑒 𝑧 −1 1−𝑒
Res(f(z);z=1)= lim𝑧→1 ( 𝑧 − 1)(𝑧(𝑧−1)(𝑧−𝑖)2 = and
2𝑖

𝑑 𝑒 𝑧 −1 1
Res(f(z);z=i)= lim𝑧→𝑖 𝑑𝑧 ((𝑧 − 𝑖)2 (𝑧(𝑧−1)(𝑧−𝑖)2 )) = 2 (3𝑒 𝑖 − 𝑖) (verify it!)

Therefore, by the Residue theorem,

e z 1 1−𝑒 1
C z( z  1)( z  i) 2 dz = 2𝜋𝑖(0 + 2𝑖 + 2 (3𝑒 𝑖 − 𝑖)
tan z 1
Example 3. Evaluate 
C
(
z2  4
 ze z
)dz where C : |𝑧| = 2 clockwise.

𝜋 3𝜋
Solution. tan z is not analytic at z=± 2 , ± , …. ,all of which are outside the contour C. The
2
𝑡𝑎𝑛𝑧
integrand has simple pole 2 and -2 since z2-4=0.z=0 is also an essential singularity of the
𝑧 2 −4
1⁄
integrand 𝑧𝑒 𝑧 which lies inside C. Then
𝑡𝑎𝑛𝑧 𝑡𝑎𝑛𝑧 𝑡𝑎𝑛2
Res(𝑧 2 −4; z= 2)=2𝜋𝑖 lim𝑧→2 (𝑧 − 2) (𝑧−2)(𝑧+2) = 4
and
𝑡𝑎𝑛𝑧 𝑡𝑎𝑛𝑧 𝑡𝑎𝑛−2 𝑡𝑎𝑛2
Res(𝑧 2 −4; z= -2)=2𝜋𝑖 lim𝑧→−2 (𝑧 + 2) (𝑧−2)(𝑧+2) = = (𝑤ℎ𝑦) .
−4 4
1⁄
Since z=0 is an essential singularity ,the only method to find the residue of 𝑧𝑒 𝑧 is finding its
Laurent series.So
1⁄ 1 1 1
𝑧𝑒 𝑧 =z(1 + 𝑧 + 2!𝑧 2 + 3!𝑧 3+….)
1 1 1 1
= z +1+ 2!𝑧 + 3!𝑧 2 + ⋯ from which its residue is the coefficient of , that is 2.
𝑧

tan z 1 𝑡𝑎𝑛2 𝑡𝑎𝑛2 1

∴  (z
C 2
4
 ze z )dz = 2𝜋𝑖 (
4
+ 4
+ 2
) = 𝜋𝑖(tan 2 + 1) by the residue theorem.

Quick exercise.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 274

 Applied Mathematics III

Evaluate the following integrals(Counterclockwise).

z  23
1. 
C z  4z  5
2
dz , C: |𝑧 − 2 − 𝑖| = 3.2

zez 
2. 
C
(
z  16
4
 ze z )dz , where C is the ellipse 9𝑥 2 + 𝑦 2 = 9

z 1
3. 
C ( z  1)( z  3) 4
dz, C: |𝑧 − 2| =2

Group Exercise
1. Find Res(f,0) for
a) 𝑓(𝑧) = d) 𝑓(𝑧) = 𝑐𝑠𝑐 2 𝑧
f) 𝑓(𝑧) = 𝑧
1
𝑧 −3 cosh 4𝑧 𝑧 4 sin(1 + 𝑧)
b) 𝑓(𝑧) = csc 𝑧 exp(4𝑧)−1
e) 𝑓(𝑧) =
1 𝑠𝑖𝑛2 𝑧
c) 𝑓(𝑧) = exp(1 + 𝑧

2. Evaluate all the singularities and the corresponding residues.

𝑒𝑧 2𝜋 1
a. d. g.
((𝑧 2 +36)2 (𝑧−𝜋𝑖)4 𝑧 2 (𝑧−9)3 (𝑧+𝑧 2 )2

1 e. cot(𝜋𝑧)
b. 1−𝑒 𝑧
1⁄ f. tanz
c. 𝑒 1−𝑧

3. Let p(z) be a polynomial of degree at most 2.Show that if a,b and c are distinct complex
numbers, then
𝑝(𝑧) 𝐴 𝐵 𝐶
f(z)=(𝑧−𝑎)(𝑧−𝑏)(𝑧−𝑐) = 𝑧−𝑎 + 𝑧−𝑏 + 𝑧−𝑐
𝑝(𝑎)
where A=Res(f, a)=(𝑎−𝑏)(𝑎−𝑐)
𝑝(𝑏)
B=Res(f, b)=(𝑏−𝑎)(𝑏−𝑐) and
𝑝(𝑐)
C=Res(f, c)=(𝑐−𝑎)(𝑐−𝑏)

