Chapter 1

First Order Ordinary Differential Equations

Learning Objectives
By the end of this chapter, students should be able to
1. define an Ordinary Differential Equation (ODE);
2. determine the order of an ODE;
3. distinguish between general and particular solutions of an ODE;
4. distinguish between separable and linear ODEs;
5. solve first order ODEs using the variable separation method or the
integrating factor method;
6. solve engineering problems modelled by first order ODEs.

D ifferential Equations (DE) are indispensable in the study of engineering

and modern science. Many engineering problems are modelled into
one of two types of differential equations, namely, Ordinary
Differential Equations (ODEs) and Partial Differential Equations (PDEs).

We begin this chapter by introducing the ODEs and their orders in Section 1.1,
followed by discussing two methods in solving first order ODEs: the variable
separation method in Section 1.2, and the integrating factor method in Section
1.3. In Section 1.4, we look at some engineering problems modelled by first
order ODES. Finally, we close the chapter with exercises in Section 1.5, where
student gets to solve first order ODEs using the two methods discussed, as well
as a taste of more advanced techniques.

1
First Order Ordinary Differential Equations 2

1.1 Introduction

Definition 1.1.1 (Ordinary Differential Equations)

An Ordinary Differential Equation (ODE) is an equation involving the
derivatives of an output variable 𝑦𝑦 with respect to the input variable 𝑥𝑥.

The order of an ODE is the order of the highest derivative in the equation.

The following are examples of ODEs with their orders.

Example 1.1.2
𝑑𝑑𝑑𝑑
(a) = 3𝑥𝑥 2 First order
𝑑𝑑𝑑𝑑
(b) 𝑥𝑥𝑦𝑦 ′ + 𝑦𝑦 = 2𝑥𝑥 First order

(c) 𝑦𝑦 ′′ = 𝑦𝑦′ Second order

𝑑𝑑𝑑𝑑
(d) = 𝑡𝑡𝑥𝑥 2 First order
𝑑𝑑𝑑𝑑

Definition 1.1.3 (Solution of a DE)

A solution of a DE is an equation that satisfies the DE but does not involve any
derivatives.

There are two possible types of solutions in DEs:

1. The general solution of an 𝑛𝑛-th order DE is a solution that involves 𝑛𝑛
arbitrary constants. It includes all possible solutions of the DE.
2. A particular solution of a DE is a solution that does NOT involve any
arbitrary constants. This happens when initial conditions are known.
First Order Ordinary Differential Equations 3

A particular solution of an ODE is obtained when additional information are

given, such as initial value(s). ODEs where initial conditions are known are
called Initial Value Problems (IVP).

Example 1.1.4
ODE/IVP Solution Type

𝑑𝑑𝑑𝑑
(a) = 3𝑥𝑥 2 𝑦𝑦 = 𝑥𝑥 3 + 𝐶𝐶 General solution
𝑑𝑑𝑑𝑑
(b) 𝑦𝑦′′ = 𝑦𝑦 𝑦𝑦 = 𝐶𝐶1 𝑒𝑒 𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −𝑥𝑥 General solution

𝑥𝑥𝑥𝑥′ + 𝑦𝑦 = 2𝑥𝑥 1
(c) � 𝑦𝑦 = 𝑥𝑥 − Particular solution
𝑦𝑦(1) = 0 𝑥𝑥

𝑦𝑦′′ = 2𝑦𝑦′
(d) �𝑦𝑦(0) = 0 𝑦𝑦 = 1 − 𝑒𝑒 2𝑥𝑥 Particular solution
𝑦𝑦 ′ (0) = −1

From Example 1.1.4(a) and (b), we observe that the general solution of a first
order ODE contains one arbitrary constant, while that of a second order ODE
contains two arbitrary constant.

From Example 1.1.4(c) and (d), we observe that a first order IVP includes one
initial value, while a second order IVP includes two initial values.

The next few examples show how general solutions to ODEs can be found by
𝑑𝑑𝑑𝑑
noting that = 𝑓𝑓(𝑥𝑥) implies 𝑦𝑦 = � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 .
𝑑𝑑𝑑𝑑
First Order Ordinary Differential Equations 4

Example 1.1.5
Solve 𝑦𝑦′ = sin 2𝑥𝑥.

Solution
1
𝑦𝑦 ′ = sin 2𝑥𝑥 ⇒ 𝑦𝑦 = � sin 2𝑥𝑥 𝑑𝑑𝑑𝑑 ⇒ 𝑦𝑦 = − cos 2𝑥𝑥 + 𝐶𝐶.
2

Example 1.1.6
𝑑𝑑𝑑𝑑
Solve 𝑥𝑥 − 1 = 2𝑥𝑥 2 .
𝑑𝑑𝑑𝑑

Solution
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 1
𝑥𝑥 − 1 = 2𝑥𝑥 2 ⇒ 𝑥𝑥 = 2𝑥𝑥 2 + 1 ⇒ = 2𝑥𝑥 +
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑥𝑥
1
⇒ 𝑦𝑦 = � �2𝑥𝑥 + � 𝑑𝑑𝑑𝑑
𝑥𝑥
⇒ 𝑦𝑦 = 𝑥𝑥 2 + ln|𝑥𝑥| + 𝐶𝐶.

Example 1.1.7
𝑑𝑑𝑑𝑑
Solve the IVP: 𝑥𝑥 − 1 = 2𝑥𝑥 2 , where 𝑦𝑦(1) = 0.
𝑑𝑑𝑑𝑑

Solution
𝑑𝑑𝑑𝑑
From the previous example, the general solution of 𝑥𝑥 − 1 = 2𝑥𝑥 2 is
𝑑𝑑𝑑𝑑
𝑦𝑦 = 𝑥𝑥 2 + ln|𝑥𝑥| + 𝐶𝐶.
𝑦𝑦(1) = 0 ⇒ 0 = 12 + ln|1| + 𝐶𝐶 ⇒ 𝐶𝐶 = −1
The particular solution is 𝑦𝑦 = 𝑥𝑥 2 + ln|𝑥𝑥| − 1.
First Order Ordinary Differential Equations 5

1.2 Separable ODEs

Definition 1.2.1 (Separable ODE)

𝑑𝑑𝑑𝑑
A separable ODE is an ODE that can be written as = 𝑓𝑓(𝑥𝑥) 𝑔𝑔(𝑦𝑦) .
𝑑𝑑𝑑𝑑

Example 1.2.2
Identify which of the following ODEs are separable:
𝑑𝑑𝑑𝑑
(a) = 𝑥𝑥𝑥𝑥
𝑑𝑑𝑑𝑑
(b) 𝑦𝑦 ′ = 𝑥𝑥 2 𝑒𝑒 −𝑦𝑦
𝑑𝑑𝑑𝑑
(c) + 2𝑦𝑦 = 2𝑥𝑥
𝑑𝑑𝑑𝑑

Solution

(a) Separable.
(b) Separable.
(c) Non-separable.

Procedure in Solving Separable ODE

𝑑𝑑𝑑𝑑
Step 1. Group 𝑔𝑔(𝑦𝑦) with 𝑑𝑑𝑑𝑑 and 𝑓𝑓(𝑥𝑥) with 𝑑𝑑𝑑𝑑 to give = 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 .
𝑔𝑔(𝑦𝑦)
1
Step 2. Integrate both sides: � 𝑑𝑑𝑑𝑑 = � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑 to give the general
𝑔𝑔(𝑦𝑦)
solution.
First Order Ordinary Differential Equations 6

Example 1.2.3
𝑑𝑑𝑑𝑑
Solve = 𝑥𝑥𝑥𝑥 .
𝑑𝑑𝑑𝑑

Solution
𝑑𝑑𝑑𝑑 1
Step 1. = 𝑥𝑥𝑥𝑥 ⇒ 𝑑𝑑𝑑𝑑 = 𝑥𝑥 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑦𝑦
1 1
Step 2. � 𝑑𝑑𝑑𝑑 = � 𝑥𝑥 𝑑𝑑𝑑𝑑 ⇒ ln |𝑦𝑦| = 𝑥𝑥 2 + 𝐶𝐶.
𝑦𝑦 2

Example 1.2.4
Solve 𝑦𝑦 ′ = 𝑦𝑦 .

Solution
𝑑𝑑𝑑𝑑 1
Step 1. = 𝑦𝑦 ⇒ 𝑑𝑑𝑑𝑑 = 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑦𝑦
1
Step 2. � 𝑑𝑑𝑑𝑑 = � 𝑑𝑑𝑑𝑑 ⇒ ln |𝑦𝑦| = 𝑥𝑥 + 𝐶𝐶 or 𝑦𝑦 = 𝐶𝐶1 𝑒𝑒 𝑥𝑥 .
𝑦𝑦

Example 1.2.5
𝑑𝑑𝑑𝑑
Solve = 𝑒𝑒 2𝑥𝑥 sec 𝑦𝑦 .
𝑑𝑑𝑑𝑑

Solution
𝑑𝑑𝑑𝑑 1
Step 1. = 𝑒𝑒 2𝑥𝑥 sec 𝑦𝑦 ⇒ 𝑑𝑑𝑑𝑑 = 𝑒𝑒 2𝑥𝑥 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 sec 𝑦𝑦
1 2𝑥𝑥 1
Step 2. � cos 𝑦𝑦 𝑑𝑑𝑑𝑑 = � 𝑒𝑒 2𝑥𝑥 𝑑𝑑𝑑𝑑 ⇒ sin 𝑦𝑦 = 𝑒𝑒 + 𝐶𝐶. �note: sec 𝑦𝑦 = �
2 cos 𝑦𝑦
First Order Ordinary Differential Equations 7

Example 1.2.6
𝑑𝑑𝑑𝑑
Solve = 𝑥𝑥 2 𝑒𝑒 −𝑦𝑦 .
𝑑𝑑𝑑𝑑

Solution
𝑑𝑑𝑑𝑑
Step 1. = 𝑥𝑥 2 𝑒𝑒 −𝑦𝑦 ⇒ 𝑒𝑒𝑦𝑦 𝑑𝑑𝑑𝑑 = 𝑥𝑥2 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
1 3
Step 2. � 𝑒𝑒 𝑦𝑦 𝑑𝑑𝑑𝑑 = � 𝑥𝑥 2 𝑑𝑑𝑑𝑑 ⇒ 𝑒𝑒 𝑦𝑦 = 𝑥𝑥 + 𝐶𝐶.
3

Example 1.2.7
𝑥𝑥 2
Solve the IVP: 𝑦𝑦 ′ = , where 𝑦𝑦(0) = 1.
𝑒𝑒 𝑦𝑦

Solution
𝑥𝑥 2
From the previous example, the general solution of 𝑦𝑦 ′ = is
𝑒𝑒 𝑦𝑦
1
𝑒𝑒 𝑦𝑦 = 3 𝑥𝑥 3 + 𝐶𝐶.

𝑦𝑦(0) = 1 ⇒ 𝑒𝑒 1 = 0 + 𝐶𝐶 ⇒ 𝐶𝐶 = 𝑒𝑒
1
The particular solution is 𝑒𝑒 𝑦𝑦 = 3 𝑥𝑥 3 + 𝑒𝑒.

1.3 Linear ODE

𝑑𝑑𝑑𝑑
A first order linear ODE is an ODE of the form 𝑃𝑃1 + 𝑃𝑃0 𝑦𝑦 = 𝐹𝐹 , where 𝑃𝑃0 ,
𝑑𝑑𝑑𝑑
𝑃𝑃1 , 𝑦𝑦, and 𝐹𝐹 are functions of 𝑥𝑥, with 𝑃𝑃1 ≠ 0. The solution can be found by using
the integrating factor. As a start, we rewrite the ODE in standard form.
First Order Ordinary Differential Equations 8

Definition 1.3.1 (Standard Form of First Order Linear ODE)

𝑑𝑑𝑑𝑑
The standard form of a first order linear ODE is + 𝑃𝑃(𝑥𝑥) 𝑦𝑦 = 𝑄𝑄(𝑥𝑥) .
𝑑𝑑𝑑𝑑

Remark. In Definition 1.3.1, 𝑃𝑃 and 𝑄𝑄 are functions of 𝑥𝑥. If either 𝑃𝑃 or 𝑄𝑄 is zero,

then the ODE is a separable ODE.

Procedure in Solving First Order Linear ODE

Step 1. Write the ODE in standard form and identify 𝑃𝑃(𝑥𝑥) and 𝑄𝑄(𝑥𝑥).
Step 2. Evaluate and simplify the integrating factor 𝐼𝐼(𝑥𝑥) = 𝑒𝑒 ∫ 𝑃𝑃(𝑥𝑥) 𝑑𝑑𝑑𝑑 .

Step 3. The general solution is 𝐼𝐼(𝑥𝑥) 𝑦𝑦 = � 𝐼𝐼(𝑥𝑥) 𝑄𝑄(𝑥𝑥) 𝑑𝑑𝑑𝑑 .

Example 1.3.2
𝑑𝑑𝑑𝑑
Solve + 2𝑦𝑦 = 𝑒𝑒 −𝑥𝑥 .
𝑑𝑑𝑑𝑑

Solution
Step 1. 𝑃𝑃(𝑥𝑥) = 2 and 𝑄𝑄(𝑥𝑥) = 𝑒𝑒 −𝑥𝑥
Step 2. 𝐼𝐼(𝑥𝑥) = 𝑒𝑒 ∫ 2𝑑𝑑𝑑𝑑 = 𝑒𝑒 2𝑥𝑥

Step 3. The general solution is 𝑒𝑒 2𝑥𝑥 𝑦𝑦 = � 𝑒𝑒 2𝑥𝑥 𝑒𝑒 −𝑥𝑥 𝑑𝑑𝑑𝑑 = � 𝑒𝑒 𝑥𝑥 𝑑𝑑𝑑𝑑

⇒ 𝑒𝑒 2𝑥𝑥 𝑦𝑦 = 𝑒𝑒 𝑥𝑥 + 𝐶𝐶
First Order Ordinary Differential Equations 9

We may use the laws of exponents and radicals to simplify the integrating
factor. In the following examples, we note that 𝑒𝑒 ln 𝐴𝐴 = 𝐴𝐴.
Example 1.3.3
𝑑𝑑𝑑𝑑 2𝑦𝑦
Solve + = 1 , where 𝑥𝑥 > 0.
𝑑𝑑𝑑𝑑 𝑥𝑥

Solution
2
Step 1. 𝑃𝑃(𝑥𝑥) = and 𝑄𝑄(𝑥𝑥) = 1
𝑥𝑥
2
Step 2. 𝐼𝐼(𝑥𝑥) = 𝑒𝑒 ∫𝑥𝑥𝑑𝑑𝑑𝑑 = 𝑒𝑒 2 ln 𝑥𝑥 = 𝑒𝑒 ln 𝑥𝑥 2 = 𝑥𝑥 2

Step 3. The general solution is 𝑥𝑥 2 𝑦𝑦 = � 𝑥𝑥 2 𝑑𝑑𝑑𝑑

1
⇒ 𝑥𝑥 2 𝑦𝑦 = 3 𝑥𝑥 3 + 𝐶𝐶

Example 1.3.4
𝑑𝑑𝑑𝑑 1 1
Solve + �1 + � 𝑦𝑦 = , where 𝑥𝑥 > 0.
𝑑𝑑𝑑𝑑 𝑥𝑥 𝑥𝑥

Solution
1 1
Step 1. 𝑃𝑃(𝑥𝑥) = 1 + and 𝑄𝑄(𝑥𝑥) =
𝑥𝑥 𝑥𝑥
1
Step 2. 𝐼𝐼(𝑥𝑥) = 𝑒𝑒 ∫�1+𝑥𝑥�𝑑𝑑𝑑𝑑 = 𝑒𝑒 𝑥𝑥+ln 𝑥𝑥 = 𝑒𝑒 𝑥𝑥 𝑒𝑒 ln 𝑥𝑥 = 𝑥𝑥𝑒𝑒 𝑥𝑥
1
Step 3. The general solution is 𝑥𝑥𝑒𝑒 𝑥𝑥 𝑦𝑦 = �(𝑥𝑥𝑒𝑒 𝑥𝑥 ) � � 𝑑𝑑𝑑𝑑 = � 𝑒𝑒 𝑥𝑥 𝑑𝑑𝑑𝑑
𝑥𝑥
⇒ 𝑥𝑥𝑒𝑒 𝑥𝑥 𝑦𝑦 = 𝑒𝑒 𝑥𝑥 + 𝐶𝐶
First Order Ordinary Differential Equations 10

1.4 Applications
In this section, we discuss engineering problems modelled by first order DEs.
The modelling processes are briefly described for examples in the life sciences,
mechanics, and electronics.

Applications in the Life Sciences

The exponential growth/decay model is often used in describing the

growth/decay of a substance. The model is based on the assumption that the
derivative 𝑄𝑄′ is directly proportional to the quantity 𝑄𝑄. In other words,
𝑑𝑑𝑑𝑑
= 𝑘𝑘𝑘𝑘,
𝑑𝑑𝑑𝑑
where 𝑘𝑘 is a non-zero constant. If 𝑘𝑘 > 0, then it is a growth model; otherwise
it is a decay model.

The growth/decay model above is a separable ODE. To solve for the particular
solution, we consider the initial quantity 𝑄𝑄(0).

Example 1.4.1
In an experiment of testing a new drug against a certain type of bacteria, a
scientist modelled the decay of the bacteria population 𝑃𝑃 by the differential
equation
𝑑𝑑𝑑𝑑
= −0.2457𝑃𝑃
𝑑𝑑𝑑𝑑
Suppose the initial population of the bacteria in a culture is approximately 2 (in
million units). Express the bacteria population 𝑃𝑃 in terms of time 𝑡𝑡 (in days).
First Order Ordinary Differential Equations 11

Solution
𝑑𝑑𝑑𝑑 1
= −0.2457𝑃𝑃 ⇒ 𝑑𝑑𝑑𝑑 = −0.2457 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑃𝑃
1
⇒ � 𝑑𝑑𝑑𝑑 = −0.2457 � 𝑑𝑑𝑑𝑑
𝑃𝑃
⇒ ln 𝑃𝑃 + 𝐶𝐶 = −0.2457 𝑡𝑡

𝑃𝑃(0) = 2 ⇒ ln 2 + 𝐶𝐶 = 0 ⇒ 𝐶𝐶 = − ln 2
Thus, ln 𝑃𝑃 − ln 2 = −0.2457 𝑡𝑡 or 𝑃𝑃 = 2 𝑒𝑒 −0.2457𝑡𝑡 .

Applications in Mechanics

In classical mechanics, we study the motion of an object subjected to various

forces. The Newton’s Law 𝐹𝐹 = 𝑚𝑚𝑚𝑚 serves as the principle in modelling
mechanical problems. Here, 𝐹𝐹 denotes the total forces involved, and 𝑚𝑚
denotes the mass of the object. Moreover, 𝑎𝑎 denotes acceleration of the
object, defined as the derivative of the velocity, 𝑣𝑣, with respect to time, 𝑡𝑡:
𝑑𝑑𝑑𝑑
𝑎𝑎 = .
𝑑𝑑𝑑𝑑
Alternatively, if the velocity 𝑣𝑣 is regarded as a function of 𝑥𝑥, the displacement
of the object, then the acceleration can be written as follows:
𝑑𝑑𝑑𝑑
𝑎𝑎 = 𝑣𝑣 .
𝑑𝑑𝑑𝑑

We shall discuss two examples of mechanics problems modelled by differential

𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
equations involving or , and solve for the particular solutions by
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
considering the initial velocity 𝑣𝑣(0).
First Order Ordinary Differential Equations 12

Example 1.4.2
A ball of unit mass moves horizontally in a straight line with initial velocity 1
m/s. The motion of the ball satisfies the differential equation
1 𝑑𝑑𝑑𝑑
− 𝑣𝑣 3 = .
2 𝑑𝑑𝑑𝑑
Express 𝑣𝑣 in terms of 𝑡𝑡.

Solution
1 𝑑𝑑𝑑𝑑 −2
− 𝑣𝑣 3 = ⇒ 𝑑𝑑𝑑𝑑 = 3 𝑑𝑑𝑑𝑑
2 𝑑𝑑𝑑𝑑 𝑣𝑣

⇒ � 𝑑𝑑𝑑𝑑 = −2 � 𝑣𝑣 −3 𝑑𝑑𝑑𝑑

⇒ 𝑡𝑡 = 𝑣𝑣 −2 + 𝐶𝐶

At 𝑡𝑡 = 0, the velocity is 𝑣𝑣 = 1 m/s. Hence, 0 = 1−2 + 𝐶𝐶 ⇒ 𝐶𝐶 = −1.

1
And thus, 𝑡𝑡 = 𝑣𝑣 −2 − 1 or 𝑣𝑣 = � .
𝑡𝑡 + 1

v (m/s)

t (second)
2 4 6 8
Figure 1.1. Graph of 𝑣𝑣 in terms of 𝑡𝑡 in Example 1.4.2
First Order Ordinary Differential Equations 13

Example 1.4.3
A particle of unit mass is projected vertically upwards with initial velocity 1
m/s, and subjected to earth’s gravity. The motion of the particle satisfies the
differential equation
𝑑𝑑𝑑𝑑 1
𝑣𝑣 = −𝑔𝑔 − 𝑔𝑔𝑣𝑣 2 ,
𝑑𝑑𝑑𝑑 3
where 𝑥𝑥 denotes the vertical (upward) displacement of the particle, and 𝑔𝑔 ≈
9.8 m/s2 is the constant of gravitational acceleration.

(a) Express 𝑥𝑥 in terms of 𝑣𝑣.

(b) Calculate the maximum displacement of the particle.