4. Evaluate the following the integrals counterclockwise.

sinh z 3z 3  2
a. C 2 z  i dz, C:|𝑧 − 2𝑖| = 2 b. C ( z  1)( z 2  9)dz, where i)

C: |𝑧 − 2| = 2 and ii) 𝐶: |𝑧| = 4

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 275

 Applied Mathematics III

1
1
c. 
z 3e z
C 1 z3
dz, C:|𝑧| = 3 j. C z  z 3  2z 2
4
dz, 𝐶: |𝑧| = 3

 (z  2) 1dz, 𝐶: |𝑧 − 1| = 2
4
3z 2  2 z  5 k.
C (3z  2) 2 (2 z  1)dz , 𝐶 𝑖𝑠 𝑡ℎ𝑒 𝑢𝑛𝑖𝑡 𝑐𝑖𝑟𝑐𝑙𝑒
C
d.
1
z 5
l. C z 4
4
dz, C: |𝑧 + 1 − 𝑖| = 1
e. 
C z  3z 3
4
dz, C: |𝑧 − 𝑖| = 4
sin z
cos z
m. C 4z 2   2
dz, C: |𝑧| = 2
f. 
C z5
dz, C:|𝑧 − 1| = 1.5
1
n.  3z  10 z 2  3
4
dz, C: |𝑧 −
 tan zdz ,
C
g. C:|𝑧| = 2
C
𝐼
𝑖√3| = 1 , C:|𝑧 − |=1
dz √3
h.  , C:|𝑧| = 2
C sinh z

z cosh z
i. 
C z  13z 2  36
4
dz, C: |𝑧| = 𝜋

5. Show that
𝑧−𝑠𝑖𝑛ℎ𝑧 𝑖
a) Res( ; 𝜋𝑖) =
𝑧 2 𝑠𝑖𝑛ℎ𝑧 𝜋
exp(𝑧𝑡) exp(𝑧𝑡)
b) Res( ; 𝜋𝑖) + Res( ; −𝜋𝑖) = −2𝑐𝑜𝑠(𝜋𝑡)
𝑠𝑖𝑛ℎ𝑧 𝑠𝑖𝑛ℎ𝑧

Evaluation of real integrals using residue.

One advantage of complex analysis over real analysis or calculus is evaluating complicated
integrals of real-valued functions using residue integration. In this section we shall see important
applications of the theory of residues.
𝟐𝝅
1. Integrals of the type ∫𝟎 𝑭(𝒄𝒐𝒔𝜽, 𝒔𝒊𝒏𝜽)𝒅𝜽 where where
F(cos𝜃, 𝑠𝑖𝑛𝜗) is a real rational of cos𝜃 𝑎𝑛𝑑 𝑠𝑖𝑛𝜗 and is finite (does not become infinite) on the
interval of integration. Setting 𝑧 = 𝑒 𝑖𝜗 we obtain
1 1 1
𝑐𝑜𝑠𝜃 = (𝑒 𝑖𝜃 + 𝑒 −𝑖𝜃 )= (𝑧 + )
2 2 𝑧
1 1 1
𝑠𝑖𝑛𝜃 = (𝑒 𝑖𝜃 − 𝑒 −𝑖𝜃 )= (𝑧 − )
2 2𝑖 𝑧

and 𝑑𝑧⁄𝑑𝜃 = 𝑖𝑒 𝑖𝜃  𝑑𝜃 = 𝑖𝑧 . Since F is rational function in z (why?) , say, f(z),then

𝑑𝑧

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 276

 Applied Mathematics III

2𝜋
dz
∫ 𝐹(𝑐𝑜𝑠𝜃, 𝑠𝑖𝑛𝜃)𝑑𝜃 = 
C
f ( z)
iz
0
and, as 𝜃 ranges from 0 to in 2𝜋 , the variable z=𝒆𝒊𝜽 ranges counterclockwise once around the
unit circle |𝑧| = 1.
Example 1 . Evaluate the following integrals.
2𝜋 𝑑𝜃
1. ∫0 5+4𝑠𝑖𝑛𝜃
𝑑𝑧 1 1
Solution. Let 𝑧 = 𝑒 𝑖𝜃 .Then 𝑑𝜃 = and using 𝑠𝑖𝑛𝜃 = (z − ) the integral becomes
𝑖𝑧 2i z

2𝜋 𝑑𝜃 dz
∫0 = where C: |𝑧| = 1
5+4𝑠𝑖𝑛𝜃 C z  z 1
iz[5  4( )]
2i
dz dz
=  .
C 2 z  5iz  2
2 C i
2( z  2i)( z  )
2
𝑖
Since z=-2i and z=− 2 are simple poles of the integrand of which -2i is outside the contour
𝑖
C,then we compute the residue at − 2 as
𝑖 𝑖 1 1
Res(f(z);z=− 2)=lim𝑧→− 𝑖 (𝑧 + 2) 𝑖 = 3𝑖.Hence by the Residue theorem
2 2(𝑧+2𝑖)(𝑧+ )
2

2𝜋 𝑑𝜃 1 2𝜋
∫0 =2𝜋𝑖(3𝑖)= 3
5+4𝑠𝑖𝑛𝜃

Example 2.
2𝜋 𝑑𝜃
2. ∫0 2+𝑐𝑜𝑠𝜃
1 1
Solution. Letting 𝑧 = 𝑒 𝑖𝜃 and 𝑐𝑜𝑠𝜃 = 2 (z + z),the given integral becomes

dz
C z  z 1
where C: |𝑧| = 1.
iz (2  )
2

dz dz  2idz
Therefore , C zz 1
=
C i ( 4 z  z  1)
2

C ( z  2) 2  3
iz (2  )
2
 2idz
=
( z  2  3 )( z  2  3 )

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 277

 Applied Mathematics III

 − 2 + √3 & − 2 − √3 are simple poles for the integrand where −2 + √3 lies

inside C. Therefore
2𝑖 𝑖
Res(f(z); z= −2 + √3) = lim𝑧→−2+√3 ( − (𝑧−(−2−√3)) = − and hence by the residue theorem
√3

 2idz 𝑖 2𝜋
 (z  2  3 )( z  2  3 )
=2𝜋𝑖 (− ) =
√3 √3

Quick exercise Show that

2𝜋 𝑑𝜃 𝜋 2𝜋 cos 2𝜃𝑑𝜃 𝜋 2𝜋 𝑑𝜃 𝜋
1. ∫0 = 2. ∫0 =6 3. ∫0 =3
5+3𝑐𝑜𝑠𝜃 2 5−4𝑐𝑜𝑠𝜃 1+15𝑠𝑖𝑛2 𝜃

Improper integrals of rational function.

∞ 𝑔(𝑥)
Consider real integrals of the type ∫−∞ 𝑓(𝑥)𝑑 𝑤ℎ𝑒𝑟𝑒 𝑓(𝑥) = ℎ(𝑥) and g(x) and h(x) are

polynomials in x and the degree of h(x) exceeds that of g(x) by at least two.To evaluate this type
𝑔(𝑧)
of integrals we let 𝑓(𝑧) = ℎ(𝑧).The poles of f(z) are determined by the zeros of the equation

h(z)=0.
Case I : No poles of f(z) lies on the real axis:

In this case we choose the curve C consisting of the interval [-r, r] on the real axis and the semi
circle 𝐶1 : |𝑧| = 𝑟 lying the upper half of the plane.Here we choose r large enough so that all the
poles lying in the upper half of the plane are in the interior of C.Then we have
r
C
f ( z )dz   f ( x)dx   f ( z )dz where C1 is the semicircle. Since the degree of h(x)-the
r C1

degree of g(x) is greater than equal to two it follows that 

C1
f ( z )dz  0 as r  0 and hence

 
C
f ( z )dz   f ( x)dx . Therefore
 

f ( x)dx can be evaluated by evaluating  C
f ( z )dz by the

residue theorem.

f ( x)dx =2𝜋𝑖 ∑ 𝑅𝑒𝑠(𝑓(𝑧), the sum is over all residues of the corresponding

∴ 


poles of f(z) in the upper half plane.

Case ii: f(z) has poles lying on the real axis.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 278

 Applied Mathematics III

Suppose that z0 is a pole lying on the real axis. In this case we indent the real axis by a semi
circle C2 of radius 𝜀 with center z0 lying on the upper half plane where 𝜀 is chosen to be
sufficiently small.