Solution
𝑑𝑑𝑑𝑑 1 2 𝑑𝑑𝑑𝑑 𝑣𝑣 2
(a) 𝑣𝑣 = −𝑔𝑔 − 𝑔𝑔𝑣𝑣 ⇒ 𝑣𝑣 = −𝑔𝑔 �1 + �
𝑑𝑑𝑑𝑑 3 𝑑𝑑𝑑𝑑 3
𝑣𝑣
⇒ 𝑑𝑑𝑑𝑑 = 𝑑𝑑𝑑𝑑
𝑣𝑣 2
−𝑔𝑔 �1 + �
3
1 3𝑣𝑣
⇒ � 𝑑𝑑𝑑𝑑 = � 𝑑𝑑𝑑𝑑
−𝑔𝑔 3 + 𝑣𝑣 2
3
⇒− ln(3 + 𝑣𝑣 2 ) + 𝐶𝐶 = 𝑥𝑥
2𝑔𝑔
3 3
At 𝑥𝑥 = 0, the velocity is 𝑣𝑣 = 1. So, − 2𝑔𝑔 ln(3 + 12 ) + 𝐶𝐶 = 0 ⇒ 𝐶𝐶 = 2𝑔𝑔 ln 4.
3 3 3 4
Thus, − 2𝑔𝑔 ln(3 + 𝑣𝑣 2 ) + 2 ln 4 = 𝑥𝑥 ⇒ 𝑥𝑥 = 2𝑔𝑔 ln �3+𝑣𝑣2�.

(b) The maximum displacement is obtained when 𝑣𝑣 = 0.

3 4
Thus, 𝑥𝑥max = 2𝑔𝑔 ln �3� ≈ 0.0440 m.
First Order Ordinary Differential Equations 14

Applications in Electronics

In circuit analysis, we study how interactions of electronics components affect

the behaviour of the current and voltage in an electrical circuit. The Kirchoff’s
Voltage Law (KVL) serves as the principle in modelling electronics problems.
KVL states that the sum of the voltage across each component in a series
circuit is equal to the voltage source, i.e.

𝑉𝑉𝑅𝑅 + 𝑉𝑉𝐿𝐿 + 𝑉𝑉𝐶𝐶 = 𝑉𝑉source .

where 𝑉𝑉𝑅𝑅 , 𝑉𝑉𝐿𝐿 , and 𝑉𝑉𝐶𝐶 denotes the voltage across resistor, inductor, and
capacitor, respectively, and they satisfy the following properties:

𝑑𝑑𝑑𝑑 𝑞𝑞
𝑉𝑉𝑅𝑅 = 𝑖𝑖𝑖𝑖, 𝑉𝑉𝐿𝐿 = 𝐿𝐿 𝑑𝑑𝑑𝑑, 𝑉𝑉𝐶𝐶 = 𝐶𝐶.

Here, 𝑖𝑖 denotes the current (in amperes), 𝑅𝑅 the resistance (in ohms), 𝐿𝐿 the
inductance (in henrys), 𝑞𝑞 the charge (in coulombs), and 𝐶𝐶 the capacitance (in
farads), and 𝑡𝑡 the time (in seconds).

In a closed series circuit, the same current passes through each electrical
component in the circuit. Moreover, the current is defined as the derivative of
the charge with respect to time:
𝑑𝑑𝑑𝑑
𝑖𝑖 = .
𝑑𝑑𝑑𝑑

We shall discuss three examples of electronics problems modelled by linear

𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
differential equations involving or , and solve for their particular
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
solutions by considering initial current 𝑖𝑖(0) or initial charge 𝑞𝑞(0).
First Order Ordinary Differential Equations 15

Example 1.4.4
A series circuit consists of an inductor of inductance 1 H, a resistor of
resistance 1 kΩ and a 9V battery. The circuit is closed at time 𝑡𝑡 = 0 and
satisfies the differential equation
𝑑𝑑𝑑𝑑
+ 1000𝑖𝑖 = 9.
𝑑𝑑𝑑𝑑
Express the current 𝑖𝑖 in terms of time 𝑡𝑡.

Solution
𝑑𝑑𝑑𝑑
+ 1000𝑖𝑖 = 9 is a standard 1st order linear ODE
𝑑𝑑𝑑𝑑
𝑃𝑃(𝑡𝑡) = 1000 and 𝑄𝑄(𝑡𝑡) = 9
𝐼𝐼(𝑡𝑡) = 𝑒𝑒 ∫ 1000 𝑑𝑑𝑑𝑑 = 𝑒𝑒 1000𝑡𝑡
9
𝑒𝑒 1000𝑡𝑡 𝑖𝑖 = � 𝑒𝑒 1000𝑡𝑡 (9)𝑑𝑑𝑑𝑑 = 𝑒𝑒 1000𝑡𝑡 + 𝐶𝐶
1000
⇒ 𝑖𝑖 = 0.009 + 𝐶𝐶𝑒𝑒 −1000𝑡𝑡

Since the circuit is closed at time 𝑡𝑡 = 0, 𝑖𝑖(0) = 0, and hence,

0 = 0.009 + 𝐶𝐶𝑒𝑒 0 ⇒ 𝐶𝐶 = −0.009
Thus, 𝑖𝑖 = 0.009 − 0.009𝑒𝑒 −1000𝑡𝑡 or 𝑖𝑖 = 0.009(1 − 𝑒𝑒 −1000𝑡𝑡 ).

10 q (milli coulomb)

8
6
4
2
t (milli second)
2 4 6 8 10
Figure 1.2. Graph of 𝑞𝑞 in terms of 𝑡𝑡 in Example 1.4.4
First Order Ordinary Differential Equations 16

Example 1.4.5
A series circuit consists of a capacitor of capacitance 0.01 F, a resistor of
resistance 0.1 kΩ and a voltage source 𝑉𝑉source = 200 𝑒𝑒 −𝑡𝑡 sin 𝑡𝑡. The capacitor is
charged with an initial charge 𝑞𝑞(0) = 1 C and satisfies the differential equation
𝑑𝑑𝑑𝑑
+ 𝑞𝑞 = 2 𝑒𝑒 −𝑡𝑡 sin 𝑡𝑡.
𝑑𝑑𝑑𝑑
Express the charge 𝑞𝑞 in terms of time 𝑡𝑡.

Solution
𝑃𝑃(𝑡𝑡) = 1 and 𝑄𝑄(𝑡𝑡) = 2 𝑒𝑒 −𝑡𝑡 sin 𝑡𝑡
𝐼𝐼(𝑡𝑡) = 𝑒𝑒 ∫ 1 𝑑𝑑𝑑𝑑 = 𝑒𝑒 𝑡𝑡

𝑒𝑒 𝑡𝑡 𝑞𝑞 = � 𝑒𝑒 𝑡𝑡 (2 𝑒𝑒 −𝑡𝑡 sin 𝑡𝑡)𝑑𝑑𝑑𝑑

⇒ 𝑒𝑒 𝑡𝑡 𝑞𝑞 = 2 � sin 𝑡𝑡 𝑑𝑑𝑑𝑑 = −2 cos 𝑡𝑡 + 𝐶𝐶

⇒ 𝑞𝑞 = −2𝑒𝑒 −𝑡𝑡 cos 𝑡𝑡 + 𝐶𝐶𝑒𝑒 −𝑡𝑡

𝑞𝑞(0) = 1 ⇒ 1 = −2𝑒𝑒 0 cos 0 + 𝐶𝐶𝑒𝑒 0 ⇒ 𝐶𝐶 = 3

Thus, 𝑞𝑞 = −2𝑒𝑒 −𝑡𝑡 cos 𝑡𝑡 + 3𝑒𝑒 −𝑡𝑡 or 𝑞𝑞 = 𝑒𝑒 −𝑡𝑡 (3 − 2 cos 𝑡𝑡).

q
1

0.5

t
1 2 3 4 5 6 7
Figure 1.3. Graph of 𝑞𝑞 in terms of 𝑡𝑡 in Example 1.4.5
First Order Ordinary Differential Equations 17

Example 1.4.6
A capacitor of capacitance 10 mF is fully charged with initial voltage 𝑉𝑉𝐶𝐶 (0) =
10 V. It is then discharged through a resistor of resistance 0.1 kΩ and satisfies
𝑑𝑑𝑉𝑉𝐶𝐶
+ 𝑉𝑉𝐶𝐶 = 0.
𝑑𝑑𝑑𝑑
Express the voltage across the capacitor 𝑉𝑉𝐶𝐶 in terms of time 𝑡𝑡.

Solution
𝑑𝑑𝑉𝑉𝐶𝐶 𝑑𝑑𝑉𝑉𝐶𝐶
+ 𝑉𝑉𝐶𝐶 = 0 ⇒ = −𝑉𝑉𝐶𝐶
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
1
⇒ 𝑑𝑑𝑉𝑉𝐶𝐶 = −𝑑𝑑𝑑𝑑
𝑉𝑉𝐶𝐶
1
⇒� 𝑑𝑑𝑉𝑉 = − � 𝑑𝑑𝑑𝑑
𝑉𝑉𝐶𝐶 𝐶𝐶
⇒ ln 𝑉𝑉𝐶𝐶 + 𝑘𝑘 = −𝑡𝑡

𝑉𝑉𝐶𝐶 (0) = 10 ⇒ ln 10 + 𝑘𝑘 = 0 ⇒ 𝑘𝑘 = − ln 10
Thus, ln 𝑉𝑉𝐶𝐶 − ln 10 = −𝑡𝑡 ⇒ ln(0.1 𝑉𝑉𝐶𝐶 ) = −𝑡𝑡 or 𝑉𝑉𝐶𝐶 = 10𝑒𝑒 −𝑡𝑡 .

10 Vc

8
6
4
2
t
2 4 6 8 10 12
Figure 1.4. Graph of 𝑉𝑉𝐶𝐶 in terms of 𝑡𝑡 in Example 1.4.6
First Order Ordinary Differential Equations 18

1.5 Exercises

Part A: Baseline Problems

1. Solve the following first order separable ODEs (see Section 1.2).
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
(a) = cos 𝑥𝑥 (f) (𝑥𝑥 2 + 1) = 𝑥𝑥 sec 𝑦𝑦
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑒𝑒 𝑥𝑥 𝑑𝑑𝑑𝑑
(b) = (g) 𝑥𝑥 = √𝑥𝑥 + 𝑥𝑥 sin 𝑥𝑥
𝑑𝑑𝑑𝑑 3𝑦𝑦 2 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
(c) 𝑥𝑥 = 2𝑦𝑦 (h) = 𝑒𝑒 2𝑡𝑡
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
(d) cos2 𝑥𝑥 −1=0 (i) = √𝑥𝑥𝑥𝑥
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑒𝑒 2𝑡𝑡−𝑥𝑥
(e) 𝑥𝑥 2 + 𝑦𝑦 2 =1 (j) =
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 𝑥𝑥

2. Solve the following first order linear ODEs (see Section 1.3).
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
(a) + 𝑦𝑦 = −𝑒𝑒 −𝑥𝑥 (f) 𝑥𝑥 + (1 + 𝑥𝑥)𝑦𝑦 = 𝑒𝑒 −𝑥𝑥 , where 𝑥𝑥 > 0
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 2 𝑑𝑑𝑑𝑑
(b) + 𝑦𝑦 = −𝑥𝑥 (g) sin 𝑥𝑥 + 𝑦𝑦 cos 𝑥𝑥 = sin 2𝑥𝑥 , where 0 < 𝑥𝑥 < 𝜋𝜋
𝑑𝑑𝑑𝑑 𝑥𝑥 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
(c) + 𝑦𝑦 = 𝑒𝑒 −𝑥𝑥 sin 2𝑥𝑥 (h) (𝑥𝑥 + 1) + 𝑦𝑦 = 𝑥𝑥 + 2, where 𝑥𝑥 > −1
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 3
(d) − 2𝑦𝑦 = 𝑒𝑒 2𝑥𝑥 cos 2𝑥𝑥 (i) (1 − 𝑥𝑥 2 ) + 4𝑦𝑦 =
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 1 + 𝑥𝑥
𝑑𝑑𝑑𝑑 cos 𝑥𝑥 𝑑𝑑𝑑𝑑
(e) 𝑥𝑥 + 2𝑦𝑦 = (j) (1 + 𝑥𝑥 2 ) + 2𝑥𝑥𝑥𝑥 = cos 4𝑥𝑥
𝑑𝑑𝑑𝑑 𝑥𝑥 𝑑𝑑𝑑𝑑
First Order Ordinary Differential Equations 19

Part B: Intermediate Problems

3. Solve the following Initial Value Problems (IVP) (see Example 1.2.7).
𝑑𝑑𝑑𝑑 1
=
(a) �𝑑𝑑𝑑𝑑 1 + 𝑥𝑥 2
𝑦𝑦(0) = 1

𝑑𝑑𝑑𝑑
= sec 2 𝑥𝑥
(b) �𝑑𝑑𝑑𝑑
𝑦𝑦(0) = −1

𝑦𝑦′ = 4𝑥𝑥𝑥𝑥
(c) �
𝑦𝑦(0) = 2

4. Solve the following Initial Value Problems (IVP).

𝑑𝑑𝑑𝑑
− 𝑦𝑦 = 𝑒𝑒 𝑥𝑥
(a) �𝑑𝑑𝑑𝑑
𝑦𝑦(0) = 4

𝑑𝑑𝑑𝑑 𝑦𝑦
+ = 𝑒𝑒 𝑥𝑥 , where 𝑥𝑥 > 0
(b) � 𝑑𝑑𝑑𝑑 𝑥𝑥
𝑦𝑦(1) = 0

𝑥𝑥𝑦𝑦 ′ + (1 + 𝑥𝑥)𝑦𝑦 = 𝑒𝑒 −𝑥𝑥 , where 𝑥𝑥 > 0

(c) �
𝑦𝑦(1) = 1
First Order Ordinary Differential Equations 20

Part C: Application Problems

(See Section 1.4)

5. In a chemical experiment, a tank is initially filled with 30 litres of pure

water. A chemical substance is being poured into the tank at the rate of 5
litres/minute. The mixture is kept uniform by constant stirring, and flows
out of the tank at the same rate as the chemical substance being poured
in. The amount of the substance in the tank satisfies the differential
equation
𝑑𝑑𝑑𝑑 5𝑥𝑥
= 10 − ,
𝑑𝑑𝑑𝑑 30
where 𝑥𝑥 denotes the amount of the substance at time 𝑡𝑡 minutes.

(a) Express 𝑥𝑥 in term of 𝑡𝑡.

(b) How much substance is present in the tank after an hour of mixing?

6. A forensic officer examined a corpse found in an air-conditioned room of

constant temperature 25°C. At the time of investigation, the corpse had
a temperature of 35°C. According to Newton’s Law of Cooling, the
cooling body satisfies the differential equation
𝑑𝑑𝑑𝑑
= −0.1054(𝑥𝑥 − 25),
𝑑𝑑𝑑𝑑
where 𝑥𝑥 denotes the temperature of the cooling body (in °C), and 𝑡𝑡 the
time (in hours).

(a) Express 𝑥𝑥 in term of 𝑡𝑡.

(b) Suppose the person died in the room and had a body temperature of
36°C at the time of death. How long had the body been in the room
since it was found?
First Order Ordinary Differential Equations 21

7. A white snooker ball of mass 0.16 kg travels at initial velocity 2 m/s in a

horizontal direction towards a blue snooker ball located 0.5 m away. The
motion of the white ball is subjected to resistance, and its velocity, 𝑣𝑣,
satisfies the differential equation
𝑑𝑑𝑑𝑑 1
=− ,
𝑑𝑑𝑑𝑑 0.32
where 𝑥𝑥 denotes the displacement of the white ball from its initial
position.

(a) Express 𝑣𝑣 in terms of 𝑥𝑥.

(b) Will the white ball hit the blue ball?

8. A ball of unit mass is projected vertically upwards with an initial velocity

3 m/s. The motion of the ball is subjected to earth’s gravity and air
resistance, and its velocity, 𝑣𝑣, satisfies the differential equation
𝑑𝑑𝑑𝑑
= −𝑔𝑔(1 + 𝑣𝑣),
𝑑𝑑𝑑𝑑
where 𝑔𝑔 ≈ 9.8 m/s2 is the constant of gravitational acceleration.

(a) Express 𝑣𝑣 in terms of 𝑡𝑡.

(b) How long does it take the ball to reach maximum height?

9. A 5-volt voltage source is connected to a 1 H inductor and a 0.5 kΩ

resistor in series. The current 𝑖𝑖 on the circuit satisfies the differential
equation
𝑑𝑑𝑑𝑑
+ 500𝑖𝑖 = 5.
𝑑𝑑𝑑𝑑
Given zero initial current in the circuit, express the current 𝑖𝑖 in terms of 𝑡𝑡.
First Order Ordinary Differential Equations 22

10. A fully charged 0.05 F capacitor is discharged through a 1 kΩ resistor. The

voltage across the capacitor 𝑉𝑉𝐶𝐶 satisfies the differential equation
𝑑𝑑𝑉𝑉𝐶𝐶
50 + 𝑉𝑉𝐶𝐶 = 0.
𝑑𝑑𝑑𝑑
Given an initial voltage of 25 volts across the capacitor, express the 𝑉𝑉𝐶𝐶 in
terms of 𝑡𝑡.

11. A series circuit consists of a capacitor of capacitance 0.0625 F, a resistor

of resistance 0.8 kΩ and a voltage source of 24 volts. The circuit is closed
and satisfies the differential equation
𝑑𝑑𝑑𝑑
10 + 2𝑞𝑞 = 3.
𝑑𝑑𝑑𝑑
Given that the capacitor has an initial charge of 0.5 C, express the charge
𝑞𝑞 in terms of 𝑡𝑡.

− END OF CHAPTER 1 −
Chapter 2
Second Order Linear ODE

Learning Objectives
By the end of this chapter, students should be able to
1. distinguish between homogeneous and non-homogeneous second order linear
ODEs;
2. solve homogeneous second order linear ODEs;
3. solve non-homogeneous second order linear ODEs using inverse D-operators;
4. solve engineering problems modelled by second order linear ODEs.

T his chapter shows how we solve second order linear ODEs with constant
coefficients. The method of using inverse D-operator is a simplified version of
the Undetermined Coefficients (UC) method.

We start with a discussion of solving homogeneous second order linear ODEs in Section
2.1. Next, we introduce the inverse D-operator to solve non-homogeneous ODEs in
Section 2.2. Applications of second order linear ODEs in engineering are demonstrated
in Section 2.3 before we end with exercises in Section 2.4.

23
Second Order Linear ODE 24

2.1 Homogeneous Second Order Linear ODE

𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
A second order linear ODE is an ODE of the form 𝑃𝑃2 2
+ 𝑃𝑃1 + 𝑃𝑃0 𝑦𝑦 = 𝑄𝑄 , where
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
𝑃𝑃2 , 𝑃𝑃1 , 𝑃𝑃0 , 𝑦𝑦, and 𝑄𝑄 are functions of 𝑥𝑥, with 𝑃𝑃2 ≠ 0. In this chapter, we consider second
order linear ODEs with constant coefficients, that is, 𝑃𝑃2 , 𝑃𝑃1 , 𝑃𝑃0 are all constants.

𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎 2 + 𝑏𝑏 + 𝑐𝑐 𝑦𝑦 = 𝑄𝑄
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
A second order linear ODE above is homogeneous if 𝑄𝑄 = 0, and non-homogeneous
otherwise.

Definition 2.1.1 (Homogeneous Second Order Linear ODE and Characteristic

Equation)
The homogeneous second order linear ODE with constant coefficients is of the form
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎2
+ 𝑏𝑏 + 𝑐𝑐 𝑦𝑦 = 0
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
and its characteristic equation is the quadratic equation
𝑎𝑎𝑚𝑚2 + 𝑏𝑏𝑏𝑏 + 𝑐𝑐 = 0
(𝑎𝑎, 𝑏𝑏, 𝑐𝑐 are constants with 𝑎𝑎 ≠ 0).

Remark. Definition 2.1.1 can be extended to higher order linear ODE with constant
coefficients. In some texts, the characteristic equation is also called the auxiliary
equation.

The solutions of a characteristic equation are called characteristic roots. Recall that
there are three types of solution for quadratic equations. The type of characteristic
roots will determine the general solution of the homogeneous linear ODE.
Second Order Linear ODE 25

Procedure in Solving Homogeneous Second Order Linear ODE

Step 1. Write the characteristic equation and solve for the characteristics roots.
Step 2. Let 𝛼𝛼 and 𝛽𝛽 be the characteristic roots. The general solution of the ODE is one
of the cases below.

Type of characteristic roots General Solution

Real with 𝛼𝛼 ≠ 𝛽𝛽 𝑦𝑦 = 𝐶𝐶1 𝑒𝑒 𝛼𝛼𝛼𝛼 + 𝐶𝐶2 𝑒𝑒 𝛽𝛽𝛽𝛽

Real with 𝛼𝛼 = 𝛽𝛽 𝑦𝑦 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 )𝑒𝑒 𝛼𝛼𝛼𝛼

Complex roots 𝑝𝑝 ± 𝑞𝑞𝑞𝑞 𝑦𝑦 = 𝑒𝑒 𝑝𝑝𝑝𝑝 (𝐶𝐶1 sin 𝑞𝑞𝑞𝑞 + 𝐶𝐶2 cos 𝑞𝑞𝑞𝑞)

𝐶𝐶1 and 𝐶𝐶2 are arbitrary constants.

Example 2.1.2
Solve 𝑦𝑦 ′′ − 4𝑦𝑦 = 0.

Solution
Step 1. Characteristic Equation: 𝑚𝑚2 − 4 = 0 ⇒ 𝑚𝑚 = 2 or 𝑚𝑚 = −2.
Step 2. The general solution is 𝑦𝑦 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥 .

Example 2.1.3
Solve 𝑦𝑦′′ + 4𝑦𝑦′ + 4𝑦𝑦 = 0.