It can be proven that C1

f ( z )dz  i Re s( f ( z ); z  z0 ) by taking limits as → ∞ & 𝜀 → 0 , we get

the value of
∞

∫ 𝑓(𝑥)𝑑𝑥 = 𝜋𝑖𝑅𝑒𝑠(𝑓(𝑧); 𝑧 = 𝑧0 )
−∞
∞ 1
Example 1. Evaluate ∫−∞ 𝑥 4 +1
𝜃+2𝑘𝜋
1
. The f(z) has four simple pole at z=√−1 = 𝑒 𝑖(
4
Solution. Let f(z)= 𝑛
𝑧 4 +1
𝜋𝑖⁄ 3𝜋𝑖⁄ −𝜋𝑖⁄ −3𝜋𝑖⁄
,n=4,𝜃 = 𝜋,k=0,1,2,3,namely z1=𝑒 4 , 𝑧2 =𝑒 4 , 𝑧3 =𝑒 4 𝑎𝑛𝑑 𝑧4 = 𝑒 4

But only 𝑧1 𝑎𝑛𝑑 𝑧2 lie in the upper half of the plane.Let h(z)=1 and k(z)=𝑧 4 + 1 so that f(z)
ℎ(𝑧)
= 𝑘(𝑧).

𝜋𝑖⁄ −3𝜋
ℎ(𝑧1 ) 1 1 1 1
Res(f(z);z=𝑒 4 )= = 4𝑧 3 =4 𝑒 4 = − 4√2 − 4√2 𝑖
𝑘 ′ (𝑧1 ) z z1

3𝜋𝑖⁄ −9𝜋
ℎ(𝑧1 ) 1 1 1 1
Res(f(z);z=𝑒 4 )= = 4𝑧 3 =4 𝑒 4 = 4√2 − 4√2 𝑖
𝑘 ′ (𝑧1 ) z z 2

∞ 1 1 1 1 1 𝜋
Therefore ∫−∞ 𝑥 4 +1 = 2𝜋𝑖 (− 4√2 − 4√2 𝑖 + 4√2 − 4√2 𝑖) =
√2
∞ 𝑑𝑥
Example 2. Evaluate ∫−∞ (𝑥 2 −5𝑥+6)(𝑥 2 +4)
𝑑𝑧
Solution. Let f(z)= (𝑧 2 −5𝑧+6)(𝑧 2 +4). f(z) has four simple poles at z= ±2𝑖, 𝑧 = 2 𝑎𝑛𝑑 𝑧 = 3 of

which 𝑧 = −2𝑖 is in the lower half plane which is of no interest here and z=2 and z=3 are on the
real axis.
1 1
Res(f(z);z=2)=lim𝑧→2 (z − 2) (z−2)(z−3)((z2 +4) = − 8
1 1
Res(f(z);z=3)=lim𝑧→3 (z − 3) (z−2)(z−3)((z2 +4) = 13
1 1 40−8i 5−i
Res(f(z);z=2i)=limz→2i (z − 2i) (z−2)(z−3)(z+2i)(z−2i) = 40+8i = = 208
1664
∞ dx 5−i 1 1 π
∴ ∫−∞ (x2 −5x+6)(x2 +4) = 2πi(208) + πi (− 8 + 13) = 104

Quick exercises. Show that

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 279

 Applied Mathematics III

∞ 𝑥 2 𝑑𝑥 𝜋 ∞ 𝑥 4 𝑑𝑥 √3𝜋
1. ∫−∞ (𝑥 2 +1)2 = 2. ∫0 =
2 𝑥 6 −1 6

Exercises Evaluate the following integrals

∞ 𝑑𝑥 ∞ 𝑥+5
a) ∫−∞ 2 g) ∫−∞ 3 𝑑𝑥
𝑥 + 6𝑥+ 13 𝑥 −𝑥
∞ 𝑥 2 𝑑𝑥 ∞ 𝑥2
b) ∫−∞ 4 h) ∫0 𝑑𝑥
𝑥 + 6𝑥 2 + 13 𝑥 6 +1
∞ 𝑥𝑑𝑥 ∞ 𝑥 2 𝑑𝑥
c) ∫−∞ ( 2 i) ∫−∞ 2
𝑥 +1)(𝑥 2 +4) (𝑥 +16)2
∞ 𝑑𝑥 ∞ 𝑥+3𝑑𝑥
d) ∫0 j) ∫−∞ 2 2
𝑥 4 +1 (𝑥 +9)
∞ 𝑥2 ∞ 𝑑𝑥
e) ∫0( 𝑑𝑥 k) ∫−∞ 2 3
𝑥 2 +4)(𝑥 2 +9) (𝑥 +4)
∞ 𝑥 4 𝑑𝑥 ∞ 𝑥 2 𝑑𝑥
f) ∫0 l) ∫−∞
𝑥 6 +1 (𝑥 2 +4)3

5.6 SUMMARY 0F UNIT 5

For arithmetic operations with complex numbers

𝑧 = 𝑥 + 𝑖𝑦 = 𝑟𝑒 𝑖𝜃 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃) (1)
𝑦
𝑟 = |𝑧| = √𝑥 2 + 𝑦 2 𝜃 = arctan(𝑥 ), and for their representation in the complex plane.

A complex function is analytic in a domain D if it has a derivative

𝑓(𝑧+∆𝑧)−𝑓(𝑧)
𝑓 ′ (𝑧) = lim∆𝑧→0 (2)
∆𝑧

everywhere in D. Also, f (z) is analytic at a point z=𝑧0 f it has a derivative in a

neighborhood of 𝑧0 (not necessarily at 𝑧0 itself).
If f(z) is analytic in D, then u(x ,y)and v(x, y) satisfy the Cauchy–Riemann equations
𝜕𝑢 𝜕𝑣 𝜕𝑢 𝜕𝑣
= , =− . (3)
𝜕𝑥 𝜕𝑦 𝜕𝑦 𝜕𝑥

everywhere in D. Then u and v also satisfy Laplace’s equation

𝑢𝑥𝑥 + 𝑢𝑦𝑦 = 0, 𝑣𝑥𝑥 + 𝑣𝑦𝑦 = 0. (4)

everywhere in D. If u(x, y) and v(x, y) are continuous and have continuous partial derivatives in
D that satisfy (3) in D, then f(z)=𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) is analytic in D.
The complex exponential function

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 280

 Applied Mathematics III

𝑒 𝑧 = exp 𝑧 = 𝑒 𝑥 (cos 𝑦 + 𝑖 sin 𝑦) (5)

reduces to 𝑒 𝑥 𝑖𝑓 𝑧 = 𝑥(𝑦 = 0). if . It is periodic with 2𝜋𝑖 and has the derivative 𝑒 𝑧 .

The trigonometric functions are

1
cos 𝑧 = (𝑒 𝑖𝑧 + 𝑒 −𝑖𝑧 ) = cos 𝑥𝑐𝑜𝑠ℎ 𝑦 − 𝑖 sin 𝑥 sinh 𝑦 (6)
2
1
sin 𝑧 = (𝑒 𝑖𝑧 + 𝑒 −𝑖𝑧 ) = sin 𝑥𝑐𝑜𝑠ℎ 𝑦 − 𝑖 cos 𝑥 sinh 𝑦
2
and furthermore,
sin 𝑧 1
tan 𝑧 = , cot 𝑧 = , 𝑒𝑡𝑐.
cos 𝑧 tan 𝑧
The hyperbolic functions are
1 1
cosh 𝑧 = (𝑒 𝑧 + 𝑒 −𝑧 ) = cos 𝑖𝑧, sinh 𝑧 = (𝑒 𝑧 − 𝑒 −𝑧 ) = −𝑖 sin 𝑖𝑧 (7)
2 2
etc. The functions (5)–(7) are entire, that is, analytic everywhere in the complex plane.
The natural logarithm is
ln 𝑧 = ln|𝑧| + 𝑖 arg 𝑧 = ln|𝑧| + 𝑖 Arg 𝑧 ± 2𝑛𝜋𝑖
where 𝑧 ≠ 0 and n=0,1,2... . Arg z is the principal value of arg z, that is,
−𝜋 < 𝐴𝑟𝑔 𝑧 ≤ 𝜋. We see that ln z is infinitely many-valued. Taking n=0 gives the principal
value Ln z of ln z; thus 𝐿𝑛 𝑧 = 𝑙𝑛|𝑧| + 𝑖 𝐴𝑟𝑔 𝑧.
General powers are defined by
𝑧 𝑐 = 𝑒 𝑐 ln 𝑧 (𝑐 𝑐𝑜𝑚𝑝𝑙𝑒𝑥 , 𝑧 ≠ 0).
The complex line integral of a function taken over a path C is denoted by

 f ( z)dz or, if C is closed, also by  f(z)dz.

C C
(8)

If f(z) is analytic in a simply connected domain D, then we can evaluate (8) as in calculus by
indefinite integration and substitution of limits, that is,

(9)
 f ( z)dz  F ( z )  F ( z ) where F ( z )  f ( z ) 
'
1 0

for every path C in D from a point 𝑧0 to a point 𝑧1 . These assumptions imply independence of
C

path, that is, (9) depends only on 𝑧0 and 𝑧1 (and on f(z) of course) but not on the choice of C.
A general method of integration, not restricted to analytic functions, uses the equation z=z(t) of
C, where 𝑎 ≤ 𝑡 ≤ 𝑏,
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 281
 Applied Mathematics III