Solution
Step 1. Characteristic Equation: 𝑚𝑚2 + 4𝑚𝑚 + 4 = 0 ⇒ 𝑚𝑚 = −2.
Step 2. The general solution is 𝑦𝑦 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 )𝑒𝑒 −2𝑥𝑥 .
Second Order Linear ODE 26

Example 2.1.4
𝑑𝑑2 𝑦𝑦
Solve + 4𝑦𝑦 = 0 .
𝑑𝑑𝑥𝑥 2

Solution
Step 1. Characteristic Equation: 𝑚𝑚2 + 4 = 0 ⇒ 𝑚𝑚 = ±2𝑗𝑗.
Step 2. The general solution is 𝑦𝑦 = 𝑒𝑒 0𝑥𝑥 (𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥)
or 𝑦𝑦 = 𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥.

Example 2.1.5
Solve the IVP:
𝑑𝑑2 𝑦𝑦
⎧ 2 + 4𝑦𝑦 = 0
⎪𝑑𝑑𝑥𝑥

⎨𝑦𝑦(0) = 1
⎪
⎩𝑦𝑦 ′ (0) = 1

Solution
𝑑𝑑2 𝑦𝑦
From previous example, the general solution of + 4𝑦𝑦 = 0 is
𝑑𝑑𝑥𝑥 2
𝑦𝑦 = 𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥.

𝑦𝑦(0) = 1 ⇒ 1 = 𝐶𝐶1 sin 0 + 𝐶𝐶2 cos 0 ⇒ 𝐶𝐶2 = 1

𝑑𝑑 𝑑𝑑
𝑦𝑦 ′ = 𝐶𝐶1 (sin 2𝑥𝑥) + 𝐶𝐶2 (cos 2𝑥𝑥) = 2𝐶𝐶1 cos 2𝑥𝑥 − 2𝐶𝐶2 sin 2𝑥𝑥
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑦𝑦 ′ (0) = 1 ⇒ 1 = 2𝐶𝐶1 cos 0 − 2𝐶𝐶2 sin 0 ⇒ 𝐶𝐶1 = 0.5
The particular solution is 𝑦𝑦 = 0.5 sin 2𝑥𝑥 + cos 2𝑥𝑥.
Second Order Linear ODE 27

2.2 Non-homogeneous Second Order Linear ODE

Definition 2.2.1 (Non-homogeneous Second Order Linear ODE)

The non-homogeneous second order linear ODE with constant coefficients is of the
form
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎 + 𝑏𝑏 + 𝑐𝑐𝑐𝑐 = 𝑄𝑄
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑
where 𝑄𝑄 is a non-zero function of 𝑥𝑥, and 𝑎𝑎, 𝑏𝑏, 𝑐𝑐 are constants with 𝑎𝑎 ≠ 0.

Observe that a non-homogeneous linear ODE can be written as

𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎 2 + 𝑏𝑏 + 𝑐𝑐𝑐𝑐 = 0 + Q.
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
If 𝑦𝑦𝑐𝑐 is the solution of the corresponding homogeneous linear ODE
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎 2 + 𝑏𝑏 + 𝑐𝑐𝑐𝑐 = 0
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
and 𝑦𝑦𝑝𝑝 is a solution of
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎 2 + 𝑏𝑏 + 𝑐𝑐𝑐𝑐 = 𝑄𝑄,
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
then the general solution of a non-homogeneous linear ODE is of the form
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝
where 𝑦𝑦𝑐𝑐 is the complementary function, and 𝑦𝑦𝑝𝑝 is the particular integral.

To find 𝑦𝑦𝑐𝑐 , we solve the corresponding homogeneous ODE as discussed in the previous
section. To find 𝑦𝑦𝑝𝑝 , we use inverse D-operators. An operator is a rule that assigns an
input function to an output function. A D-operator takes differentiable functions as its
inputs, and assigns the derivative of the input as the outputs, e.g. 𝐷𝐷(sin 𝑥𝑥) = cos 𝑥𝑥.

We use the abbreviated form 𝐷𝐷2 𝑦𝑦 to denote the second derivative of 𝑦𝑦. For instance,
𝐷𝐷2 (sin 𝑥𝑥) = 𝐷𝐷�𝐷𝐷(sin 𝑥𝑥)� = 𝐷𝐷(cos 𝑥𝑥) = − sin 𝑥𝑥.
Second Order Linear ODE 28

For our purpose, the linear ODE

𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
𝑎𝑎 2
+ 𝑏𝑏 + 𝑐𝑐𝑐𝑐 = 𝑄𝑄
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
is written in the D-operator form as
(𝑎𝑎𝐷𝐷2 + 𝑏𝑏𝑏𝑏 + 𝑐𝑐)𝑦𝑦 = 𝑄𝑄,
and 𝑦𝑦𝑝𝑝 is found as follows.
1
𝑦𝑦𝑝𝑝 = (𝑄𝑄)
𝑎𝑎𝐷𝐷2 + 𝑏𝑏𝑏𝑏 + 𝑐𝑐

1
The expression above is an example of an inverse D-operator, which
𝑎𝑎𝐷𝐷2
+ 𝑏𝑏𝑏𝑏 + 𝑐𝑐
1
takes the function 𝑄𝑄 as its input. For convenience, we use the notation to
𝐹𝐹(𝐷𝐷)
denote an inverse D-operator.

We can determine the particular integral 𝑦𝑦𝑝𝑝 if we know how the inverse D-operator
acts on various types of input function 𝑄𝑄:
1. 𝑄𝑄 = 𝑘𝑘, a constant function
2. 𝑄𝑄 = 𝑒𝑒 𝑘𝑘𝑘𝑘 , an exponential function
3. 𝑄𝑄 = 𝑒𝑒 𝑘𝑘𝑘𝑘 𝑉𝑉(𝑥𝑥), a product of 𝑒𝑒 𝑘𝑘𝑘𝑘 and constant/sine/cosine
4. 𝑄𝑄 = sin 𝑘𝑘𝑘𝑘 or 𝑄𝑄 = cos 𝑘𝑘𝑘𝑘 sine or cosine function

We will discuss all the cases above one by one.

Second Order Linear ODE 29

Inverse D-operator of a Constant Function

1 1
(𝑘𝑘) = (𝑘𝑘)
𝐹𝐹(𝐷𝐷) 𝐹𝐹(0)

In the formula above, 𝐹𝐹(0) is the value obtained by replacing 𝐷𝐷 in 𝐹𝐹(𝐷𝐷) with 0,
provided that 𝐹𝐹(0) ≠ 0. If 𝐹𝐹(0) = 0, we modify the method (see Example 2.2.3).

Example 2.2.2
𝑑𝑑2 𝑦𝑦
Solve − 4𝑦𝑦 = 8 .
𝑑𝑑𝑥𝑥 2

Solution
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑐𝑐 : Consider − 4𝑦𝑦 = 0 . From Example 2.1.2, we have
𝑑𝑑𝑥𝑥 2
𝑦𝑦𝑐𝑐 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑝𝑝 : Consider − 4𝑦𝑦 = 8 .
𝑑𝑑𝑥𝑥 2
1
(𝐷𝐷2 − 4)𝑦𝑦 = 8 ⇒ 𝑦𝑦𝑝𝑝 = (8)
𝐷𝐷2
−4
1 8
⇒ 𝑦𝑦𝑝𝑝 = 2 (8) = = −2
0 −4 −4
Thus, the general solution is
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥 − 2.
Second Order Linear ODE 30

1
By the definition of the D-operator, it is obvious that �𝑓𝑓(𝑥𝑥)� is the anti-derivative
𝐷𝐷
(indefinite integral) of 𝑓𝑓(𝑥𝑥), that is
1
�𝑓𝑓(𝑥𝑥)� = � 𝑓𝑓(𝑥𝑥) 𝑑𝑑𝑑𝑑.
𝐷𝐷

We apply the above formula to the next example.

Example 2.2.3
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
Solve 2
−4 =8 .
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑

Solution
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
Find 𝑦𝑦𝑐𝑐 : Consider − 4 =0 .
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑
Characteristic equation: 𝑚𝑚2 − 4𝑚𝑚 = 0 ⇒ 𝑚𝑚 = 0 or 𝑚𝑚 = 4.
𝑦𝑦𝑐𝑐 = 𝐶𝐶1 𝑒𝑒 0 + 𝐶𝐶2 𝑒𝑒 4𝑥𝑥 or 𝑦𝑦𝑐𝑐 = 𝐶𝐶1 + 𝐶𝐶2 𝑒𝑒 4𝑥𝑥
𝑑𝑑2 𝑦𝑦 𝑑𝑑𝑑𝑑
Find 𝑦𝑦𝑝𝑝 : Consider − 4 =8 .
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑
1
(𝐷𝐷2 − 4𝐷𝐷)𝑦𝑦 = 8 ⇒ 𝑦𝑦𝑝𝑝 = (8)
𝐷𝐷2 − 4𝐷𝐷
1 1 1 8
⇒ 𝑦𝑦𝑝𝑝 = � (8)� = � � = � −2 𝑑𝑑𝑑𝑑 = −2𝑥𝑥
𝐷𝐷 𝐷𝐷 − 4 𝐷𝐷 0 − 4
Thus, the general solution is
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = 𝐶𝐶1 + 𝐶𝐶2 𝑒𝑒 4𝑥𝑥 − 2𝑥𝑥.
Second Order Linear ODE 31

Inverse D-operator of an Exponential Function

1 1
(𝑒𝑒 𝑘𝑘𝑘𝑘 ) = (𝑒𝑒 𝑘𝑘𝑘𝑘 )
𝐹𝐹(𝐷𝐷) 𝐹𝐹(𝑘𝑘)

In the formula above, 𝐹𝐹(𝑘𝑘) is the value obtained by replacing 𝐷𝐷 in 𝐹𝐹(𝐷𝐷) with 𝑘𝑘,
provided that 𝐹𝐹(𝑘𝑘) ≠ 0. If 𝐹𝐹(𝑘𝑘) = 0, we use a different formula (see Example 2.2.5).

Example 2.2.4
𝑑𝑑 2 𝑦𝑦
Solve − 4𝑦𝑦 = 𝑒𝑒 4𝑥𝑥 .
𝑑𝑑𝑥𝑥 2

Solution
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑐𝑐 : Consider − 4𝑦𝑦 = 0 . From Example 2.1.2, we have
𝑑𝑑𝑥𝑥 2
𝑦𝑦𝑐𝑐 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥 .
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑝𝑝 : Consider − 4𝑦𝑦 = 𝑒𝑒 4𝑥𝑥 .
𝑑𝑑𝑥𝑥 2
1
(𝐷𝐷2 − 4)𝑦𝑦 = 𝑒𝑒 4𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = (𝑒𝑒 4𝑥𝑥 )
𝐷𝐷2
−4
1 1 4𝑥𝑥
⇒ 𝑦𝑦𝑝𝑝 = 2 (𝑒𝑒 4𝑥𝑥 ) = 𝑒𝑒
4 −4 12
Thus, the general solution is
1 4𝑥𝑥
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥 + 𝑒𝑒 .
12
Second Order Linear ODE 32

Inverse D-operator of a Product involving Exponential Function

1 1
�𝑒𝑒 𝑘𝑘𝑘𝑘 𝑉𝑉(𝑥𝑥)� = 𝑒𝑒 𝑘𝑘𝑘𝑘 �𝑉𝑉(𝑥𝑥)�
𝐹𝐹(𝐷𝐷) 𝐹𝐹(𝐷𝐷 + 𝑘𝑘)

In the formula above, 𝐹𝐹(𝐷𝐷 + 𝑘𝑘) is obtained by replacing 𝐷𝐷 in 𝐹𝐹(𝐷𝐷) with 𝐷𝐷 + 𝑘𝑘. This
new inverse D-operator then acts on the input 𝑉𝑉(𝑥𝑥), which may be a constant, sine, or
cosine function.

Example 2.2.5
𝑑𝑑 2 𝑦𝑦
Solve − 4𝑦𝑦 = 𝑒𝑒 2𝑥𝑥 .
𝑑𝑑𝑥𝑥 2

Solution
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑐𝑐 : Consider − 4𝑦𝑦 = 0 . From Example 2.1.2, we have
𝑑𝑑𝑥𝑥 2
𝑦𝑦𝑐𝑐 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥 .
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑝𝑝 : Consider 2
− 4𝑦𝑦 = 𝑒𝑒 2𝑥𝑥 .
𝑑𝑑𝑥𝑥
1
(𝐷𝐷2 − 4)𝑦𝑦 = 𝑒𝑒 2𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = (𝑒𝑒 2𝑥𝑥 )
𝐷𝐷2 −4
1 1
⇒ 𝑦𝑦𝑝𝑝 = 𝑒𝑒 2𝑥𝑥 (1) = 𝑒𝑒 2𝑥𝑥
(1)
(𝐷𝐷 + 2)2 − 4 𝐷𝐷2 + 4𝐷𝐷
1 1 1 1
⇒ 𝑦𝑦𝑝𝑝 = 𝑒𝑒 2𝑥𝑥 � (1)� = 𝑒𝑒 2𝑥𝑥 � 𝑑𝑑𝑑𝑑 = 𝑥𝑥 𝑒𝑒 2𝑥𝑥
𝐷𝐷 𝐷𝐷 + 4 4 4

Thus, the general solution is

1
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = 𝐶𝐶1 𝑒𝑒 2𝑥𝑥 + 𝐶𝐶2 𝑒𝑒 −2𝑥𝑥 + 𝑥𝑥𝑒𝑒 2𝑥𝑥 .
4
Second Order Linear ODE 33

Inverse D-operators of Sine and Cosine Functions

1 1
(sin 𝑘𝑘𝑘𝑘) = (sin 𝑘𝑘𝑘𝑘)
𝐹𝐹(𝐷𝐷 )
2 𝐹𝐹(−𝑘𝑘 2 )

1 1
(cos 𝑘𝑘𝑘𝑘) = (cos 𝑘𝑘𝑘𝑘)
𝐹𝐹(𝐷𝐷 )
2 𝐹𝐹(−𝑘𝑘 2 )

In the formula above, obtaining 𝐹𝐹(−𝑘𝑘 2 ) from 𝐹𝐹(𝐷𝐷2 ) means replacing only 𝐷𝐷2 by −𝑘𝑘 2 ,
provided that 𝐹𝐹(−𝑘𝑘 2 ) ≠ 0. If 𝐹𝐹(−𝑘𝑘 2 ) = 0, we use different formula (see Example
2.2.7).

Example 2.2.6
𝑑𝑑 2 𝑦𝑦
Solve + 4𝑦𝑦 = sin 𝑥𝑥 .
𝑑𝑑𝑥𝑥 2

Solution
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑐𝑐 : Consider + 4𝑦𝑦 = 0 . From Example 2.1.4, we have
𝑑𝑑𝑥𝑥 2
𝑦𝑦𝑐𝑐 = 𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥.
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑝𝑝 : Consider + 4𝑦𝑦 = sin 𝑥𝑥 .
𝑑𝑑𝑥𝑥 2
1
(𝐷𝐷2 + 4)𝑦𝑦 = sin 𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = (sin 𝑥𝑥)
𝐷𝐷2 + 4
1 1
⇒ 𝑦𝑦𝑝𝑝 = (sin 𝑥𝑥) = sin 𝑥𝑥
−12 + 4 3
Thus, the general solution is
1
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = 𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥 + sin 𝑥𝑥.
3
Second Order Linear ODE 34

Take note of the following special forms for the inverse D-operators of sine and cosine
functions.

Special Forms for Inverse D-operators of Sine and Cosine Functions

1 −𝑥𝑥 cos 𝑘𝑘𝑘𝑘
(sin 𝑘𝑘𝑘𝑘) =
𝐷𝐷2 + 𝑘𝑘 2 2𝑘𝑘

1 𝑥𝑥 sin 𝑘𝑘𝑘𝑘
(cos 𝑘𝑘𝑘𝑘) =
𝐷𝐷2 + 𝑘𝑘 2 2𝑘𝑘

Example 2.2.7
𝑑𝑑 2 𝑦𝑦
Solve + 4𝑦𝑦 = sin 2𝑥𝑥 .
𝑑𝑑𝑥𝑥 2

Solution
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑐𝑐 : Consider + 4𝑦𝑦 = 0 . From Example 2.1.4, we have
𝑑𝑑𝑥𝑥 2
𝑦𝑦𝑐𝑐 = 𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥.
𝑑𝑑 2 𝑦𝑦
Find 𝑦𝑦𝑝𝑝 : Consider + 4𝑦𝑦 = sin 2𝑥𝑥 .
𝑑𝑑𝑥𝑥 2
1
(𝐷𝐷2 + 4)𝑦𝑦 = sin 2𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = (sin 2𝑥𝑥)
𝐷𝐷2 + 4
−𝑥𝑥 cos 2𝑥𝑥
⇒ 𝑦𝑦𝑝𝑝 =
4
Thus, the general solution is
𝑥𝑥 cos 2𝑥𝑥
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = 𝐶𝐶1 sin 2𝑥𝑥 + 𝐶𝐶2 cos 2𝑥𝑥 − .
4
Second Order Linear ODE 35

In the example below, the inverse D-operator contains both 𝐷𝐷2 and 𝐷𝐷 terms.

Example 2.2.8
𝑑𝑑 2 𝑦𝑦 𝑑𝑑𝑑𝑑
Solve + 4 + 4𝑦𝑦 = sin 2𝑥𝑥 .
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑

Solution
𝑑𝑑 2 𝑦𝑦 𝑑𝑑𝑑𝑑
Find 𝑦𝑦𝑐𝑐 : Consider 2
+4 + 4𝑦𝑦 = 0 . From Example 2.1.3, we have
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
𝑦𝑦𝑐𝑐 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 ) 𝑒𝑒 −2𝑥𝑥 .
𝑑𝑑 2 𝑦𝑦 𝑑𝑑𝑑𝑑
Find 𝑦𝑦𝑝𝑝 : Consider + 4 + 4𝑦𝑦 = sin 2𝑥𝑥 .
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑
1
(𝐷𝐷2 + 4𝐷𝐷 + 4)𝑦𝑦 = sin 2𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = (sin 2𝑥𝑥)
𝐷𝐷2+ 4𝐷𝐷 + 4
1
⇒ 𝑦𝑦𝑝𝑝 = (sin 2𝑥𝑥)
−22 + 4𝐷𝐷 + 4
1
⇒ 𝑦𝑦𝑝𝑝 = (sin 2𝑥𝑥)
4𝐷𝐷
1 cos 2𝑥𝑥
⇒ 𝑦𝑦𝑝𝑝 = � sin 2𝑥𝑥 𝑑𝑑𝑑𝑑 = −
4 8
Thus, the general solution is
cos 2𝑥𝑥
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 ) 𝑒𝑒 −2𝑥𝑥 − .
8

When dealing with sine or cosine functions, we consider only 𝐷𝐷2 terms in the
expression of the inverse D-operator. If the expression does not contain any 𝐷𝐷2 terms,
a simple modification using the conjugate pair is necessary.
Second Order Linear ODE 36

1 1 1
Observe that � (sin 𝑘𝑘𝑘𝑘)� = 2 (sin 𝑘𝑘𝑘𝑘) implies
𝐷𝐷 − 𝑛𝑛 𝐷𝐷 + 𝑛𝑛 𝐷𝐷 − 𝑛𝑛2
1 1
(sin 𝑘𝑘𝑘𝑘) = (𝐷𝐷 − 𝑛𝑛) � 2 (sin 𝑘𝑘𝑘𝑘)�.
𝐷𝐷 + 𝑛𝑛 𝐷𝐷 − 𝑛𝑛2
The roles of (𝐷𝐷 + 𝑛𝑛) and (𝐷𝐷 − 𝑛𝑛) are interchangeable.

Example 2.2.9
Solve 𝑦𝑦 ′′ + 4𝑦𝑦 ′ + 4𝑦𝑦 = sin 𝑥𝑥.

Solution
Find 𝑦𝑦𝑐𝑐 : Consider 𝑦𝑦 ′′ + 4𝑦𝑦 ′ + 4𝑦𝑦 = 0. From Example 2.1.3, we have
𝑦𝑦𝑐𝑐 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 ) 𝑒𝑒 −2𝑥𝑥 .

Find 𝑦𝑦𝑝𝑝 : Consider 𝑦𝑦 ′′ + 4𝑦𝑦 ′ + 4𝑦𝑦 = sin 𝑥𝑥.

1
(𝐷𝐷2 + 4𝐷𝐷 + 4)𝑦𝑦 = sin 𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = (sin 𝑥𝑥)
𝐷𝐷2 + 4𝐷𝐷 + 4
1 1
⇒ 𝑦𝑦𝑝𝑝 = (sin 𝑥𝑥) = (sin 𝑥𝑥)
−12 + 4𝐷𝐷 + 4 4𝐷𝐷 + 3
1 1
⇒ 𝑦𝑦𝑝𝑝 = (4𝐷𝐷 − 3) � 2
(sin 𝑥𝑥)� = (4𝐷𝐷 − 3) � (sin 𝑥𝑥)�
16𝐷𝐷 − 9 −16 − 9
4 3 4 3
⇒ 𝑦𝑦𝑝𝑝 = 𝐷𝐷(sin 𝑥𝑥) − (sin 𝑥𝑥) = − cos 𝑥𝑥 + sin 𝑥𝑥
−25 −25 25 25

Thus, the general solution is

4 3
𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 ) 𝑒𝑒 −2𝑥𝑥 − cos 𝑥𝑥 + sin 𝑥𝑥.
25 25
Second Order Linear ODE 37

The following example uses a combination of a few formulas.

Example 2.2.10
Solve 𝑦𝑦 ′′ + 4𝑦𝑦 ′ + 4𝑦𝑦 = 𝑒𝑒 −2𝑥𝑥 sin 𝑥𝑥.

Solution
Find 𝑦𝑦𝑐𝑐 : Consider 𝑦𝑦 ′′ + 4𝑦𝑦 ′ + 4𝑦𝑦 = 0. From Example 2.1.3, we have
𝑦𝑦𝑐𝑐 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 ) 𝑒𝑒 −2𝑥𝑥 .

Find 𝑦𝑦𝑝𝑝 : Consider 𝑦𝑦 ′′ + 4𝑦𝑦 ′ + 4𝑦𝑦 = 𝑒𝑒 −2𝑥𝑥 sin 𝑥𝑥.