10
b

C
f ( z )dz   f ( z (t )) z ' (t )dt
a

Cauchy’s integral theorem is the most important theorem in this chapter. It states that if f(z) is
analytic in a simply connected domain D, then for every closed path C in D

 f ( z)dz  0 .
C

Under the same assumptions and for any 𝑧0 in D and closed path C in D containing 𝑧0 in its
interior we also have Cauchy’s integral formula
1 f ( z)
f ( z0 )  
2i C z  z 0
dz. (11)

Furthermore, under these assumptions f(z) has derivatives of all orders in D that are themselves
analytic functions in D and
𝑛! f ( z)
𝑓 𝑛 (𝑧0 ) = 2𝜋𝑖  (z  z 0)
n 1
dz , n  1, 2 ,.... (12)

This implies Morera’s theorem (the converse of Cauchy’s integral theorem) and Cauchy’s
inequality , which in turn implies Liouville’s theorem that an entire function that is bounded in
the whole complex plane must be constant.
A Laurent series is a series of the form
∞ ∞
𝑏𝑛
𝑓(𝑧) = ∑ 𝑎𝑛 (𝑧 − 𝑧0 )𝑛 + ∑ (13)
(𝑧 − 𝑧0 )𝑛
𝑛=0 𝑛=1
or, more briefly written

1 f (z* )
𝑓(𝑧) = ∑∞ 𝑛
𝑛=−∞ 𝑎𝑛 (𝑧 − 𝑧0 ) , 𝑎𝑛 = 2𝜋𝑖  ( z *  z ) n 1
dz *
C 0

where n=0,1,2, ... .This series converges in an open annulus (ring) A with center 𝑧0. In A the
function f(z) is analytic. At points not in A it may have singularities. The first series in (12) is a
power series. In a given annulus, a Laurent series of f(z) is unique, but f(z) may have different
Laurent series in different annuli with the same center.
Of particular importance is the Laurent series (12) that converges in a neighborhood of z0 except
at z0 itself, say, for 0 < |𝑧 − 𝑧0 | < 𝑅 𝑓𝑜𝑟 𝑠𝑢𝑖𝑡𝑎𝑏𝑙𝑒 𝑅 > 0. The series (or finite sum) of the

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 282

 Applied Mathematics III

negative powers in this Laurent series is called the principal part of f(z) at z0. The coefficient
1
𝑏1 of 𝑧−𝑧 in this series is called the residue of f(z) at z0 and is given by
0

1
2i C 
𝒃𝟏 = 𝑹𝒆𝒔(𝒇(𝒛); 𝒛 = 𝒛𝟎 ) = f ( z * ) dz * (𝟏𝟒). Thus f ( z * )dz *  2i Re s( f ( z ), z  z0 ).
C

𝒃𝟏 can be used for integration as shown in (14) because it can be found from
1 𝒅𝒎−𝟏
𝑹𝒆𝒔(𝒇(𝒛); 𝒛 = 𝒛𝟎 ) = 𝐥𝐢𝐦 ( 𝒎−𝟏 [(𝒛 − 𝒛𝟎 )𝒎 𝒇(𝒛)]), (𝟏𝟓)
(m  1)! 𝒛→𝒛𝟎 𝒅𝒛
provided f(z) has z0 at a pole of order m; by definition this means that principal part has
1
as its highest negative power. Thus for a simple pole (m-1)
(𝑧−𝑧0 )𝑚

𝑝(𝑧) 𝑝(𝑧0 )
𝑅𝑒𝑠(𝑓(𝑧); 𝑧 = 𝑧0 ) = lim (𝑧 − 𝑧0 )𝑓(𝑧); 𝑎𝑙𝑠𝑜, 𝑅𝑒𝑠 ( ; 𝑧 = 𝑧0 ) = ′ .
𝑧→𝑧0 𝑞(𝑧) 𝑞 (𝑧0 )
If the principal part is an infinite series, the singularity of f(z) at 𝑧0 is called an essential
singularity.
Residue integration may also be used to evaluate certain classes of complicated real integrals

CHAPTER 5 REVIEW QUESTIONS

1. Write the following complex numbers (1-4) in the form of +𝑖𝑦 .

1. (𝟑 − 𝟒𝐢)𝟐 3. √𝐢
𝟐−𝟑𝐢 𝟐𝛑𝐢
2. 4. 𝐞 𝟑
𝟏=𝟔𝐢
3. Find all values of the following functions(5-6)
4 3
5. √−24𝑖 6. √−1 , √8
Verify that the given functions (7-9 ) are Harmonic and find its Harmonic conjugates
7. 𝑢 = 𝑒 −𝑥 [(𝑥 2 − 𝑦 2 ) cos 𝑦 + 2𝑥𝑦 sin 𝑦)
8. u=𝑥 3 − 3𝑥𝑦 2 + 3𝑥 + 2
9. 𝑣 = 𝑦 sinh 2𝑥 cos 2𝑦 + 𝑥𝑐𝑜𝑠ℎ 2𝑥 sin 2𝑦
Find the value of the functions(10-11)
10. sin(2 − 𝑖) 11. cosh(𝜋 + 𝜋𝑖)
Solve the following equations(12-15)
12. 𝑒 𝑧 = 3𝑖 13. ln 𝑧 =
𝜋𝑖
2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 283