1
(𝐷𝐷2 + 4𝐷𝐷 + 4)𝑦𝑦 = 𝑒𝑒 −2𝑥𝑥 sin 𝑥𝑥 ⇒ 𝑦𝑦𝑝𝑝 = 2
(𝑒𝑒 −2𝑥𝑥 sin 𝑥𝑥)
𝐷𝐷 + 4𝐷𝐷 + 4
1
⇒ 𝑦𝑦𝑝𝑝 = 𝑒𝑒 −2𝑥𝑥 (sin 𝑥𝑥)
(𝐷𝐷 − 2)2 + 4(𝐷𝐷 − 2) + 4

1
⇒ 𝑦𝑦𝑝𝑝 = 𝑒𝑒 −2𝑥𝑥 (sin 𝑥𝑥)
(𝐷𝐷2 − 4𝐷𝐷 + 4) + (4𝐷𝐷 − 8) + 4
1
⇒ 𝑦𝑦𝑝𝑝 = 𝑒𝑒 −2𝑥𝑥 (sin 𝑥𝑥)
𝐷𝐷2
1
⇒ 𝑦𝑦𝑝𝑝 = 𝑒𝑒 −2𝑥𝑥 � sin 𝑥𝑥� = −𝑒𝑒 −2𝑥𝑥 sin 𝑥𝑥
−1

Thus, the general solution is

𝑦𝑦 = 𝑦𝑦𝑐𝑐 + 𝑦𝑦𝑝𝑝 = (𝐶𝐶1 𝑥𝑥 + 𝐶𝐶2 ) 𝑒𝑒 −2𝑥𝑥 − 𝑒𝑒 −2𝑥𝑥 sin 𝑥𝑥.
Second Order Linear ODE 38

2.3 Applications
In this section, we explore engineering problems modelled by second order linear
differential equations. Examples in electronics and mechanics are presented together
with brief descriptions on the modelling process.

Applications in Electronics

Recall the Kirchoff’s Voltage Law (KVL) used in circuit analysis:

𝑉𝑉𝑅𝑅 + 𝑉𝑉𝐿𝐿 + 𝑉𝑉𝐶𝐶 = 𝑉𝑉source .
Using the following formulas
𝑑𝑑𝑑𝑑 𝑞𝑞 𝑑𝑑𝑑𝑑
𝑉𝑉𝑅𝑅 = 𝑖𝑖𝑖𝑖, 𝑉𝑉𝐿𝐿 = 𝐿𝐿 𝑑𝑑𝑑𝑑, 𝑉𝑉𝐶𝐶 = 𝐶𝐶, 𝑖𝑖 = 𝑑𝑑𝑑𝑑 ,
the KVL translates to the second order linear differential equation
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑 1
𝐿𝐿 2 + 𝑅𝑅 + 𝑞𝑞 = 𝑉𝑉source .
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑 𝐶𝐶

The next example is an electronics problem modelled by a linear differential equation

𝑑𝑑 2 𝑞𝑞
involving . We solve for the particular solution by considering initial charge 𝑞𝑞(0)
𝑑𝑑𝑡𝑡 2
and initial current 𝑖𝑖(0) = 𝑞𝑞′(0).

Example 2.3.1
A series circuit consists of a 1 H inductor, a 1 kΩ resistor, and a 6.25µF capacitor. The
capacitor was initially charged at 1.5 mC. The circuit is then closed at time 𝑡𝑡 = 0,
allowing the capacitor to discharge, and satisfies the differential equation
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑
2
+ 1000 + 160000𝑞𝑞 = 0.
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑
Express the charge 𝑞𝑞 in terms of time 𝑡𝑡.
Second Order Linear ODE 39

Solution
Characteristic equation: 𝑚𝑚2 + 1000𝑚𝑚 + 160000 = 0 ⇒ 𝑚𝑚 = −800 or 𝑚𝑚 = −200.
So, 𝑞𝑞 = 𝐶𝐶1 𝑒𝑒 −800𝑡𝑡 + 𝐶𝐶2 𝑒𝑒 −200𝑡𝑡 .

𝑞𝑞(0) = 0.0015 ⇒ 0.0015 = 𝐶𝐶1 𝑒𝑒 0 + 𝐶𝐶2 𝑒𝑒 0 ⇒ 𝐶𝐶1 + 𝐶𝐶2 = 0.0015

𝑞𝑞 ′ = −800 𝐶𝐶1 𝑒𝑒 −800𝑡𝑡 − 200 𝐶𝐶2 𝑒𝑒 −200𝑡𝑡

𝑞𝑞 ′ (0) = 0 ⇒ 0 = −800 𝐶𝐶1 𝑒𝑒 0 − 200 𝐶𝐶2 𝑒𝑒 0 ⇒ 4𝐶𝐶1 + 𝐶𝐶2 = 0

Solving the linear system for 𝐶𝐶1 and 𝐶𝐶2 , we have 𝐶𝐶1 = −0.0005 and 𝐶𝐶2 = 0.002.
Thus, 𝑞𝑞 = −0.0005𝑒𝑒 −800𝑡𝑡 + 0.002𝑒𝑒 −200𝑡𝑡 .

q (mC)

1.5

0.5

t (msec)
2 4 6 8 10 12 14 16 18 20

Figure 2.1. Graph of 𝑞𝑞 (millicoulomb) vs. 𝑡𝑡 (millisecond) in Example 2.3.1

Second Order Linear ODE 40

Applications in Mechanics

An interesting topic in classical mechanics is the harmonic motion, e.g. an object moves
back and forth due to the potential energy of a spring.

(a) vertical (b) horizontal

Figure 2.2 Vertical and horizontal harmonic montions

In studying harmonic motion, the Hooke’s Law, 𝐹𝐹𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = −𝑘𝑘𝑘𝑘, is considered. Here, 𝑥𝑥
denotes the displacement of the object from its initial position, and 𝑘𝑘 is a spring
constant that determines the stiffness of the spring.

In the next two examples, we see problems in harmonic motion modelled by second
𝑑𝑑2 𝑥𝑥
order differential equations involving . We solve for the particular solution by
𝑑𝑑𝑡𝑡 2
considering initial displacement 𝑥𝑥(0) and initial velocity 𝑣𝑣(0) = 𝑥𝑥 ′ (0).

Example 2.3.2
A spring with spring constant of 1 N/m is lying horizontally on a smooth surface with
one end fixed to a wall as shown in Figure 2.2(b) above. An object of unit mass is
attached to the other end of the spring without stretching it. Let 𝑥𝑥 be the displacement
of the object from its equilibrium position. The object projected horizontally away from
the wall with initial velocity 1 m/s satisfies the differential equation
𝑑𝑑2 𝑥𝑥
+ 𝑥𝑥 = 0.
𝑑𝑑𝑡𝑡 2
Express the displacement 𝑥𝑥 as a function of time 𝑡𝑡.
Second Order Linear ODE 41

Solution
Characteristic equation: 𝑚𝑚2 + 1 = 0 ⇒ 𝑚𝑚 = 𝑗𝑗 or 𝑚𝑚 = −𝑗𝑗.
So, 𝑥𝑥 = 𝐶𝐶1 sin 𝑡𝑡 + 𝐶𝐶2 cos 𝑡𝑡.

Initially, the spring is not stretched ⇒ 𝑥𝑥(0) = 0.

𝑥𝑥(0) = 0 ⇒ 0 = 𝐶𝐶1 sin 0 + 𝐶𝐶2 cos 0 ⇒ 𝐶𝐶2 = 0

Initial velocity is 1 m/s ⇒ 𝑥𝑥 ′ (0) = 1.

𝑥𝑥 ′ = 𝐶𝐶1 cos 𝑡𝑡 − 𝐶𝐶2 sin 𝑡𝑡
𝑥𝑥 ′ (0) = 1 ⇒ 1 = 𝐶𝐶1 cos 0 − 𝐶𝐶2 sin 0 ⇒ 𝐶𝐶1 = 1

Thus, 𝑥𝑥 = sin 𝑡𝑡.

The solution in Example 2.3.2 indicates that the object is moving back and forth,
pushing and pulling the spring alternately forever. This is an ideal situation with no
resistance involved in the harmonic motion. In a real situation, this is not possible.

Example 2.3.3
A system consists of a spring of stiffness 0.625 N/m and a 25 gram ball suspended by
the spring as shown in Figure 2.2(a) in the previous page. The other end of the spring is
fixed onto a ceiling. The ball is pulled downwards 50 cm from its equilibrium position,
and released. Let 𝑣𝑣 be the velocity of the ball, and 𝑥𝑥 its displacement from its
equilibrium position. The system experiences a resistance 0.25𝑣𝑣, and satisfies the
differential equation
𝑑𝑑2 𝑥𝑥 𝑑𝑑𝑑𝑑
+ 10 + 25𝑥𝑥 = 0.
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑
Express the displacement 𝑥𝑥 as a function of time 𝑡𝑡.
Second Order Linear ODE 42

Solution
Characteristic equation: 𝑚𝑚2 + 10𝑚𝑚 + 25 = 0 ⇒ 𝑚𝑚 = −5.
So, 𝑥𝑥 = (𝐶𝐶1 + 𝐶𝐶2 𝑡𝑡)𝑒𝑒 −5𝑡𝑡 .

Initially, the spring is pulled 50 cm downwards ⇒ 𝑥𝑥(0) = −0.5.

𝑥𝑥(0) = −0.5 ⇒ −0.5 = (𝐶𝐶1 + 0)𝑒𝑒 0 ⇒ 𝐶𝐶1 = −0.5

The ball was released after being pulled ⇒ 𝑥𝑥 ′ (0) = 0.

𝑥𝑥 ′ = (−5 𝐶𝐶1 − 5 𝐶𝐶2 𝑡𝑡 + 𝐶𝐶2 )𝑒𝑒 −5𝑡𝑡
𝑥𝑥 ′ (0) = 0 ⇒ 0 = (−5 𝐶𝐶1 − 0 + 𝐶𝐶2 )𝑒𝑒 0 ⇒ 𝐶𝐶2 = −2.5

Thus, 𝑥𝑥 = (−0.5 − 2.5 𝑡𝑡)𝑒𝑒 −5𝑡𝑡 .

x
0.1
t
1 2
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6

Figure 2.2. Graph of 𝑥𝑥 vs. 𝑡𝑡 for harmonic motion in Example 2.3.3

Second Order Linear ODE 43

2.4 Exercises

Part A: Baseline Problems

1. Solve the following homogeneous second order linear ODEs, see Section 2.1.
(a) 𝑦𝑦 ′′ + 4𝑦𝑦 ′ − 5𝑦𝑦 = 0
(b) 𝑦𝑦 ′′ + 6𝑦𝑦 ′ + 9𝑦𝑦 = 0
(c) 𝑦𝑦 ′′ − 16𝑦𝑦 = 0
𝑑𝑑 2 𝑦𝑦 𝑑𝑑𝑑𝑑
(d) +2 + 2𝑦𝑦 = 0
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑
𝑑𝑑 2 𝑦𝑦 𝑑𝑑𝑑𝑑
(e) 6 − 11 + 3𝑦𝑦 = 0
𝑑𝑑𝑥𝑥 2 𝑑𝑑𝑑𝑑

2. Solve the following homogeneous second order linear ODEs, see Section 2.1.
𝑑𝑑 2 𝑥𝑥 𝑑𝑑𝑑𝑑
(a) + =0
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑
𝑑𝑑 2 𝑥𝑥 𝑑𝑑𝑑𝑑
(b) −6 + 10𝑥𝑥 = 0
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑
𝑑𝑑 2 𝑞𝑞
(c) + 4𝑞𝑞 = 0
𝑑𝑑𝑡𝑡 2
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑
(d) +2 − 𝑞𝑞 = 0
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑
𝑑𝑑 2 𝑧𝑧 𝑑𝑑𝑑𝑑
(e) 4 2 − 12 + 9𝑧𝑧 = 0
𝑑𝑑𝑥𝑥 𝑑𝑑𝑑𝑑
Second Order Linear ODE 44

Part B: Intermediate Problems

3. Solve the following Initial Value Problems (IVP), see Example 2.1.5.
𝑦𝑦 ′′ + 4𝑦𝑦 ′ − 5𝑦𝑦 = 0
(a) �𝑦𝑦(0) = 1
𝑦𝑦 ′ (0) = 1
2
⎧𝑑𝑑 𝑥𝑥 + 𝑑𝑑𝑑𝑑 = 0
⎪ 𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑
(b)
⎨𝑥𝑥(0) = 0
⎪ ′
⎩𝑥𝑥 (0) = 1
2
⎧𝑑𝑑 𝑞𝑞 + 4𝑞𝑞 = 0
⎪ 𝑑𝑑𝑡𝑡 2
(c)
⎨𝑞𝑞(0) = 1
⎪ ′
⎩𝑞𝑞 (0) = 0

For Questions 4 – 6, see Examples 2.2.2 and 2.2.3.

1
4. (a) Find (15).
𝐷𝐷2 + 4𝐷𝐷 − 5
(b) Solve 𝑦𝑦 ′′ + 4𝑦𝑦′ − 5𝑦𝑦 = 15.

1
5. (a) Find (21).
𝐷𝐷2 + 6𝐷𝐷 + 9
(b) Solve 𝑦𝑦′′ + 6𝑦𝑦′ + 9𝑦𝑦 = 21.
Second Order Linear ODE 45

1
6. (a) Find (2).
𝐷𝐷2 + 𝐷𝐷
𝑑𝑑 2 𝑥𝑥 𝑑𝑑𝑑𝑑
(b) Solve + = 2.
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑

For Questions 7 – 9, see Example 2.2.4.

1
7. (a) Find (𝑒𝑒 𝑥𝑥 ).
𝐷𝐷2 + 6𝐷𝐷 + 9
(b) Solve 𝑦𝑦′′ + 6𝑦𝑦′ + 9𝑦𝑦 = 8𝑒𝑒 𝑥𝑥 .

1
8. (a) Find (𝑒𝑒 2𝑥𝑥 ).
𝐷𝐷2 − 1
(b) Solve 𝑦𝑦 ′′ − 𝑦𝑦 = 𝑒𝑒 2𝑥𝑥 .

1
9. (a) Find (𝑒𝑒 5𝑡𝑡 ).
𝐷𝐷2 + 𝐷𝐷
𝑑𝑑 2 𝑥𝑥 𝑑𝑑𝑑𝑑
(b) Solve + = 20𝑒𝑒 5𝑡𝑡 .
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑

For Questions 10 – 11, see Example 2.2.5.

1
10. (a) Find (𝑒𝑒 −3𝑥𝑥 ).
𝐷𝐷2 + 6𝐷𝐷 + 9
(b) Solve 𝑦𝑦′′ + 6𝑦𝑦′ + 9𝑦𝑦 = 𝑒𝑒 −3𝑥𝑥 .
Second Order Linear ODE 46

1
11. (a) Find (𝑒𝑒 −𝑡𝑡 ).
𝐷𝐷2 + 𝐷𝐷
𝑑𝑑 2 𝑥𝑥 𝑑𝑑𝑑𝑑
(b) Solve + = 𝑒𝑒 −𝑡𝑡 .
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑

For Questions 12 – 14, see Examples 2.2.6 and 2.2.7.

1
12. (a) Find (4 cos 𝑥𝑥).
𝐷𝐷2 −1
(b) Solve 𝑦𝑦 ′′ − 𝑦𝑦 = 4 cos 𝑥𝑥.

1
13. (a) Find (4 cos 𝑥𝑥).
𝐷𝐷2 + 1
(b) Solve 𝑦𝑦 ′′ + 𝑦𝑦 = 4 cos 𝑥𝑥.

1
14. (a) Find (cos 3𝑡𝑡).
𝐷𝐷2 +9
𝑑𝑑 2 𝑞𝑞
(b) Solve + 9𝑞𝑞 = cos 3𝑡𝑡
𝑑𝑑𝑡𝑡 2

For Questions 15 – 17, see Examples 2.2.8 and 2.2.9.

1
15. (a) Find (sin 3𝑥𝑥).
𝐷𝐷2 + 6𝐷𝐷 + 9
(b) Solve 𝑦𝑦′′ + 6𝑦𝑦′ + 9𝑦𝑦 = sin 3𝑥𝑥.
Second Order Linear ODE 47

1
16. (a) Find (sin 𝑥𝑥).
𝐷𝐷2 + 2𝐷𝐷 + 2
(b) Solve 𝑦𝑦′′ + 2𝑦𝑦′ + 2𝑦𝑦 = sin 𝑥𝑥.

1
17. (a) Find (cos 3𝑡𝑡).
𝐷𝐷2 + 𝐷𝐷
𝑑𝑑 2 𝑥𝑥 𝑑𝑑𝑑𝑑
(b) Solve + = cos 3𝑡𝑡.
𝑑𝑑𝑡𝑡 2 𝑑𝑑𝑑𝑑

For Questions 18 – 19, see Example 2.2.10.

1
18. (a) Find (𝑒𝑒 −3𝑥𝑥 sin 𝑥𝑥).
𝐷𝐷2 + 6𝐷𝐷 + 9
(b) Solve 𝑦𝑦′′ + 6𝑦𝑦′ + 9𝑦𝑦 = 𝑒𝑒 −3𝑥𝑥 sin 𝑥𝑥.

1
19. (a) Find (𝑒𝑒 −𝑥𝑥 cos 𝑥𝑥).
𝐷𝐷2 + 2𝐷𝐷 + 2
(b) Solve 𝑦𝑦′′ + 2𝑦𝑦′ + 2𝑦𝑦 = 𝑒𝑒 −𝑥𝑥 cos 𝑥𝑥.

20. Solve the following non-homogeneous second order linear ODEs.

(a) 𝑦𝑦 ′′ + 4𝑦𝑦 = sin 𝑥𝑥 + 𝑒𝑒 2𝑥𝑥 + 1
(b) 𝑦𝑦 ′′ − 16𝑦𝑦 = 6 sin 2𝑥𝑥 − 8 cos 𝑥𝑥 + 4
(c) 𝑦𝑦 ′′ + 6𝑦𝑦 ′ + 9𝑦𝑦 = sin 3𝑥𝑥 − 2𝑒𝑒 −3𝑥𝑥
Second Order Linear ODE 48

Part C: Application Problems

See Section 2.3.

21. A series circuit consists of a capacitor of capacitance 0.4 mF, an inductor of

inductance 1 H, a resistor of resistance 0.1 kΩ. The capacitor was initially charged
at 0.01 C. The circuit is then closed at time 𝑡𝑡 = 0 and satisfies the differential
equation
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑
2
+ 100 + 2500𝑞𝑞 = 0.
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑
Express the charge 𝑞𝑞 in terms of 𝑡𝑡.

22. A 2 kg metal ball is suspended by a spring of stiffness 10 N/m. The ball is pulled
down 0.1 m below its equilibrium position and released. Let 𝑥𝑥 be the
displacement of the object from its equilibrium position. The harmonic motion of
the ball is subjected to air resistance, and satisfies the differential equation
𝑑𝑑2 𝑥𝑥 𝑑𝑑𝑑𝑑
+2 + 5𝑥𝑥 = 0.
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
Express the displacement 𝑥𝑥 as a function of 𝑡𝑡.

− END OF CHAPTER 2 −
Chapter 3
Laplace Transform

Learning Objectives
By the end of this chapter, students should be able to
1. evaluate Laplace and inverse Laplace transforms of elementary functions;
2. evaluate Laplace and inverse Laplace transforms of a product involving 𝑒𝑒 𝑎𝑎𝑎𝑎 ;
3. evaluate Laplace and inverse Laplace transforms of a truncated shift of a function;
4. solve initial value problems (IVP) using Laplace transforms.

L aplace transforms derive their name from the French scholar Pierre-Simon
Laplace (1749 – 1827), who did pivotal work in mathematics, statistics, physics,
and astronomy. These transforms are useful for solving engineering problems
involving initial value problems (IVP).

Section 3.1 introduces the definition of the Laplace transform and its basic properties.
This is followed by a discussion on the Laplace transforms of elementary functions
(polynomials, exponential, and trigonometric functions), as well as the unit-step
function, the delta function, and convolution of functions.

Sections 3.2 and 3.3 are dedicated to evaluating Laplace transforms of products
involving exponential function 𝑒𝑒 𝑎𝑎𝑎𝑎 or unit-step function 𝑈𝑈(𝑡𝑡 − 𝑎𝑎); whereas Section 3.4
presents the Laplace transform as an elegant algebraic tool for solving IVPs of linear
differential equations. As usual, we conclude the chapter with exercises in Section 3.5.

49
Laplace Transforms 50

3.1 Definition and Formulas

In engineering applications, we commonly use a mathematical function in time (or a
function in 𝑡𝑡-domain) to represent a signal, for example, electrical signal. The Laplace
transform maps input functions in 𝑡𝑡-domain to output functions in 𝑠𝑠-domain (see
Figure 3.1 and Definition 3.1.1).

We use lower-case letters to denote functions in the 𝑡𝑡-domain, and upper-case letters
to denote functions in the 𝑠𝑠-domain. Figure 3.1 illustrates the role of Laplace
transform.

ℒ
𝑓𝑓(𝑡𝑡) 𝐹𝐹(𝑠𝑠)

Figure 3.1. The Laplace transform maps functions in the 𝑡𝑡-domain to functions in the 𝑠𝑠-
domain

Definition 3.1.1 (Laplace Transform)

The Laplace transform of 𝑓𝑓(𝑡𝑡), is a function 𝐹𝐹(𝑠𝑠), defined by
𝑅𝑅
𝐹𝐹(𝑠𝑠) = lim � 𝑒𝑒 −𝑠𝑠𝑠𝑠 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑
𝑅𝑅→∞ 0

We use the notation ℒ[𝑓𝑓(𝑡𝑡)] to denote the Laplace transform of 𝑓𝑓(𝑡𝑡).