 Applied Mathematics III

14. arctan 2𝑖 = 𝑧 15. cos 𝑧 = cosh 2

Evaluate the following integrals(16-22 ) using Theorem 5.9 or Theorem 5.10
i
16. 
C
z sinh( z 2 )dz from 0 to
2
17.  ( z  z)dz clokwise around the unit circle
C

18.  Re z dz
C
from 0 to 3  27i along y  x 3

Lnz 1
19. 
C ( z  2i ) 2
dz counterclo ckwise around z  1 
2

2 1
20.  ( ( z  2i)  z  4i )dz
C
clockwise around z  1  2.5

 (z  z 2 )dz from z  0 horizontally to z  2, then vertically upward to 2  2i

2
21.
C

 (z  z 2 )dz from 0 to 2  2i, shortest path.

2
22.
C

Integrate the following complex functions(23-26) counterclockwise around C residue integration

method.
cos 6𝑧 cos 𝑧
23. , 𝐶: |𝑧| = 𝑒 25. , 𝑛 = 0,1,2, … 𝐶: |𝑧| = 1.
𝑧3 𝑧𝑛
2𝑧+7 3
24. 𝑧 3 −16𝑧 , 𝐶: |𝑧 − 1| = 2𝜋 26. cot 4𝑧 , 𝐶: |𝑧| = 4

Evaluate the following real integrals(27-39)

2𝜋 sin 𝜃 ∞ 𝑑𝑥 ∞ 𝑠𝑖𝑛2 𝑥
27. ∫0 𝑑𝜃 31. ∫−∞ 𝑥 6 +𝑎6 (𝑎 > 0) 35. ∫0 𝑑𝑥
3+cos 𝜃 𝑥2
2𝜋 1 ∞ 𝑥 2 𝑑𝑥 ∞ cos 𝑘𝑥
28. ∫0 𝑑𝜃 32. ∫0 36. ∫0 𝑑𝑥 , 𝑘 ≥ 0
1+𝑠𝑖𝑛2 𝜃 𝑥 6 +1 𝑥 4 +𝑎2
2𝜋 𝑑𝜃 ∞ 𝑑𝑥 ∞ 𝑥 sin 2𝑥𝑑𝑥
29. ∫0 1 33. ∫−∞ (𝑥 2 +1)3 37. ∫0
(1+ cos 𝜃)2 𝑥 2 +4
2
∞ cos 2𝜋𝑥𝑑𝑥 ∞ cos 3𝑥𝑑𝑥
∞ 𝑥 2 𝑑𝑥 34. ∫0 38. ∫−∞
30. ∫−∞ (𝑥 2 +1)2 (𝑥 2 +2𝑥+2) 𝑥 4 +4 (𝑥 2 +1)2

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 284

 Applied Mathematics III

References

1. Erwin Kreyszig: Advanced Engineering Mathematics, Eighth

Edition

2. Derrick & Grossman: A First Course in Differential Equations

With Applications, Third Edition

3. Peter V.O’Neil: Advanced Engineering Mathematics, Fourth

Edition.