𝑅𝑅 ∞
Remark. The notation lim � 𝑒𝑒 −𝑠𝑠𝑠𝑠
𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑 above can be written as � 𝑒𝑒 −𝑠𝑠𝑠𝑠 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑 ,
𝑅𝑅→∞ 0 0
i.e. as an improper integral.
Laplace Transforms 51

The inverse Laplace transform has the opposite role, as illustrated by Figure 3.2.

𝑓𝑓(𝑡𝑡) 𝐹𝐹(𝑠𝑠)

ℒ −1

Figure 3.2. The inverse Laplace transform maps functions in the 𝑠𝑠-domain to functions
in the 𝑡𝑡-domains

The Laplace transforms of elementary functions are shown without proof in Table 3.1.
We shall use them directly for subsequent problems.

𝑓𝑓(𝑡𝑡) ℒ[𝑓𝑓(𝑡𝑡)]
𝑛𝑛!
𝑡𝑡 𝑛𝑛
𝑠𝑠 𝑛𝑛+1
1
𝑒𝑒 𝑎𝑎𝑎𝑎
𝑠𝑠 − 𝑎𝑎
𝑎𝑎
sin 𝑎𝑎𝑎𝑎
𝑠𝑠 2 + 𝑎𝑎2
𝑠𝑠
cos 𝑎𝑎𝑎𝑎
𝑠𝑠 2 + 𝑎𝑎2
ℒ −1 [𝐹𝐹(𝑠𝑠)] 𝐹𝐹(𝑠𝑠)
Table 3.1 Laplace transforms of elementary functions

Remark. For 𝑛𝑛 = 0, 1, 2, …, the notation 𝑛𝑛! is read 𝑛𝑛 factorial. It is the product of all
integers from 1 to 𝑛𝑛, in other words, 𝑛𝑛! = 𝑛𝑛 ⋅ (𝑛𝑛 − 1) ⋅ (𝑛𝑛 − 2) ⋯ 2 ⋅ 1.
Laplace Transforms 52

The first two formulas in Table 3.1 can be written as follows.

𝑛𝑛! 𝑛𝑛!
ℒ[𝑡𝑡 𝑛𝑛 ] = or ℒ −1 � � = 𝑡𝑡 𝑛𝑛
𝑠𝑠 𝑛𝑛+1 𝑠𝑠 𝑛𝑛+1
1 1
ℒ[𝑒𝑒 𝑎𝑎𝑎𝑎 ] = or ℒ −1 � � = 𝑒𝑒 𝑎𝑎𝑎𝑎
𝑠𝑠 − 𝑎𝑎 𝑠𝑠 − 𝑎𝑎

Example 3.1.2
3! 6
(a) ℒ[𝑡𝑡 3 ] = =
𝑠𝑠 3+1 𝑠𝑠 4
4! 24
(b) ℒ[𝑡𝑡 4 ] = =
𝑠𝑠 4+1 𝑠𝑠 5
6 3!
(c) ℒ −1 � 4 � = ℒ −1 � 3+1 � = 𝑡𝑡 3
𝑠𝑠 𝑠𝑠
1 1 2! 1
(d) ℒ −1 � 3 � = ℒ −1 � 2+1 � = 𝑡𝑡 2
𝑠𝑠 2! 𝑠𝑠 2

Example 3.1.3
1
(a) ℒ[𝑒𝑒 𝑡𝑡 ] =
𝑠𝑠 − 1
1 1
(b) ℒ[𝑒𝑒 −𝑡𝑡 ] = =
𝑠𝑠 − (−1) 𝑠𝑠 + 1
1
(c) ℒ −1 � � = 𝑒𝑒 𝑡𝑡
𝑠𝑠 − 1
1 1
(d) ℒ −1 � � = ℒ −1 � � = 𝑒𝑒 −2𝑡𝑡
𝑠𝑠 + 2 𝑠𝑠 − (−2)
Laplace Transforms 53

The Laplace transform for sine and cosine functions can be written as follows.

𝑎𝑎 𝑎𝑎
ℒ[sin 𝑎𝑎𝑎𝑎] = or ℒ −1 � � = sin 𝑎𝑎𝑎𝑎
𝑠𝑠 2 + 𝑎𝑎2 𝑠𝑠 2 + 𝑎𝑎2
𝑠𝑠 𝑠𝑠
ℒ[cos 𝑎𝑎𝑎𝑎] = or ℒ −1 � � = cos 𝑎𝑎𝑎𝑎
𝑠𝑠 2 + 𝑎𝑎2 𝑠𝑠 2 + 𝑎𝑎2

Example 3.1.4
2 2
(a) ℒ[sin 2𝑡𝑡] = = 2
𝑠𝑠 2 +22 𝑠𝑠 + 4
s 𝑠𝑠
(b) ℒ[cos 2𝑡𝑡] = = 2
𝑠𝑠 2 +22 𝑠𝑠 + 4
1 1 −1 2 1
(c) ℒ −1 � � = ℒ � � = sin 2𝑡𝑡
𝑠𝑠 2 + 4 2 𝑠𝑠 2 + 22 2
s 𝑠𝑠
(d) ℒ −1 � � = ℒ −1
� � = cos 2𝑡𝑡
𝑠𝑠 2 + 4 𝑠𝑠 2 + 22

The unit-step function, or Heaviside function, is a piecewise defined function given by

0 for 𝑡𝑡 < 𝑎𝑎
𝑈𝑈(𝑡𝑡 − 𝑎𝑎) = �
1 for 𝑡𝑡 ≥ 𝑎𝑎

1
t
a
Figure 3.3. Graph of 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)
Laplace Transforms 54

Given a function 𝑓𝑓(𝑡𝑡), the product 𝑓𝑓(𝑡𝑡)𝑈𝑈(𝑡𝑡 − 𝑎𝑎) describes the truncated function at
𝑡𝑡 = 𝑎𝑎, that is, the function is cut-off prior to 𝑡𝑡 = 𝑎𝑎 and resumes from 𝑡𝑡 = 𝑎𝑎 onwards.
Figure 3.4 shows the graphical comparisons between 𝑓𝑓(𝑡𝑡) and 𝑓𝑓(𝑡𝑡)𝑈𝑈(𝑡𝑡 − 𝑎𝑎).

y y
y = f(t)

t a t

y = f(t)U(t-a)

Figure 3.4. Graphs of 𝑓𝑓(𝑡𝑡) and 𝑓𝑓(𝑡𝑡)𝑈𝑈(𝑡𝑡 − 𝑎𝑎)

Another function that is very useful in engineering is called Dirac’s delta function,
𝛿𝛿(𝑡𝑡 − 𝑎𝑎), named after the British physicist, Paul Adrien Maurice Dirac (1902 – 1984). It
is also known as the impulse function. Mathematically speaking, Dirac’s delta function
is not really a function in the formal sense. It satisfies the following two properties:
∞
∞ for 𝑡𝑡 = 𝑎𝑎
𝛿𝛿(𝑡𝑡 − 𝑎𝑎) = � and � 𝛿𝛿(𝑡𝑡 − 𝑎𝑎)𝑑𝑑𝑑𝑑 = 1.
0 for 𝑡𝑡 ≠ 𝑎𝑎 −∞

The delta function can be considered as the derivative of the unit-step function. The
Laplace transform of the unit-step and delta functions is as follows.

𝑓𝑓(𝑡𝑡) ℒ[𝑓𝑓(𝑡𝑡)]
𝑒𝑒 −𝑎𝑎𝑎𝑎
𝑈𝑈(𝑡𝑡 − 𝑎𝑎) for 𝑎𝑎 ≥ 0
𝑠𝑠
𝛿𝛿(𝑡𝑡 − 𝑎𝑎) for 𝑎𝑎 ≥ 0 𝑒𝑒 −𝑎𝑎𝑎𝑎

ℒ −1 [𝐹𝐹(𝑠𝑠)] 𝐹𝐹(𝑠𝑠)
Table 3.2 Laplace transforms of unit-step and delta functions
Laplace Transforms 55

The formulas in Table 3.2 can be written as follows.

𝑒𝑒 −𝑎𝑎𝑎𝑎 𝑒𝑒 −𝑎𝑎𝑎𝑎
ℒ[𝑈𝑈(𝑡𝑡 − 𝑎𝑎)] = or ℒ −1 � � = 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)
𝑠𝑠 𝑠𝑠

ℒ[𝛿𝛿(𝑡𝑡 − 𝑎𝑎)] = 𝑒𝑒 −𝑎𝑎𝑎𝑎 or ℒ −1 [𝑒𝑒 −𝑎𝑎𝑎𝑎 ] = 𝛿𝛿(𝑡𝑡 − 𝑎𝑎)

Example 3.1.5
𝑒𝑒 −2𝑠𝑠 𝑒𝑒 −𝑠𝑠
(a) ℒ[𝑈𝑈(𝑡𝑡 − 2)] = (c) ℒ −1
� � = 𝑈𝑈(𝑡𝑡 − 1)
𝑠𝑠 𝑠𝑠
(b) ℒ[𝛿𝛿(𝑡𝑡 − 2)] = 𝑒𝑒 −2𝑠𝑠 (d) ℒ −1 [𝑒𝑒 −𝑠𝑠 ] = 𝛿𝛿(𝑡𝑡 − 1)

We can construct a step function by adding or subtracting unit-step functions. Example

3.1.6 shows how this is done.

Example 3.1.6
The graph of 𝑓𝑓(𝑡𝑡) is given below. Express 𝑓𝑓(𝑡𝑡) in terms of unit-step functions, and
evaluate ℒ[𝑓𝑓(𝑡𝑡)].
y

1
t
1 2 3 4 5 6 7 8

Solution
𝑓𝑓(𝑡𝑡) = 2𝑈𝑈(𝑡𝑡 − 1) − 𝑈𝑈(𝑡𝑡 − 5)
2𝑒𝑒 −𝑠𝑠 𝑒𝑒 −5𝑠𝑠
ℒ[𝑓𝑓(𝑡𝑡)] = 2ℒ[𝑈𝑈(𝑡𝑡 − 1)] − ℒ[𝑈𝑈(𝑡𝑡 − 5)] = −
𝑠𝑠 𝑠𝑠
Laplace Transforms 56

Laplace transforms and inverse Laplace transforms take on the linear properties of the
integral.

Theorem 3.1.7 (Linear Properties of Laplace Transform)

ℒ[𝑓𝑓(𝑡𝑡) + 𝑔𝑔(𝑡𝑡)] = ℒ[𝑓𝑓(𝑡𝑡)] + ℒ[𝑔𝑔(𝑡𝑡)]

ℒ[𝑘𝑘𝑘𝑘(𝑡𝑡)] = 𝑘𝑘 ℒ[𝑓𝑓(𝑡𝑡)] for any constant 𝑘𝑘

Example 3.1.8
2 3 2 3
(a) ℒ[2𝑒𝑒 −𝑡𝑡 + sin 3𝑡𝑡] = 2ℒ[𝑒𝑒 −𝑡𝑡 ] + ℒ[sin 3𝑡𝑡] = + 2 2
= + 2
𝑠𝑠 − (−1) 𝑠𝑠 + 3 𝑠𝑠 + 1 𝑠𝑠 + 9
1 1
(b) ℒ[1 + 𝛿𝛿(𝑡𝑡)] = ℒ[𝑒𝑒 0 ] + ℒ[𝛿𝛿(𝑡𝑡)] = + 𝑒𝑒 −0𝑠𝑠 = + 1
𝑠𝑠 − 0 𝑠𝑠
1 1
(c) ℒ −1 �𝑒𝑒 −𝑠𝑠 + 1 + � = ℒ −1 [𝑒𝑒 −𝑠𝑠 ] + ℒ −1 [𝑒𝑒 −0𝑠𝑠 ] + ℒ −1 � � = 𝛿𝛿(𝑡𝑡 − 1) + 𝛿𝛿(𝑡𝑡) + 1
𝑠𝑠 𝑠𝑠
s+3 𝑠𝑠 3
(d) ℒ −1 � 2
� = ℒ −1 � 2 2
� + ℒ −1 � 2 � = cos 3𝑡𝑡 + sin 3𝑡𝑡
𝑠𝑠 + 9 𝑠𝑠 + 3 𝑠𝑠 + 32

Given two functions 𝑓𝑓(𝑡𝑡) and 𝑔𝑔(𝑡𝑡), the convolution (𝑓𝑓 ∗ 𝑔𝑔)(𝑡𝑡) is defined by
𝑡𝑡
(𝑓𝑓 ∗ 𝑔𝑔)(𝑡𝑡) = � 𝑓𝑓(𝑢𝑢)𝑔𝑔(𝑡𝑡 − 𝑢𝑢)𝑑𝑑𝑑𝑑
0

The Laplace transform of the convolution of 𝑓𝑓 and 𝑔𝑔 is the product of the Laplace
transform of 𝑓𝑓 and the Laplace transform of 𝑔𝑔, that is,
ℒ[(𝑓𝑓 ∗ 𝑔𝑔)(𝑡𝑡)] = ℒ[𝑓𝑓(𝑡𝑡)] ⋅ ℒ[𝑔𝑔(𝑡𝑡)].
In other words, ℒ[𝑓𝑓 ∗ 𝑔𝑔] = 𝐹𝐹(𝑠𝑠)𝐺𝐺(𝑠𝑠).
Laplace Transforms 57

Example 3.1.9
1! 1 1
(a) ℒ[𝑡𝑡 ∗ sin 𝑡𝑡] = ℒ[𝑡𝑡] ⋅ ℒ[sin 𝑡𝑡] = ⋅ = 2 2
𝑠𝑠1+1 𝑠𝑠 2 +12 𝑠𝑠 (𝑠𝑠 + 1)
1 1 1
(b) ℒ[𝑡𝑡 ∗ 𝑒𝑒 𝑡𝑡 ] = ℒ[𝑡𝑡] ⋅ ℒ[𝑒𝑒 𝑡𝑡 ] = ⋅ = 2
𝑠𝑠 𝑠𝑠 − 1 𝑠𝑠 (𝑠𝑠 − 1)
2

1 𝑠𝑠 𝑠𝑠
(c) ℒ[𝑒𝑒 𝑡𝑡 ∗ cos 𝑡𝑡] = ℒ[𝑒𝑒 𝑡𝑡 ] ⋅ ℒ[cos 𝑡𝑡] = ⋅ 2 =
𝑠𝑠 − 1 𝑠𝑠 + 1 (𝑠𝑠 − 1)(𝑠𝑠 2 + 1)

Note that the symbol ∗ is used for convolution, not multiplication. Hence, the
expressions 𝑡𝑡 ∗ sin 𝑡𝑡 and 𝑡𝑡 sin 𝑡𝑡 are completely different. To find the Laplace transform
of 𝑡𝑡 sin 𝑡𝑡, or in general, 𝑡𝑡 𝑓𝑓(𝑡𝑡), we apply the formula
𝑑𝑑
ℒ[𝑡𝑡 𝑓𝑓(𝑡𝑡)] = − (ℒ[𝑓𝑓(𝑡𝑡)]).
𝑑𝑑𝑑𝑑
Alternatively, we write ℒ[𝑡𝑡 𝑓𝑓(𝑡𝑡)] = −𝐹𝐹 ′ (𝑠𝑠).

Example 3.1.10
Evaluate ℒ[𝑡𝑡 sin 𝑡𝑡].

Solution
1
Since ℒ[sin 𝑡𝑡] = ,
𝑠𝑠 2 + 1

𝑑𝑑 1 𝑑𝑑
ℒ[𝑡𝑡 sin 𝑡𝑡] =− � 2 � = − ((𝑠𝑠 2 + 1)−1 )
𝑑𝑑𝑑𝑑 𝑠𝑠 + 1 𝑑𝑑𝑑𝑑

= (𝑠𝑠 2 + 1)−2 (2𝑠𝑠)

2𝑠𝑠
=
(𝑠𝑠 2 + 1)2
Laplace Transforms 58

3.2 Laplace Transforms of Products of 𝒆𝒆𝒂𝒂𝒂𝒂

The following theorem considers the Laplace transform of a product involving
exponential function 𝑒𝑒 𝑎𝑎𝑎𝑎 . It is also known as the 𝑠𝑠-shift theorem or First Shift Theorem.

Theorem 3.2.1
If 𝐹𝐹(𝑠𝑠) = ℒ[𝑓𝑓(𝑡𝑡)], then
ℒ[𝑒𝑒 𝑎𝑎𝑎𝑎 𝑓𝑓(𝑡𝑡)] = 𝐹𝐹(𝑠𝑠 − 𝑎𝑎).

Proof
From Definition 3.1.1, we have
𝑅𝑅
ℒ[𝑒𝑒 𝑎𝑎𝑎𝑎 𝑓𝑓(𝑡𝑡)] = lim � 𝑒𝑒 −𝑠𝑠𝑠𝑠 ⋅ 𝑒𝑒 𝑎𝑎𝑎𝑎 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑
𝑅𝑅→∞ 0

𝑅𝑅
= lim � 𝑒𝑒 −(𝑠𝑠−𝑎𝑎)𝑡𝑡 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑
𝑅𝑅→∞ 0

= 𝐹𝐹(𝑠𝑠 − 𝑎𝑎).

Applying Theorem 3.2.1 for Laplace Transform

Step 1. Identify 𝑎𝑎 and 𝑓𝑓(𝑡𝑡).

Step 2. Evaluate 𝐹𝐹(𝑠𝑠) = ℒ[𝑓𝑓(𝑡𝑡)].

Step 3. Conclude ℒ[𝑒𝑒 𝑎𝑎𝑎𝑎 𝑓𝑓(𝑡𝑡)] = 𝐹𝐹(𝑠𝑠 − 𝑎𝑎).

Laplace Transforms 59

Example 3.2.2
Evaluate ℒ[𝑒𝑒 𝑡𝑡 sin 𝑡𝑡].

Solution
Step 1. 𝑎𝑎 = 1 and 𝑓𝑓(𝑡𝑡) = sin 𝑡𝑡
1
Step 2. 𝐹𝐹(𝑠𝑠) = ℒ[sin 𝑡𝑡] =
𝑠𝑠 2 +1
1
Step 3. ℒ[𝑒𝑒 𝑡𝑡 sin 𝑡𝑡] = 𝐹𝐹(𝑠𝑠 − 1) =
(𝑠𝑠 − 1)2 + 1

Example 3.2.3
Evaluate ℒ[𝑡𝑡 2 𝑒𝑒 −𝑡𝑡 ].

Solution
Step 1. 𝑎𝑎 = −1 and 𝑓𝑓(𝑡𝑡) = 𝑡𝑡 2
2! 2
Step 2. 𝐹𝐹(𝑠𝑠) = ℒ[𝑡𝑡 2 ] = =
𝑠𝑠 2+1 𝑠𝑠 3
2
Step 3. ℒ[𝑡𝑡 2 𝑒𝑒 −𝑡𝑡 ] = 𝐹𝐹(𝑠𝑠 + 1) =
(𝑠𝑠 + 1)3

Example 3.2.4
Evaluate ℒ[𝑒𝑒 2𝑡𝑡 𝛿𝛿(𝑡𝑡 − 1)].

Solution
Step 1. 𝑎𝑎 = 2 and 𝑓𝑓(𝑡𝑡) = 𝛿𝛿(𝑡𝑡 − 1)
Step 2. 𝐹𝐹(𝑠𝑠) = ℒ[𝛿𝛿(𝑡𝑡 − 1)] = 𝑒𝑒 −𝑠𝑠
Step 3. ℒ[𝑒𝑒 2𝑡𝑡 𝛿𝛿(𝑡𝑡 − 1)] = 𝐹𝐹(𝑠𝑠 − 2) = 𝑒𝑒 −(𝑠𝑠−2)
Laplace Transforms 60

Theorem 3.2.1 for inverse Laplace transform is

ℒ −1 [𝐹𝐹(𝑠𝑠 − 𝑎𝑎)] = 𝑒𝑒 𝑎𝑎𝑎𝑎 𝑓𝑓(𝑡𝑡).

Applying Theorem 3.2.1 for Inverse Laplace Transform

Step 1. Identify 𝑎𝑎 and find 𝐹𝐹(𝑠𝑠) from 𝐹𝐹(𝑠𝑠 − 𝑎𝑎).

Step 2. Evaluate 𝑓𝑓(𝑡𝑡) = ℒ −1 [𝐹𝐹(𝑠𝑠)].

Step 3. Conclude ℒ −1 [𝐹𝐹(𝑠𝑠 − 𝑎𝑎)] = 𝑒𝑒 𝑎𝑎𝑎𝑎 𝑓𝑓(𝑡𝑡).

Remark. To identify 𝑎𝑎, we observe the factor (𝑠𝑠 − 𝑎𝑎) in the denominator, e.g. (𝑠𝑠 − 1)2 ,
(𝑠𝑠 + 2), (𝑠𝑠 − 2)3 , etc.

Example 3.2.5
1
Evaluate ℒ −1 � �.
(𝑠𝑠 − 1)2 + 1

Solution
1 1
Step 1. 𝑎𝑎 = 1 and 𝐹𝐹(𝑠𝑠 − 1) = . So, 𝐹𝐹(𝑠𝑠) =
(𝑠𝑠 − 1)2 + 1 𝑠𝑠 2 + 1
1
Step 2. 𝑓𝑓(𝑡𝑡) = ℒ −1 � � = sin 𝑡𝑡
𝑠𝑠 2 +1
1
Step 3. ℒ −1 � � = 𝑒𝑒𝑡𝑡 sin 𝑡𝑡
(𝑠𝑠 − 1)2 + 1
Laplace Transforms 61

Example 3.2.6
2
Evaluate ℒ −1 � �.
(𝑠𝑠 + 1)3

Solution
2 2
Step 1. 𝑎𝑎 = −1 and 𝐹𝐹(𝑠𝑠 + 1) = . So, 𝐹𝐹(𝑠𝑠) =
(𝑠𝑠 + 1)3 𝑠𝑠 3
2 2!
Step 2. 𝑓𝑓(𝑡𝑡) = ℒ −1 � 3 � = ℒ −1 � 2+1 � = 𝑡𝑡 2
𝑠𝑠 𝑠𝑠
2
Step 3. ℒ −1 � � = 𝑒𝑒−𝑡𝑡 𝑡𝑡2
(𝑠𝑠 + 1)3

Example 3.2.7
Evaluate ℒ −1 [𝑒𝑒 2−𝑠𝑠 ].

Solution
Step 1. 𝑒𝑒 2−𝑠𝑠 = 𝑒𝑒 −(𝑠𝑠−2) ⇒ 𝑎𝑎 = 2 and 𝐹𝐹(𝑠𝑠 − 2) = 𝑒𝑒 −(𝑠𝑠−2) . So, 𝐹𝐹(𝑠𝑠) = 𝑒𝑒 −𝑠𝑠
Step 2. 𝑓𝑓(𝑡𝑡) = ℒ −1 [𝑒𝑒 −𝑠𝑠 ] = 𝛿𝛿(𝑡𝑡 − 1)
Step 3. ℒ −1 [𝑒𝑒 2−𝑠𝑠 ] = 𝑒𝑒 2𝑡𝑡 𝛿𝛿(𝑡𝑡 − 1)

An alternative answer for ℒ −1 [𝑒𝑒 2−𝑠𝑠 ] is

ℒ −1 [𝑒𝑒 2−𝑠𝑠 ] = 𝑒𝑒 2 ℒ −1 [𝑒𝑒 −𝑠𝑠 ] = 𝑒𝑒 2 𝛿𝛿(𝑡𝑡 − 1).
The functions 𝑒𝑒 2𝑡𝑡 𝛿𝛿(𝑡𝑡 − 1) and 𝑒𝑒 2 𝛿𝛿(𝑡𝑡 − 1) are equivalent because of the property of
the delta function.
Laplace Transforms 62

3.3 Laplace Transform of a Truncated Shift of a Function

For a given function 𝑓𝑓(𝑡𝑡) and positive constant 𝑎𝑎, the truncated shift of 𝑓𝑓 by 𝑎𝑎 units is
given by
𝑓𝑓(𝑡𝑡 − 𝑎𝑎) 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)
Figure 3.5 below compares the graphs of 𝑓𝑓(𝑡𝑡), 𝑓𝑓(𝑡𝑡 − 𝑎𝑎), and 𝑓𝑓(𝑡𝑡 − 𝑎𝑎)𝑈𝑈(𝑡𝑡 − 𝑎𝑎).

y = f(t)
t

Original signal 𝑦𝑦 = 𝑓𝑓(𝑡𝑡)

y = f(t-a)
t

𝑦𝑦 = 𝑓𝑓(𝑡𝑡 − 𝑎𝑎) is the shift of 𝑦𝑦 = 𝑓𝑓(𝑡𝑡) by a units to the right

y = f(t-a) U(t-a)
𝑦𝑦 = 𝑓𝑓(𝑡𝑡 − 𝑎𝑎)𝑈𝑈(𝑡𝑡 − 𝑎𝑎) is the truncated shift of 𝑦𝑦 = 𝑓𝑓(𝑡𝑡)

Figure 3.5. Graphs of 𝑓𝑓(𝑡𝑡), 𝑓𝑓(𝑡𝑡 − 𝑎𝑎), and 𝑓𝑓(𝑡𝑡 − 𝑎𝑎)𝑈𝑈(𝑡𝑡 − 𝑎𝑎)
Laplace Transforms 63

Theorem 3.3.1 considers the Laplace transform of a truncated shift of a function. It is

also known as the Second Shift Theorem or 𝑡𝑡-shift Theorem.

Theorem 3.3.1
If 𝐹𝐹(𝑠𝑠) = ℒ[𝑓𝑓(𝑡𝑡)], then
ℒ[𝑓𝑓(𝑡𝑡 − 𝑎𝑎) 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)] = 𝑒𝑒−𝑎𝑎𝑎𝑎 𝐹𝐹(𝑠𝑠)

Proof
From Definition 3.1.1, we have
𝑅𝑅
ℒ[𝑓𝑓(𝑡𝑡 − 𝑎𝑎) 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)] = lim � 𝑒𝑒 −𝑠𝑠𝑠𝑠 ⋅ 𝑓𝑓(𝑡𝑡 − 𝑎𝑎) 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)𝑑𝑑𝑑𝑑
𝑅𝑅→∞ 0

𝑅𝑅 𝑅𝑅
= lim � 𝑒𝑒 −𝑠𝑠𝑠𝑠 𝑓𝑓(𝑡𝑡 − 𝑎𝑎)𝑑𝑑𝑑𝑑 = lim � 𝑒𝑒 −𝑠𝑠(𝑢𝑢+𝑎𝑎) 𝑓𝑓(𝑢𝑢)𝑑𝑑𝑑𝑑
𝑅𝑅→∞ 𝑎𝑎 𝑅𝑅→∞ 0

𝑅𝑅
−𝑎𝑎𝑎𝑎
= 𝑒𝑒 lim � 𝑒𝑒 −𝑠𝑠𝑠𝑠 𝑓𝑓(𝑢𝑢)𝑑𝑑𝑑𝑑 = 𝑒𝑒 −𝑎𝑎𝑎𝑎 𝐹𝐹(𝑠𝑠)
𝑅𝑅→∞ 0

Applying Theorem 3.3.1 for Laplace Transform

Step 1. Identify 𝑎𝑎 and find 𝑓𝑓(𝑡𝑡) from 𝑓𝑓(𝑡𝑡 − 𝑎𝑎).

Step 2. Evaluate 𝐹𝐹(𝑠𝑠) = ℒ[𝑓𝑓(𝑡𝑡)].

Step 3. Conclude ℒ[𝑓𝑓(𝑡𝑡 − 𝑎𝑎) 𝑈𝑈(𝑡𝑡 − 𝑎𝑎)] = 𝑒𝑒 −𝑎𝑎𝑎𝑎 𝐹𝐹(𝑠𝑠).

Remark. To identify 𝑎𝑎, we observe the factor 𝑈𝑈(𝑡𝑡 − 𝑎𝑎).

Laplace Transforms 64

Example 3.3.2
Evaluate ℒ[𝑡𝑡 𝑈𝑈(𝑡𝑡 − 1)].

Solution
Step 1. 𝑎𝑎 = 1 and 𝑓𝑓(𝑡𝑡 − 1) = 𝑡𝑡. So, 𝑓𝑓(𝑡𝑡) = 𝑡𝑡 + 1
1 1
Step 2. 𝐹𝐹(𝑠𝑠) = ℒ[𝑡𝑡 + 1] = +
𝑠𝑠 2 𝑠𝑠
1 1
Step 3. ℒ[𝑡𝑡 𝑈𝑈(𝑡𝑡 − 1)] = 𝑒𝑒−𝑠𝑠 � 2 + �
𝑠𝑠 𝑠𝑠

Example 3.3.3
Evaluate ℒ[sin 𝑡𝑡 𝑈𝑈(𝑡𝑡 − 𝜋𝜋)].

Solution
Step 1. 𝑎𝑎 = 𝜋𝜋 and 𝑓𝑓(𝑡𝑡 − 𝜋𝜋) = sin 𝑡𝑡. So, 𝑓𝑓(𝑡𝑡) = sin(𝑡𝑡 + 𝜋𝜋) = − sin 𝑡𝑡
−1
Step 2. 𝐹𝐹(𝑠𝑠) = ℒ[− sin 𝑡𝑡] =
𝑠𝑠 2+1
−1
Step 3. ℒ[sin 𝑡𝑡 𝑈𝑈(𝑡𝑡 − 𝜋𝜋)] = 𝑒𝑒−𝜋𝜋𝜋𝜋 � 2 �
𝑠𝑠 + 1

Example 3.3.4
Evaluate ℒ[(𝑡𝑡 2 − 1) 𝑈𝑈(𝑡𝑡 − 1)].

Solution
Step 1. 𝑎𝑎 = 1 and 𝑓𝑓(𝑡𝑡 − 1) = 𝑡𝑡 2 − 1. So, 𝑓𝑓(𝑡𝑡) = (𝑡𝑡 + 1)2 − 1 = 𝑡𝑡 2 + 2𝑡𝑡
2 2
Step 2. 𝐹𝐹(𝑠𝑠) = ℒ[𝑡𝑡 2 + 2𝑡𝑡] = +
𝑠𝑠 3 𝑠𝑠 2
2 2
Step 3. ℒ[(𝑡𝑡 2 − 1) 𝑈𝑈(𝑡𝑡 − 1)] = 𝑒𝑒−𝑠𝑠 � + �
𝑠𝑠3 𝑠𝑠2
Laplace Transforms 65

Theorem 3.3.1 for inverse Laplace transform is

ℒ −1 [𝑒𝑒 −𝑎𝑎𝑎𝑎 𝐹𝐹(𝑠𝑠)] = 𝑓𝑓(𝑡𝑡 − 𝑎𝑎) 𝑈𝑈(𝑡𝑡 − 𝑎𝑎).

Applying Theorem 3.3.1 for Inverse Laplace Transform

Step 1. Identify 𝑎𝑎 and 𝐹𝐹(𝑠𝑠).

Step 2. Evaluate 𝑓𝑓(𝑡𝑡) = ℒ −1 [𝐹𝐹(𝑠𝑠)].

Step 3. Find 𝑓𝑓(𝑡𝑡 − 𝑎𝑎) and conclude ℒ −1 [𝑒𝑒 −𝑎𝑎𝑎𝑎 𝐹𝐹(𝑠𝑠)] = 𝑓𝑓(𝑡𝑡 − 𝑎𝑎)𝑈𝑈(𝑡𝑡 − 𝑎𝑎).

Remark. We observe the exponential factor 𝑒𝑒 −𝑎𝑎𝑎𝑎 to identify the value of 𝑎𝑎.

Example 3.3.5
1 1
Evaluate ℒ −1 �𝑒𝑒−𝑠𝑠 � + �� .
𝑠𝑠2 𝑠𝑠

Solution
1 1
Step 1. 𝑎𝑎 = 1 and 𝐹𝐹(𝑠𝑠) = +
𝑠𝑠 2 𝑠𝑠
1 1
Step 2. 𝑓𝑓(𝑡𝑡) = ℒ −1 � + � = 𝑡𝑡 + 1
𝑠𝑠 2 𝑠𝑠
Step 3. 𝑓𝑓(𝑡𝑡 − 1) = (𝑡𝑡 − 1) + 1 = 𝑡𝑡 and
1 1
ℒ −1 �𝑒𝑒 −𝑠𝑠 � 2 + �� = 𝑡𝑡 𝑈𝑈(𝑡𝑡 − 1)
𝑠𝑠 𝑠𝑠
Laplace Transforms 66

Example 3.3.6
−1
Evaluate ℒ −1 �𝑒𝑒−𝜋𝜋𝜋𝜋 � ��.
𝑠𝑠2+1

Solution
−1
Step 1. 𝑎𝑎 = 𝜋𝜋 and 𝐹𝐹(𝑠𝑠) =
𝑠𝑠 2+1
−1
Step 2. 𝑓𝑓(𝑡𝑡) = ℒ −1 � 2 � = − sin 𝑡𝑡
𝑠𝑠 + 1
Step 3. 𝑓𝑓(𝑡𝑡 − 𝜋𝜋) = − sin(𝑡𝑡 − 𝜋𝜋) = sin 𝑡𝑡 and
−1
ℒ −1 �𝑒𝑒 −𝜋𝜋𝜋𝜋 � 2 �� = sin 𝑡𝑡 𝑈𝑈(𝑡𝑡 − 𝜋𝜋)
𝑠𝑠 + 1

Example 3.3.7
2 2
Evaluate ℒ −1 �𝑒𝑒−𝑠𝑠 � 3 + 2 �� .
𝑠𝑠 𝑠𝑠

Solution
2 2
Step 1. 𝑎𝑎 = 1 and 𝐹𝐹(𝑠𝑠) = 3
+ 2
𝑠𝑠 𝑠𝑠
2 2 2! 1!
Step 2. 𝑓𝑓(𝑡𝑡) = ℒ −1 � 3 + 2 � = ℒ −1 � 2+1 � + 2ℒ −1 � 1+1 � = 𝑡𝑡 2 + 2𝑡𝑡
𝑠𝑠 𝑠𝑠 𝑠𝑠 𝑠𝑠
Step 3. 𝑓𝑓(𝑡𝑡 − 1) = (𝑡𝑡 − 1)2 + 2(𝑡𝑡 − 1) = 𝑡𝑡 2 − 1 and
2 2
ℒ −1 �𝑒𝑒 −𝑠𝑠 � 3
+ 2 �� = (𝑡𝑡 2 − 1) 𝑈𝑈(𝑡𝑡 − 1)
𝑠𝑠 𝑠𝑠
Laplace Transforms 67

3.4 Application
The Laplace transform and its inverse is useful for solving initial value problems (IVP)
involving linear differential equations. This algebraic technique provides an alternative
to those in Chapters 1 and 2. It is particularly useful when dealing with the unit-step
and delta functions. Figure 3.6 below illustrates the concept of the Laplace transform
(and its inverse) in solving IVPs.

ℒ ℒ −1
IVP algebraic form solution of IVP

Figure 3.6. The Laplace transform technique for solving IVPs

The Laplace transform technique allows us to find the particular solution of higher
order linear ODEs efficiently. To use this technique, we need the Laplace transform of
the first and second order derivatives of the function 𝑓𝑓.

𝑑𝑑𝑑𝑑
ℒ � � = 𝑠𝑠𝑠𝑠(𝑠𝑠) − 𝑓𝑓(0)
𝑑𝑑𝑑𝑑
𝑑𝑑 2 𝑓𝑓
ℒ� � = 𝑠𝑠 2 𝐹𝐹(𝑠𝑠) − 𝑠𝑠𝑠𝑠(0) − 𝑓𝑓 ′ (0)
𝑑𝑑𝑡𝑡 2

Procedure to Solve IVP using Laplace Transform

Step 1. Apply Laplace transforms on both sides of the differential equation.

Step 2. Substitute initial values 𝑓𝑓(0) and 𝑓𝑓′(0) and express 𝐹𝐹(𝑠𝑠) as the subject.

Step 3. Evaluate 𝑓𝑓(𝑡𝑡) = ℒ −1 [𝐹𝐹(𝑠𝑠)].

Compare the technique used in Example 3.4.1 below with that in Example 1.4.6 of
Chapter 1.
Laplace Transforms 68

Example 3.4.1
Solve the IVP:
𝑑𝑑𝑑𝑑
+ 𝑓𝑓 = 0
� 𝑑𝑑𝑑𝑑
𝑓𝑓(0) = 10

Solution

𝑑𝑑𝑑𝑑
ℒ� � + ℒ[𝑓𝑓] = ℒ[0] ⇒ �𝑠𝑠 𝐹𝐹(𝑠𝑠) − 𝑓𝑓(0)� + 𝐹𝐹(𝑠𝑠) = 0
𝑑𝑑𝑑𝑑
⇒ (𝑠𝑠 𝐹𝐹(𝑠𝑠) − 10) + 𝐹𝐹(𝑠𝑠) = 0

10
⇒ (𝑠𝑠 + 1) 𝐹𝐹(𝑠𝑠) = 10 ⇒ 𝐹𝐹(𝑠𝑠) =
𝑠𝑠 + 1

10
𝑓𝑓(𝑡𝑡) = ℒ −1 �𝑠𝑠+1� = 10 𝑒𝑒 −𝑡𝑡

In many cases, it is necessary to evaluate the inverse Laplace transform (in the last step
above) through Partial Fraction Decomposition (PFD), see Example 3.4.2 below.

Example 3.4.2
1
Use PFD to evaluate ℒ −1 � �.
𝑠𝑠(𝑠𝑠2+ 1)

Solution
1 𝐴𝐴 𝐵𝐵𝐵𝐵 + 𝐶𝐶
2
= + 2 ⇒ 1 = 𝐴𝐴(𝑠𝑠 2 + 1) + (𝐵𝐵𝐵𝐵 + 𝐶𝐶)𝑠𝑠
𝑠𝑠(𝑠𝑠 + 1) 𝑠𝑠 𝑠𝑠 + 1
Using coefficients comparison (or other method) we have 𝐴𝐴 = 1, 𝐵𝐵 = −1, 𝐶𝐶 = 0.
1 1 −𝑠𝑠 + 0 1 𝑠𝑠
ℒ −1 � � = ℒ −1
� + � = ℒ −1
� � − ℒ −1
� � = 1 − cos 𝑡𝑡
𝑠𝑠(𝑠𝑠 2 + 1) 𝑠𝑠 𝑠𝑠 2 + 1 𝑠𝑠 𝑠𝑠 2 + 1
Laplace Transforms 69

Example 3.4.3
Solve the IVP:
𝑑𝑑2 𝑓𝑓
⎧ 2 + 𝑓𝑓 = 1
⎪ 𝑑𝑑𝑡𝑡

⎨𝑓𝑓(0) = 0
⎪
⎩𝑓𝑓 ′ (0) = 0

Solution
𝑑𝑑 2 𝑓𝑓
ℒ � 2 � + ℒ[𝑓𝑓] = ℒ[1]
𝑑𝑑𝑡𝑡
1
⇒ �𝑠𝑠 2 𝐹𝐹(𝑠𝑠) − 𝑠𝑠𝑠𝑠(0) − 𝑓𝑓 ′ (0)� + 𝐹𝐹(𝑠𝑠) =
𝑠𝑠
1
⇒ (𝑠𝑠 2 𝐹𝐹(𝑠𝑠) − 0 − 0) + 𝐹𝐹(𝑠𝑠) =
𝑠𝑠
1
⇒ (𝑠𝑠 2 + 1)𝐹𝐹(𝑠𝑠) =
𝑠𝑠
1
⇒ 𝐹𝐹(𝑠𝑠) =
𝑠𝑠(𝑠𝑠 2 + 1)

From Example 3.4.2, we have

1
𝑓𝑓(𝑡𝑡) = ℒ −1 � � = 1 − cos 𝑡𝑡
𝑠𝑠(𝑠𝑠 2+ 1)
Laplace Transforms 70

Example 3.4.4
Solve the IVP:
𝑑𝑑2 𝑓𝑓
⎧ 2 = 2𝑈𝑈(𝑡𝑡 − 1)
⎪ 𝑑𝑑𝑡𝑡

⎨𝑓𝑓(0) = 0
⎪
⎩𝑓𝑓 ′ (0) = 0

Solution
𝑑𝑑 2 𝑓𝑓
ℒ � 2 � = 2ℒ[𝑈𝑈(𝑡𝑡 − 1)]
𝑑𝑑𝑡𝑡

2 ′ (0)
𝑒𝑒 −𝑠𝑠
⇒ 𝑠𝑠 𝐹𝐹(𝑠𝑠) − 𝑠𝑠𝑠𝑠(0) − 𝑓𝑓 =2
𝑠𝑠
𝑒𝑒 −𝑠𝑠
⇒ 𝑠𝑠 2 𝐹𝐹(𝑠𝑠) − 0 − 0 = 2
𝑠𝑠
𝑒𝑒 −𝑠𝑠 𝑒𝑒 −𝑠𝑠
⇒ 𝑠𝑠 2 𝐹𝐹(𝑠𝑠) = 2 ⇒ 𝐹𝐹(𝑠𝑠) = 2 3
𝑠𝑠 𝑠𝑠

2
𝑓𝑓(𝑡𝑡) = ℒ −1 �𝑒𝑒 −𝑠𝑠 � 3 �� = 𝑔𝑔(𝑡𝑡 − 1)𝑈𝑈(𝑡𝑡 − 1)
𝑠𝑠
2
𝐺𝐺(𝑠𝑠) = 3 ⇒ 𝑔𝑔(𝑡𝑡) = 𝑡𝑡 2 ⇒ 𝑔𝑔(𝑡𝑡 − 1) = (𝑡𝑡 − 1)2
𝑠𝑠
Thus, 𝑓𝑓(𝑡𝑡) = (𝑡𝑡 − 1)2 𝑈𝑈(𝑡𝑡 − 1).
Laplace Transforms 71

Example 3.4.5
A series circuit consists of an inductor of inductance 1H, a resistor of resistance 0.4 kΩ,
and a capacitor of capacitance 25 𝜇𝜇F. An impulse is applied onto the closed circuit so
that the charge 𝑞𝑞 (in coulomb) on the capacitor satisfies the differential equation
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑
2
+ 400 + 40000𝑞𝑞 = 100𝛿𝛿(𝑡𝑡 − 5)
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑
Assuming zero initial conditions, express 𝑞𝑞 in terms of 𝑡𝑡.