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 285

 Applied Mathematics III

Contents
Module Introduction ....................................................................................................................... 1
UNIT I............................................................................................................................................. 3
1. ORDINARY DIFFERENTIAL EQUATIONS .......................................................................... 3
1.1 Introduction to Differential Equations .................................................................................. 4
1.1.1 Basic Concepts, terminology and Physical examples of Differential Equations ........... 4
1.1.2 Nature of Solutions of ODE: particular and general solutions ....................................... 8
1.2 First Order Ordinary Differential Equations ....................................................................... 12
1.2.1 Separable Differential Equations; Reduction to Separable Form ................................. 13
1.2.2 Homogeneous equations ............................................................................................... 19
1.2.3 Exact Differential Equations ........................................................................................ 27
1.2.3.1 Inexact equations; Integrating Factors ...................................................................... 32
1.2.4 Linear First Order Differential Equations; Reduction to Linear Form ......................... 37
1.2.5 Application Involving First Order Differential Equations ........................................... 44
1.3 Ordinary Linear Differential Equations of the Second order ......................................... 48
1.3. 1 Homogenous Equations .............................................................................................. 50
1.3.2 Non Homogeneous Linear Differential Equations ................................................. 71
1.4. Systems of Differential Equations.................................................................................. 88
1.5 Unit Summary ................................................................................................................ 90
UNIT II ......................................................................................................................................... 95
2. LAPLACE TRANSFORMS.................................................................................................. 95
2.1 The Laplace Transform; Inverse Transform and Linearity ................................................. 96
2.2 Laplace Transforms of Derivatives and Integrals ............................................................. 109
2.2.1 Transforms of Derivatives .......................................................................................... 109
2.2.2 Application To Differential Equations; Initial Value Problems ................................. 111
2.2.3 Laplace Transforms of Integrals . ............................................................................... 114
2.3 Unit Step Function and Second Shifting Theorem............................................................ 116
2.4 Differentiation and Integration of Transforms .................................................................. 123
2.4.1 Differentiation of Transforms ............................................................................... 123
2.4.2 Integration of Transforms ..................................................................................... 128
2.5. Convolution and Integral Equations ............................................................................. 130
2.5.1. Convolution ............................................................................................................. 130
2.5.2 Integral Equation .............................................................................................. 134
2.6 Unit Summary ................................................................................................................... 137
UNIT III ...................................................................................................................................... 140
3. FOURIER SERIES; FOURIER INTEGRALS AND .......................................................... 140
FOURIER TRANSFORMATIONS ........................................................................................... 140
3.1 Periodic and Orthogonal Funcions ............................................................................... 141
3.1.1. Periodic Functions and Trignometric Series ............................................................. 141
3.1.2. Orthogonal functions ................................................................................................. 144

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 286

 Applied Mathematics III

3.2. Fourier Series of Functions ............................................................................................. 144

3.2.1. Fourier Series of Functions with Periodicity 2 ................................................... 145
3.2.2 Convergence and Sum of Fourier Series .................................................................... 150
3.2.3 Fourier Series of Functions of Any Period p  2L ............................................ 154
3.2.4 The Fourier Series of Even and Odd Functions .................................................... 159
3.2.5 Half Range Expansions .............................................................................................. 166
3.2.6. Complex Exponential Fourier Series ......................................................................... 172
3.3. The Fourier Integrals ................................................................................................ 176
3.3.1 Fourier Cosine and Fourier Sine Integrals ........................................................... 177
3.4 Fourier Transformations ................................................................................................... 181
3.4.1 Fourier Cosine and Sine Transformations ............................................................ 181
3.4.2 Fourier Transformation............................................................................................... 187
3.5 Unit Summary .............................................................................................................. 191
UNIT FOUR .............................................................................................................................. 194
VECTOR DIFFERENTIAL CALCULUS ................................................................................. 194
4.1 Vector Fields ................................................................................................................... 195
4.2 The Divergence and Curl of a Vector Field ...................................................................... 197
4.3 Line and surface Integrals ................................................................................................. 202
4.4 Green’s and Stoke’s Theorem: .......................................................................................... 206
4.5 Gauss Divergence Theorem .............................................................................................. 211
4.6 Unit Summary ................................................................................................................... 214
UNIT 5 ........................................................................................................................................ 216
5. COMPLEX ANALYSIS ......................................................................................................... 216
5.1 Revision on complex numbers ......................................................................................... 216
5.1.1 Operations on complex numbers ................................................................................ 217
5.1.2 Polar coordinates of a complex number ..................................................................... 219
5.2 COMPLEX ANALYTIC FUNCTIONS ........................................................................... 223
5.2.1 Limit and Continuity .................................................................................................. 225
5.2.2 Derivative ................................................................................................................... 227
5.2.3 Cauchy–Riemann Equations....................................................................................... 231
5.2.4 Laplace’s Equation. Harmonic Functions .................................................................. 235
5.2.5 Basic elementary functions ........................................................................................... 238
5.3 Complex Integration .......................................................................................................... 245
5.3.1 Contour integral: Line integral in the complex plane ................................................. 246
5.3.2 Cauchy’s Integral Theorem ........................................................................................ 251
5.3.3 Cauchy’s Integral Formula ......................................................................................... 253
5.3.4 Derivative of analytic functions ..................................................................................... 255
5.4 Complex series,Taylor and Laurent series, residues and poles ......................................... 259
5.5 Residue integration method .................................................................................................. 269
5.6 SUMMARY 0F UNIT 5 ................................................................................................... 280
Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 287
 Applied Mathematics III

Kassahun Nigatu(Msc),Mesfin Teshome(Msc) and Yitagesu Daba(Msc) 288