Solution
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑
ℒ � 2 � + 400ℒ � � + 40000ℒ[𝑞𝑞] = 100ℒ[𝛿𝛿(𝑡𝑡 − 5)]
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑
⇒ �𝑠𝑠 2 𝑄𝑄(𝑠𝑠) − 𝑠𝑠𝑠𝑠(0) − 𝑞𝑞 ′ (0)� + 400�𝑠𝑠𝑠𝑠(𝑠𝑠) − 𝑞𝑞(0)� + 40000 𝑄𝑄(𝑠𝑠) = 100𝑒𝑒 −5𝑠𝑠
⇒ 𝑠𝑠 2 𝑄𝑄(𝑠𝑠) + 400𝑠𝑠 𝑄𝑄(𝑠𝑠) + 40000 𝑄𝑄(𝑠𝑠) = 100𝑒𝑒 −5𝑠𝑠
⇒ (𝑠𝑠 2 + 400𝑠𝑠 + 40000) 𝑄𝑄(𝑠𝑠) = 100𝑒𝑒 −5𝑠𝑠
100 𝑒𝑒 −5s 100
⇒ 𝑄𝑄(𝑠𝑠) = 2 = 𝑒𝑒 −5𝑠𝑠 � �
𝑠𝑠 + 400𝑠𝑠 + 40000 (𝑠𝑠 + 200)2

100
𝑞𝑞(𝑡𝑡) = ℒ −1 �𝑒𝑒 −5𝑠𝑠 � �� = 𝑓𝑓(𝑡𝑡 − 5)𝑈𝑈(𝑡𝑡 − 5)
(𝑠𝑠 + 200)2
100 100
𝐹𝐹(𝑠𝑠) = ⇒ 𝑓𝑓(𝑡𝑡) = ℒ −1 � � = 𝑒𝑒 −200𝑡𝑡 𝑔𝑔(𝑡𝑡)
(𝑠𝑠 + 200) 2 (𝑠𝑠 + 200)2
100 100
𝐺𝐺(𝑠𝑠 + 200) = ⇒ 𝐺𝐺(𝑠𝑠) = 2 ⇒ 𝑔𝑔(𝑡𝑡) = 100𝑡𝑡
(𝑠𝑠 + 200) 2 𝑠𝑠
100
𝑓𝑓(𝑡𝑡) = ℒ −1 � � = 𝑒𝑒 −200𝑡𝑡 (100𝑡𝑡)
(𝑠𝑠 + 200)2
Thus, 𝑞𝑞(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 − 5)𝑈𝑈(𝑡𝑡 − 5) = 100 𝑒𝑒 −200(𝑡𝑡−5) (𝑡𝑡 − 5) 𝑈𝑈(𝑡𝑡 − 5).
Laplace Transforms 72

3.5 Exercises

Part A: Baseline Problems

1. Evaluate the following Laplace transforms (see Examples 3.1.2 and 3.1.3).
(a) ℒ[1 + 𝑡𝑡] (c) ℒ[2𝑡𝑡 2 + 𝑒𝑒 2𝑡𝑡 ]

(b) ℒ[𝑡𝑡 2 + 2𝑡𝑡 − 1] (d) ℒ[𝑒𝑒 −2𝑡𝑡 − 𝑒𝑒 3𝑡𝑡 ]

2. Evaluate the following inverse Laplace transforms (see Examples 3.1.2 and 3.1.3).
2 1 4 3 1
(a) ℒ −1 � 3 + 2 � (c) ℒ −1 � 5 + 4 − �
𝑠𝑠 𝑠𝑠 𝑠𝑠 𝑠𝑠 𝑠𝑠 + 5
1 1 2 1
(b) ℒ −1 � − � (d) ℒ −1 � − �
𝑠𝑠 𝑠𝑠 − 4 𝑠𝑠 + 1 𝑠𝑠 − 2

3. Evaluate the following Laplace transforms (see Example 3.1.4).

(a) ℒ[sin 3𝑡𝑡] (c) ℒ[cos 4𝑡𝑡 + 3 sin 2𝑡𝑡]

(b) ℒ[sin 𝑡𝑡 − cos 𝑡𝑡] (d) ℒ[sin 5𝑡𝑡 − sin 𝑡𝑡 + cos 2𝑡𝑡]

4. Evaluate the following inverse Laplace transforms (see Examples 3.1.3 and 3.1.4).
2 2 𝑠𝑠 + 3
(a) ℒ −1 � 2 + � (c) ℒ −1 � 2 �
𝑠𝑠 + 1 𝑠𝑠 + 1 𝑠𝑠 + 9
1 1 2 − 3𝑠𝑠
(b) ℒ −1 � + 2 � (d) ℒ −1 � �
𝑠𝑠 + 4 𝑠𝑠 + 4 𝑠𝑠 2 + 25
Laplace Transforms 73

5. Evaluate the following Laplace transforms (see Example 3.1.5).

(a) ℒ[𝑈𝑈(𝑡𝑡 − 1) − 𝑈𝑈(𝑡𝑡 − 3)] (c) ℒ[3𝛿𝛿(𝑡𝑡 − 2)]

(b) ℒ[2𝑈𝑈(𝑡𝑡 − 3)] (d) ℒ[5𝑈𝑈(𝑡𝑡 − 2) + 𝛿𝛿(𝑡𝑡 − 4)]

6. Evaluate the following inverse Laplace transforms (see Example 3.1.5).

𝑒𝑒 −3𝑠𝑠
(a) ℒ −1 [𝑒𝑒 −2𝑠𝑠 ] (c) ℒ −1 �2𝑒𝑒 −3𝑠𝑠 + 3 �
𝑠𝑠

𝑒𝑒 −2𝑠𝑠 2𝑒𝑒 −2𝑠𝑠

(b) ℒ −1
� � (d) ℒ −1
�𝑒𝑒 −𝑠𝑠
+ �
𝑠𝑠 𝑠𝑠

7. Evaluate the following Laplace transforms (see Example 3.1.9).

(a) ℒ[𝑡𝑡 ∗ sin 4𝑡𝑡]

(b) ℒ[𝑡𝑡 2 ∗ cos 2𝑡𝑡]

(c) ℒ[𝑡𝑡 ∗ 𝑒𝑒 −𝑡𝑡 ]

(d) ℒ[𝑒𝑒 2𝑡𝑡 ∗ 𝑈𝑈(𝑡𝑡 − 3)]

(e) ℒ[𝑡𝑡 ∗ 𝑒𝑒 2𝑡𝑡 ∗ sin 𝑡𝑡]

Laplace Transforms 74

Part B: Intermediate Problems

8. The graphs of 𝑓𝑓(𝑡𝑡) is given below. Express 𝑓𝑓(𝑡𝑡) in terms of unit-step functions,
and evaluate ℒ[𝑓𝑓(𝑡𝑡)] (see Example 3.1.6).
y y
3
2
2
(a) 1 (c)
1
t t
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8

y y

2 2

(b) 1 (d) 1
t t
1 2 3 4 5 6 7 8 -1 1 2 3 4 5 6 7

9. Evaluate the following Laplace transforms (see Examples 3.2.2 – 3.2.4).

(a) ℒ[𝑒𝑒 𝑡𝑡 sin 2𝑡𝑡] (c) ℒ[𝑡𝑡 𝑒𝑒 −2𝑡𝑡 ]

(b) ℒ[𝑒𝑒 −𝑡𝑡 cos 𝑡𝑡] (d) ℒ[𝑡𝑡 3 𝑒𝑒 3𝑡𝑡 ]

10. Evaluate the following inverse Laplace transforms (see Examples 3.2.5 – 3.2.7).
1 −1
𝑠𝑠
(a) ℒ −1 � � (c) ℒ � �
(𝑠𝑠 − 2)2 + 1 (𝑠𝑠 − 1)2 + 4
2 𝑠𝑠 − 2
(b) ℒ −1 � � (d) ℒ −1 � �
(𝑠𝑠 − 1)3 (𝑠𝑠 + 2)2 + 4
Laplace Transforms 75

11. Evaluate the following Laplace transforms (see Examples 3.3.2 – 3.3.4).
(a) ℒ[𝑡𝑡 2 𝑈𝑈(𝑡𝑡 − 2)] (c) ℒ[(2𝑡𝑡 + 1) 𝑈𝑈(𝑡𝑡 − 5)]
𝜋𝜋
(b) ℒ �sin 𝑡𝑡 𝑈𝑈 �𝑡𝑡 − �� (d) ℒ[(𝑡𝑡 3 − 1) 𝑈𝑈(𝑡𝑡 − 1)]
2

12. Evaluate the following inverse Laplace transforms (see Examples 3.3.5 – 3.3.7).
𝑒𝑒−𝑠𝑠 𝑠𝑠 + 3
(a) ℒ −1 � 3 � (c) ℒ −1 �𝑒𝑒−𝜋𝜋𝜋𝜋 � 2 ��
𝑠𝑠 𝑠𝑠 + 9
2 1 2𝑠𝑠 − 5
(b) ℒ −1 �𝑒𝑒−2𝑠𝑠 � 3 + 2 �� (d) ℒ −1 �𝑒𝑒−2𝜋𝜋𝜋𝜋 � 2 ��
𝑠𝑠 𝑠𝑠 𝑠𝑠 + 4

13. Evaluate the following Laplace transforms (see Example 3.1.10).

(a) ℒ[𝑡𝑡 sin 2𝑡𝑡] (c) ℒ[𝑡𝑡 cos 2𝑡𝑡]

(b) ℒ[𝑡𝑡 cos 𝑡𝑡] (d) ℒ[𝑡𝑡 2 sin 𝑡𝑡]

14. Evaluate the following Laplace transforms.

(a) ℒ[𝑡𝑡 𝑒𝑒 2𝑡𝑡 𝑈𝑈(𝑡𝑡 − 1)] (c) ℒ[𝑡𝑡 2 𝑒𝑒 −𝑡𝑡 𝑈𝑈(𝑡𝑡 − 2)]
𝜋𝜋 𝜋𝜋
(b) ℒ �𝑒𝑒 𝑡𝑡 sin 𝑡𝑡 𝑈𝑈 �𝑡𝑡 − 2 �� (d) ℒ �𝑒𝑒 −2𝑡𝑡 cos 2𝑡𝑡 𝑈𝑈 �𝑡𝑡 − 2 ��

15. Evaluate the following inverse Laplace transforms (see Example 3.4.2).
𝑠𝑠 − 1 3
(a) ℒ −1 � 2 � (c) ℒ −1 � 2 �
𝑠𝑠(𝑠𝑠 + 1) 𝑠𝑠 (𝑠𝑠 − 3)
𝑠𝑠 6
(b) ℒ −1 � � (d) ℒ −1 � �
𝑠𝑠 2 − 1 (𝑠𝑠 2 + 1)(𝑠𝑠 2 + 4)
Laplace Transforms 76

16. Evaluate the following inverse Laplace transforms.

𝑠𝑠 − 1 2
(a) ℒ −1 �𝑒𝑒 −4𝑠𝑠 � (c) ℒ −1 �𝑒𝑒 −2𝑠𝑠 2 �
2
𝑠𝑠(𝑠𝑠 + 1) 𝑠𝑠 (𝑠𝑠 − 3)
𝑠𝑠 6
(b) ℒ −1 �𝑒𝑒 −𝑠𝑠 � (d) ℒ −1 �𝑒𝑒 −3𝑠𝑠 �
𝑠𝑠 2 −1 (𝑠𝑠 2 + 1)(𝑠𝑠 2 + 4)

Part C: Application Problems

(See Section 3.4)

17. Solve the following IVPs using Laplace transform (see Examples 3.4.3 – 3.4.4).
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
= 𝑈𝑈(𝑡𝑡 − 1) = 𝑡𝑡 𝑈𝑈(𝑡𝑡 − 2)
(a) � 𝑑𝑑𝑑𝑑 (c) � 𝑑𝑑𝑑𝑑
𝑓𝑓(0) = 3 𝑓𝑓(0) = 4

𝑑𝑑 2 𝑓𝑓 𝑑𝑑2 𝑓𝑓
⎧ 2 + 𝑓𝑓 = 1 − 𝑈𝑈(𝑡𝑡 − 1) ⎧ 2 + 𝑓𝑓 = 𝛿𝛿(𝑡𝑡 − 1)
⎪ 𝑑𝑑𝑡𝑡 ⎪ 𝑑𝑑𝑡𝑡
(b) (d)
⎨𝑓𝑓(0) = 0 ⎨𝑓𝑓(0) = 0
⎪ ⎪
⎩𝑓𝑓 ′ (0) = 0 ⎩𝑓𝑓 ′ (0) = 2

18. A series circuit consists of an inductor of inductance 2H, a resistor of resistance

0.1 kΩ, and a capacitor of capacitance 0.8 mF. An impulse is applied onto the
closed circuit so that the charge 𝑞𝑞 (in coulomb) on the capacitor satisfies the
differential equation
𝑑𝑑 2 𝑞𝑞 𝑑𝑑𝑑𝑑
2
+ 50 + 625𝑞𝑞 = 110 𝛿𝛿(𝑡𝑡 − 2)
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑
Assuming zero initial conditions, express 𝑞𝑞 in terms of 𝑡𝑡.
Laplace Transforms 77

19. A series circuit consists of a capacitor of capacitance 10 mF and an inductor of

inductance 1H. A timer is applied onto the closed circuit so that the charge 𝑞𝑞 (in
coulomb) on the capacitor satisfies the differential equation
𝑑𝑑 2 𝑞𝑞
+ 100𝑞𝑞 = 12 𝑈𝑈(𝑡𝑡 − 𝜋𝜋)
𝑑𝑑𝑡𝑡 2
Assuming zero initial conditions, express 𝑞𝑞 in terms of 𝑡𝑡.

20. A spring of stiffness 10 N/m lies on a smooth horizontal surface with a smooth
ball of unit mass is attached to its end, while the other end is fixed to a wall. Let 𝑥𝑥
be the displacement of the ball from its equilibrium position. An impulse force
acts on the ball so that its harmonic motion satisfies the differential equation
𝑑𝑑 2 𝑥𝑥
+ 4𝑥𝑥 = 𝛿𝛿(𝑡𝑡 − 𝜋𝜋).
𝑑𝑑𝑡𝑡 2
Express the displacement 𝑥𝑥 as a function of 𝑡𝑡.

− END OF CHAPTER 3 −
Chapter 4
Fourier Series

Learning Objectives
By the end of this chapter, students should be able to
1. define periodic functions;
2. describe the symmetric properties of even and odd periodic functions;
3. find the Fourier series of a periodic function.

T he French mathematician Joseph Fourier (1768 – 1830) first used an

infinite series of sine and cosine functions to solve a long-standing
problem on heat transfer. This trigonometric series, later known as
Fourier series, was found to fill in the gap for approximating functions where
the Taylor (and Maclaurin) series fail. The Taylor and Maclaurin series
approximate function values very well near a point, while the Fourier series is
well-suited for approximating functions on wide intervals, particularly
functions with repeating cycles, i.e. periodic functions. This is useful because
periodic functions abound in science and engineering, such as heartbeat, tides,
vibrating strings, radio signals and alternating currents, etc.

We begin with Section 4.1 by defining the periodic function. The Fourier series
formula follows immediately in Section 4.2, where we work through a few
typical examples. Section 4.3 is devoted to the Fourier series of even/odd
functions; their Fourier series take simpler forms. A simple application of
Fourier series is presented in Section 4.4 and exercises are given in Section 4.5
for further practice.

78
Fourier Series 79

4.1 Periodic Functions

Definition 4.1.1 (Periodic Function)

A function 𝑓𝑓(𝑡𝑡) is a periodic function if there is a positive number 𝑇𝑇 such that
𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 𝑇𝑇)
for all 𝑡𝑡 in the domain of 𝑓𝑓. The period of 𝑓𝑓 is the smallest such 𝑇𝑇.

A periodic function with period 𝑇𝑇 is referred to as 𝑇𝑇-periodic function.

Example 4.1.2
Sketch the following functions and determine their periods:
(a) 𝑦𝑦 = sin 𝑡𝑡
(b) 𝑦𝑦 = cos 𝑡𝑡

Solution
(a) y = sin t y (b) y = cos t y
1 1

t t
-3π -2π -π π 2π 3π -3π -2π -π π 2π 3π

-1 -1

𝑦𝑦 = sin 𝑡𝑡 is 2𝜋𝜋-periodic 𝑦𝑦 = cos 𝑡𝑡 is 2𝜋𝜋-periodic

A function 𝑓𝑓(𝑡𝑡) defined on the interval −𝐿𝐿 < 𝑡𝑡 < 𝐿𝐿 can be extended to a 2𝐿𝐿-
periodic function by imposing the condition 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝐿𝐿) for all 𝑡𝑡, as
shown in Examples 4.1.3 and 4.1.4.
Fourier Series 80

Example 4.1.3
0 for −𝜋𝜋 < 𝑡𝑡 < 0
(a) Sketch the graph of 𝑓𝑓(𝑡𝑡) = � on the interval
1 for 0 < 𝑡𝑡 < 𝜋𝜋
−𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋.
(b) Hence, sketch the periodic extension 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋) for all 𝑡𝑡 on the
interval −3𝜋𝜋 < 𝑡𝑡 < 3𝜋𝜋.

Solution
(a) y

t
-3π -2π -π π 2π 3π

(b) y

t
-3π -2π -π π 2π 3π
Fourier Series 81

Example 4.1.4
Sketch three cycles of the graph of the following periodic functions and
determine their periods.
0 for −𝜋𝜋 < 𝑡𝑡 < 0
(a) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
𝑡𝑡 for 0 < 𝑡𝑡 < 𝜋𝜋
−1 for −1 < 𝑡𝑡 < 0
(b) 𝑔𝑔(𝑡𝑡) = � and 𝑔𝑔(𝑡𝑡) = 𝑔𝑔(𝑡𝑡 + 2)
1 for 0 < 𝑡𝑡 < 1

Solution
(a) y period = 2𝜋𝜋
π

y=t
π/2

y=0 t
-3π -2π -π π 2π 3π

(b) y period = 2
1
y=1

t
-3 -2 -1 1 2 3

y = −1
-1
Fourier Series 82

4.2 The Fourier Series

The Fourier series is an infinite series of sine and cosine functions, and is
suitable for approximating periodic functions.

Definition 4.2.1 (Fourier Series)

Let 𝑓𝑓(𝑡𝑡) be a 𝑇𝑇-periodic function. The Fourier series of 𝑓𝑓(𝑡𝑡) is an expansion of
𝑓𝑓(𝑡𝑡) in the form of the trigonometric series
∞
𝑎𝑎0 2𝑛𝑛𝑛𝑛𝑛𝑛 2𝑛𝑛𝑛𝑛𝑛𝑛
𝑓𝑓(𝑡𝑡) = + � �𝑎𝑎𝑛𝑛 cos � � + 𝑏𝑏𝑛𝑛 sin � ��
2 𝑇𝑇 𝑇𝑇
𝑛𝑛=1

where 𝑎𝑎0 , 𝑎𝑎𝑛𝑛 , and 𝑏𝑏𝑛𝑛 are the Fourier coefficients given by

2 𝑇𝑇/2
𝑎𝑎0 = � 𝑓𝑓(𝑡𝑡) 𝑑𝑑𝑑𝑑
𝑇𝑇 −𝑇𝑇/2

2 𝑇𝑇/2 2𝑛𝑛𝑛𝑛𝑛𝑛
𝑎𝑎𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) cos � � 𝑑𝑑𝑑𝑑 for 𝑛𝑛 = 1,2,3, …
𝑇𝑇 −𝑇𝑇/2 𝑇𝑇

2 𝑇𝑇/2 2𝑛𝑛𝑛𝑛𝑛𝑛
𝑏𝑏𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) sin � � 𝑑𝑑𝑑𝑑 for 𝑛𝑛 = 1,2,3, …
𝑇𝑇 −𝑇𝑇/2 𝑇𝑇

Remark. As an infinite series, the convergence of the Fourier series is a very

important concern. However, discussion on this convergence is beyond the
scope of this text as it involves very advanced mathematics. We shall assume
convergence of the Fourier series for all points except at discontinuous points.

These results are often used when computing Fourier series.

sin(𝑛𝑛𝑛𝑛) = 0 and cos(𝑛𝑛𝑛𝑛) = (−1)𝑛𝑛 for all integers 𝑛𝑛
Fourier Series 83

When calculating the Fourier series of a periodic function, the following

procedure is a good guide.

Procedure to Calculate the Fourier Series of a Periodic Function

Step 1. Identify the period 𝑇𝑇.
Step 2. Calculate the Fourier coefficients 𝑎𝑎0 , 𝑎𝑎𝑛𝑛 , 𝑏𝑏𝑛𝑛 .
Step 3. Write the Fourier series of 𝑓𝑓 in sigma notation (see Definition 4.2.1).
Step 4. Where possible, simplify the coefficients.

In Step 4 of the procedure above, the Fourier coefficients 𝑎𝑎𝑛𝑛 or 𝑏𝑏𝑛𝑛 can be
simplified if they contain either the factor of (1 − (−1)𝑛𝑛 ) or (1 + (−1)𝑛𝑛 ). This
is done by observing its alternating values as follows.

2 for 𝑛𝑛 = 2𝑘𝑘 − 1 i.e. 𝑛𝑛 is odd

1 − (−1)𝑛𝑛 = �
0 for 𝑛𝑛 = 2𝑘𝑘 i.e. 𝑛𝑛 is even

0 for 𝑛𝑛 = 2𝑘𝑘 − 1 i.e. 𝑛𝑛 is odd

1 + (−1)𝑛𝑛 = �
2 for 𝑛𝑛 = 2𝑘𝑘 i.e. 𝑛𝑛 is even

Example 4.2.2
∞ ∞
1 − (−1)𝑛𝑛 2
(a) �� � sin(𝑛𝑛𝑛𝑛𝑛𝑛) = � sin�(2𝑘𝑘 − 1)𝜋𝜋𝜋𝜋�
𝑛𝑛𝑛𝑛 (2𝑘𝑘 − 1)𝜋𝜋
𝑛𝑛=1 𝑘𝑘=1
∞ ∞
1 + (−1)𝑛𝑛 2
(b) �� � sin(𝑛𝑛𝑛𝑛𝑛𝑛) = � sin(2𝑘𝑘𝑘𝑘𝑘𝑘)
𝑛𝑛𝑛𝑛 2𝑘𝑘𝑘𝑘
𝑛𝑛=1 𝑘𝑘=1
Fourier Series 84

Example 4.2.3
Calculate the Fourier series of 𝑓𝑓(𝑡𝑡) given by
0 for −1 < 𝑡𝑡 < 0
𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2).
1 for 0 < 𝑡𝑡 < 1

Solution
Step 1. 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2) ⇒ 𝑇𝑇 = 2

1 1
2
Step 2. 𝑎𝑎0 = � 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑 = 0 + � 1 𝑑𝑑𝑑𝑑 = 1 .
2
−1 0

1 1
2 2𝑛𝑛𝑛𝑛𝑛𝑛
𝑎𝑎𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) cos � � 𝑑𝑑𝑑𝑑 = 0 + � 1 ∙ cos(𝑛𝑛𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑
2 2
−1 0
1
sin(𝑛𝑛𝑛𝑛𝑛𝑛) sin(𝑛𝑛𝑛𝑛) sin(0)
=� � = − = 0.
𝑛𝑛𝑛𝑛 0
𝑛𝑛𝑛𝑛 𝑛𝑛𝑛𝑛

1 1
2 2𝑛𝑛𝑛𝑛𝑛𝑛
𝑏𝑏𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) sin � � 𝑑𝑑𝑑𝑑 = 0 + � 1 ∙ sin(𝑛𝑛𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑
2 2
−1 0
1
− cos(𝑛𝑛𝑛𝑛𝑛𝑛) cos(𝑛𝑛𝑛𝑛) cos(0) 1 − (−1)𝑛𝑛
=� � =− + = .
𝑛𝑛𝑛𝑛 0
𝑛𝑛𝑛𝑛 𝑛𝑛𝑛𝑛 𝑛𝑛𝑛𝑛

∞
1 1 − (−1)𝑛𝑛
Step 3. The Fourier series of 𝑓𝑓(𝑡𝑡) is 𝑓𝑓(𝑡𝑡) = + � � � sin(𝑛𝑛𝑛𝑛𝑛𝑛) .
2 𝑛𝑛𝑛𝑛
𝑛𝑛=1

∞
1 2
Step 4. 𝑓𝑓(𝑡𝑡) = + � sin�(2𝑘𝑘 − 1)𝜋𝜋𝜋𝜋� , cf. Example 4.2.2(a).
2 (2𝑘𝑘 − 1)𝜋𝜋
𝑘𝑘=1
Fourier Series 85

Let us consider the Fourier series of the function 𝑓𝑓(𝑡𝑡) in Example 4.2.3.
Denoting the 𝑛𝑛-th partial sum of the Fourier series as
𝑛𝑛
1 2
𝑆𝑆𝑛𝑛 (𝑡𝑡) = + � sin�(2𝑘𝑘 − 1)𝜋𝜋𝜋𝜋�,
2 (2𝑘𝑘 − 1)𝜋𝜋
𝑘𝑘=1

we can see graphically how 𝑆𝑆𝑛𝑛 (𝑡𝑡) approximates 𝑓𝑓(𝑡𝑡) as 𝑛𝑛 gets larger, as shown
in Figure 4.1 below.
y y

1 1

0 t 0 t
-4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 2 3 4

(a) 𝑓𝑓(𝑡𝑡) vs. 𝑆𝑆1 (𝑡𝑡) (b) 𝑓𝑓(𝑡𝑡) vs. 𝑆𝑆2 (𝑡𝑡)
y y

1 1

0 t t
0
-4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 2 3 4

(c) 𝑓𝑓(𝑡𝑡) vs. 𝑆𝑆3 (𝑡𝑡) (d) 𝑓𝑓(𝑡𝑡) vs. 𝑆𝑆6 (𝑡𝑡)

Figure 4.1. Graphs of 𝑓𝑓(𝑡𝑡) versus 𝑆𝑆𝑛𝑛 (𝑡𝑡)

We observe from the graphs in Figure 4.1 above that as 𝑛𝑛 gets larger, the
graphs of 𝑆𝑆𝑛𝑛 (𝑡𝑡) resemble that of 𝑓𝑓(𝑡𝑡) more and more.
Fourier Series 86

Example 4.2.4
Calculate the Fourier series of 𝑓𝑓(𝑡𝑡) given by
0 for −𝜋𝜋 < 𝑡𝑡 < 0
𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋).
𝑡𝑡 for 0 < 𝑡𝑡 < 𝜋𝜋

Solution
Step 1. 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋) ⇒ 𝑇𝑇 = 2𝜋𝜋.

𝜋𝜋 𝝅𝝅 𝜋𝜋
2 1 1 𝑡𝑡 2 𝜋𝜋
Step 2. 𝑎𝑎0 = 2𝜋𝜋 � 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑 = 𝜋𝜋 �0 + � 𝑡𝑡 𝑑𝑑𝑑𝑑� = 𝜋𝜋 � 2 � = 2 .
−𝜋𝜋 0 0

𝜋𝜋 𝜋𝜋
2 2𝑛𝑛𝑛𝑛𝑛𝑛 1
𝑎𝑎𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) cos � � 𝑑𝑑𝑑𝑑 = �0 + � 𝑡𝑡 ∙ cos(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑�
2𝜋𝜋 2𝜋𝜋 𝜋𝜋
−𝜋𝜋 0
𝜋𝜋
1 𝑡𝑡 sin(𝑛𝑛𝑛𝑛) cos(𝑛𝑛𝑛𝑛) 1 𝜋𝜋 sin(𝑛𝑛𝑛𝑛) cos(𝑛𝑛𝑛𝑛) 1 cos 0
= � + 2 � = � + 2 � − �0 + 2 �
𝜋𝜋 𝑛𝑛 𝑛𝑛 0
𝜋𝜋 𝑛𝑛 𝑛𝑛 𝜋𝜋 𝑛𝑛
1 (−1)𝑛𝑛 1 (−1)𝑛𝑛 − 1
= �0 + − 2� = .
𝜋𝜋 𝑛𝑛2 𝑛𝑛 𝜋𝜋𝑛𝑛2

−(−1)𝑛𝑛
Similarly, 𝑏𝑏𝑛𝑛 = .
𝑛𝑛

Step 3. The Fourier series of 𝑓𝑓(𝑡𝑡) is

∞
𝜋𝜋 (−1)𝑛𝑛 − 1 (−1)𝑛𝑛
𝑓𝑓(𝑡𝑡) = + � �� � cos(𝑛𝑛𝑛𝑛) − sin(𝑛𝑛𝑛𝑛)�
4 𝜋𝜋𝑛𝑛2 𝑛𝑛
𝑛𝑛=1

∞ ∞
𝜋𝜋 2 cos(2𝑘𝑘 − 1)𝑡𝑡 (−1)𝑛𝑛
Step 4. 𝑓𝑓(𝑡𝑡) = − � − � sin(𝑛𝑛𝑛𝑛).
4 𝜋𝜋 (2𝑘𝑘 − 1)2 𝑛𝑛
𝑘𝑘=1 𝑛𝑛=1
Fourier Series 87

4.3 Fourier series of Even/Odd Functions

The computations of the Fourier series coefficients of even or odd functions
are simpler. Thus, an effort to determine whether a function is even or odd
helps make computations faster. We will first define what odd and even
functions are and then look at some of their properties.

Definition 4.3.1 (Even and Odd Functions)

A function 𝑓𝑓 is said to be even if 𝑓𝑓(−𝑡𝑡) = 𝑓𝑓(𝑡𝑡) for all 𝑡𝑡 in the domain of 𝑓𝑓.
A function 𝑓𝑓 is said to be odd if 𝑓𝑓(−𝑡𝑡) = −𝑓𝑓(𝑡𝑡) for all 𝑡𝑡 in the domain of 𝑓𝑓.

Remark. The graph of an even function is symmetrical about the vertical axis,
while the graph of an odd function is symmetrical about the origin. There are
also functions that are neither even nor odd.

Example 4.3.2
(a) The function ℎ(𝑡𝑡) given by
ℎ(𝑡𝑡) = |𝑡𝑡| for −1 < 𝑡𝑡 < 1 and ℎ(𝑡𝑡) = ℎ(𝑡𝑡 + 2)
is an even function. Its graph is symmetric about the vertical axis in
Figure 4.2.
y
1

0.5

t
-3 -2 -1 1 2 3
Figure 4.2. Graph of ℎ(𝑡𝑡)
Fourier Series 88

(b) The function 𝑦𝑦(𝑡𝑡) given by

𝑦𝑦(𝑡𝑡) = 𝑡𝑡 for −𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋 and 𝑦𝑦(𝑡𝑡) = 𝑦𝑦(𝑡𝑡 + 2𝜋𝜋)
is an odd function. Its graph is symmetric about the origin in Figure 4.3.
π y

t
-3π -2π -π π 2π 3π

-π
Figure 4.3. Graph of 𝑦𝑦(𝑡𝑡)

(c) The function 𝑓𝑓(𝑡𝑡) defined in Example 4.1.4(a) is neither even nor odd.
Observe that its graph if not symmetrical about the vertical axis, neither
is it symmetrical about the origin.

(d) 𝑦𝑦 = sin 𝑡𝑡 is an odd function and 𝑦𝑦 = cos 𝑡𝑡 is an even function, cf.

Example 4.1.2. In general, 𝑦𝑦 = sin 𝑛𝑛𝑛𝑛 is an odd function and 𝑦𝑦 = cos 𝑛𝑛𝑛𝑛 is
an even function for any non-zero number 𝑛𝑛.

We will use graphical observation as in Example 4.3.2 to conclude whether a

given function is even, odd or neither. For our purpose, we do not require a
formal proof based on Definition 4.3.1.

The symmetries of even/odd functions give rise to the following properties.

Theorem 4.3.3 (Products of Even/Odd Functions)

(a) The product of two even functions is even.
(b) The product of two odd functions is even.
(c) The product of an even function and an odd function is odd.
Fourier Series 89

Theorem 4.4.3 is analogous to the product of positive/negative numbers, i.e.

EVEN function ∼ POSITIVE number, and ODD function ∼ NEGATIVE number.

Theorem 4.3.4 (Integral of Even/Odd Functions)

Let 𝐿𝐿 > 0.
𝐿𝐿 𝐿𝐿

(a) If 𝑔𝑔(𝑡𝑡) is an even function, then � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = 2 � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 .

−𝐿𝐿 0

𝐿𝐿

(b) If 𝑔𝑔(𝑡𝑡) is an odd function, then � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = 0 .

−𝐿𝐿

Proof
𝐿𝐿 0 𝐿𝐿 0 𝐿𝐿

(a) � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 + � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = − � 𝑔𝑔(−𝑢𝑢)𝑑𝑑𝑑𝑑 + � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑

−𝐿𝐿 −𝐿𝐿 0 𝐿𝐿 0
𝐿𝐿 𝐿𝐿 𝐿𝐿

= � 𝑔𝑔(𝑢𝑢)𝑑𝑑𝑑𝑑 + � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = 2 � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑

0 0 0

𝐿𝐿 0 𝐿𝐿 0 𝐿𝐿

(b) � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 + � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = − � 𝑔𝑔(−𝑢𝑢)𝑑𝑑𝑑𝑑 + � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑

−𝐿𝐿 −𝐿𝐿 0 𝐿𝐿 0
𝐿𝐿 𝐿𝐿

= − � 𝑔𝑔(𝑢𝑢)𝑑𝑑𝑑𝑑 + � 𝑔𝑔(𝑡𝑡)𝑑𝑑𝑑𝑑 = 0
0 0
Fourier Series 90

Example 4.3.5
𝜋𝜋

(a) 𝑦𝑦 = sin 𝑡𝑡 cos 𝑡𝑡 is an odd function, and � sin 𝑡𝑡 cos 𝑡𝑡 𝑑𝑑𝑑𝑑 = 0 .

−𝜋𝜋

𝜋𝜋 𝜋𝜋

(b) 𝑦𝑦 = sin 𝑡𝑡 sin 𝑛𝑛𝑛𝑛 is an even function, and � sin 𝑡𝑡 cos 𝑛𝑛𝑛𝑛 𝑑𝑑𝑑𝑑 = 2 � sin 𝑡𝑡 cos 𝑛𝑛𝑛𝑛 𝑑𝑑𝑑𝑑 .
−𝜋𝜋 0

𝜋𝜋

(c) 𝑦𝑦 = 𝑡𝑡 cos 𝑛𝑛𝑛𝑛 is an odd function, and � 𝑡𝑡 cos 𝑛𝑛𝑛𝑛 𝑑𝑑𝑑𝑑 = 0 .

−𝜋𝜋

𝜋𝜋 𝜋𝜋

(d) 𝑦𝑦 = 𝑡𝑡 2 cos 𝑛𝑛𝑛𝑛 is an even function, and � 𝑡𝑡 2 cos 𝑛𝑛𝑛𝑛 𝑑𝑑𝑑𝑑 = 2 � 𝑡𝑡 2 cos 𝑛𝑛𝑛𝑛 𝑑𝑑𝑑𝑑 .
−𝜋𝜋 0

The following is a result from Theorem 4.3.4.

Corollary 4.3.6 (Fourier Coefficients of Even/Odd Functions)

Let 𝑓𝑓(𝑡𝑡) be a 𝑇𝑇-periodic function with Fourier coefficients 𝑎𝑎0 , 𝑎𝑎𝑛𝑛 , 𝑏𝑏𝑛𝑛 .
(a) If 𝑓𝑓(𝑡𝑡) is an even function, then 𝑏𝑏𝑛𝑛 = 0 for all positive integers 𝑛𝑛.
(b) If 𝑓𝑓(𝑡𝑡) is an odd function, then 𝑎𝑎0 = 𝑎𝑎𝑛𝑛 = 0 for all positive integers 𝑛𝑛.

Proof
From Theorem 4.3.4, it is sufficient to show that for each case, the
corresponding Fourier coefficient has odd function integrand.
2𝑛𝑛𝑛𝑛𝑛𝑛
(a) 𝑓𝑓(𝑡𝑡) is an even function ⇒ 𝑓𝑓(𝑡𝑡) sin � � is an odd function.
𝑇𝑇
2𝑛𝑛𝑛𝑛𝑛𝑛
(b) 𝑓𝑓(𝑡𝑡) is an odd function ⇒ 𝑓𝑓(𝑡𝑡) cos � � is an odd function.
𝑇𝑇
Fourier Series 91

Example 4.3.7
Calculate the Fourier series of 𝑓𝑓(𝑡𝑡) given by
−1 if −𝜋𝜋 < 𝑡𝑡 < 0
𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋).
1 if 0 < 𝑡𝑡 < 𝜋𝜋

Solution
Step 1. 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋) ⇒ 𝑇𝑇 = 2𝜋𝜋.

Step 2. The graph of 𝑓𝑓(𝑡𝑡) is symmetrical about the origin (cf. Example
4.1.4(b)), so 𝑓𝑓(𝑡𝑡) is an odd function, and hence, 𝑎𝑎0 = 𝑎𝑎𝑛𝑛 = 0 for all
positive integers of 𝑛𝑛.

𝜋𝜋
1
𝑏𝑏𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) sin(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑
𝜋𝜋
−𝜋𝜋
𝜋𝜋
2
= � 𝑓𝑓(𝑡𝑡) sin(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑 since 𝑓𝑓(𝑡𝑡) sin(𝑛𝑛𝑛𝑛) is an even function
𝜋𝜋
0
𝜋𝜋 𝜋𝜋
2 2 − cos(𝑛𝑛𝑛𝑛) 2 − cos(𝑛𝑛𝑛𝑛) + cos 0
= � 1 ⋅ sin(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑 = � � = � �
𝜋𝜋 𝜋𝜋 𝑛𝑛 0
𝜋𝜋 𝑛𝑛
0

2 1 − (−1)𝑛𝑛
= � � for all positive integers 𝑛𝑛.
𝜋𝜋 𝑛𝑛
∞
2 1 − (−1)𝑛𝑛
Step 3. The Fourier series of 𝑓𝑓(𝑡𝑡) is 𝑓𝑓(𝑡𝑡) = � � � sin(𝑛𝑛𝑛𝑛) .
𝜋𝜋 𝑛𝑛
𝑛𝑛=1

∞
4 sin�(2𝑘𝑘 − 1)𝑡𝑡�
Step 4. 𝑓𝑓(𝑡𝑡) = �
𝜋𝜋 2𝑘𝑘 − 1
𝑘𝑘=1
Fourier Series 92

Example 4.3.8
Calculate the Fourier series of Sawtooth function 𝑓𝑓(𝑡𝑡) given by
𝑓𝑓(𝑡𝑡) = 𝑡𝑡 for −𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋 and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋).

Solution
Step 1. 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋) ⇒ 𝑇𝑇 = 2𝜋𝜋.

Step 2. The graph of 𝑓𝑓(𝑡𝑡) is symmetrical about the origin (cf. Example
4.3.2(b)), so 𝑓𝑓(𝑡𝑡) is an odd function, and hence, 𝑎𝑎0 = 𝑎𝑎𝑛𝑛 = 0 for all
positive integers of 𝑛𝑛.

𝜋𝜋
1
𝑏𝑏𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) sin(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑
𝜋𝜋
−𝜋𝜋
𝜋𝜋
2
= � 𝑓𝑓(𝑡𝑡) sin(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑 since 𝑓𝑓(𝑡𝑡) sin(𝑛𝑛𝑛𝑛) is an even function
𝜋𝜋
0
𝜋𝜋 𝜋𝜋
2 2 𝑡𝑡 cos(𝑛𝑛𝑛𝑛) sin(𝑛𝑛𝑛𝑛)
= � 𝑡𝑡 sin(𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑 = �− + �
𝜋𝜋 𝜋𝜋 𝑛𝑛 𝑛𝑛2 0
0

2 𝜋𝜋 cos(𝑛𝑛𝑛𝑛) sin(𝑛𝑛𝑛𝑛) sin 0

= ��− + 2
� − �0 + 2 ��
𝜋𝜋 𝑛𝑛 𝑛𝑛 𝑛𝑛

2 𝜋𝜋(−1)𝑛𝑛 0 2(−1)𝑛𝑛+1
= ��− + �� = for all positive integers 𝑛𝑛.
𝜋𝜋 𝑛𝑛 𝑛𝑛 𝑛𝑛
∞
(−1)𝑛𝑛+1
Step 3. The Fourier series of 𝑓𝑓(𝑡𝑡) is 𝑓𝑓(𝑡𝑡) = 2 � sin(𝑛𝑛𝑛𝑛).
𝑛𝑛
𝑛𝑛=1
Fourier Series 93

Example 4.3.9
Calculate the Fourier series of 𝑓𝑓(𝑡𝑡) given by
𝑓𝑓(𝑡𝑡) = |𝑡𝑡| for −1 < 𝑡𝑡 < 1 and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2).

Solution
Step 1. 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2) ⇒ 𝑇𝑇 = 2.
Step 2. The graph of 𝑓𝑓(𝑡𝑡) is symmetrical about the vertical axis (cf.
Example 4.3.2(a)), so 𝑓𝑓(𝑡𝑡) is an even function, and hence, 𝑏𝑏𝑛𝑛 = 0
for all positive integers of 𝑛𝑛.

1 1 1
1
𝑎𝑎0 = � 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑 = 2 � 𝑓𝑓(𝑡𝑡)𝑑𝑑𝑑𝑑 = 2 � 𝑡𝑡 𝑑𝑑𝑑𝑑 = �𝑡𝑡 2 � = 1
0
−1 0 0

1 1

𝑎𝑎𝑛𝑛 = � 𝑓𝑓(𝑡𝑡) cos(𝑛𝑛𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑 = 2 � 𝑓𝑓(𝑡𝑡) cos(𝑛𝑛𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑

−1 0
1 1
𝑡𝑡 sin(𝑛𝑛𝑛𝑛𝑛𝑛) cos(𝑛𝑛𝑛𝑛𝑛𝑛)
= 2 � 𝑡𝑡 cos(𝑛𝑛𝑛𝑛𝑛𝑛) 𝑑𝑑𝑑𝑑 = 2 � + �
𝑛𝑛𝑛𝑛 (𝑛𝑛𝑛𝑛)2 0
0

(−1)𝑛𝑛 − 1
= 2� � for all positive integers 𝑛𝑛.
(𝑛𝑛𝑛𝑛)2

Step 3. The Fourier series of 𝑓𝑓(𝑡𝑡) is

∞
1 (−1)𝑛𝑛 − 1
𝑓𝑓(𝑡𝑡) = + 2 � � � cos(𝑛𝑛𝑛𝑛𝑛𝑛).
2 (𝑛𝑛𝑛𝑛)2
𝑛𝑛=1

∞
1 4 cos�(2𝑘𝑘 − 1)𝜋𝜋𝜋𝜋�
Step 4. 𝑓𝑓(𝑡𝑡) = − 2 � (2𝑘𝑘 − 1)2
.
2 𝜋𝜋
𝑛𝑛=1
Fourier Series 94

4.4 Applications
In this section, we will see how Fourier series can be used to find the exact
value of certain infinite series.

Example 4.4.1
Use the Fourier series of the Sawtooth function in Example 4.3.8 to calculate
the exact value of the infinite series
1 1 1 1
1 − + − + − ⋯,
3 5 7 9
leaving the answer in terms of 𝜋𝜋.

Solution
From Example 4.3.8,
∞
(−1)𝑛𝑛+1
𝑡𝑡 = 2 � sin(𝑛𝑛𝑛𝑛) for − 𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋
𝑛𝑛
𝑛𝑛=1
𝜋𝜋
Substituting 𝑡𝑡 = 2 into the Fourier series above, we have
∞
𝜋𝜋 (−1)𝑛𝑛+1 𝑛𝑛𝑛𝑛
=2 � sin � �
2 𝑛𝑛 2
𝑛𝑛=1

𝜋𝜋 3𝜋𝜋 5𝜋𝜋
sin 2 sin 𝜋𝜋 sin 2 sin 2𝜋𝜋 sin 2 sin 3𝜋𝜋
= 2� − + − + − +⋯�
1 2 3 4 5 6

−1 1 −1
= 2 �1 − 0 + −0+ −0+ −0+⋯�
3 5 7
So,
𝜋𝜋 1 1 1
= 1− + − +⋯
4 3 5 7
Fourier Series 95

4.5 Exercises

Part A: Baseline Problems

1. Sketch three cycles of the following periodic functions, and determine

their periods (See Example 4.1.4).
−1 for −𝜋𝜋 < 𝑡𝑡 < 0
(a) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
0 for 0 < 𝑡𝑡 < 𝜋𝜋
2 for −𝜋𝜋 < 𝑡𝑡 < 0
(b) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
−2 for 0 < 𝑡𝑡 < 𝜋𝜋
2 if −1 < 𝑡𝑡 < 0
(c) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
1 if 0 < 𝑡𝑡 < 1
−3 if −1 < 𝑡𝑡 < 0
(d) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
2 if 0 < 𝑡𝑡 < 1

2(a) to (d). Calculate the Fourier series of the functions in Problem 1(a) to (d),
respectively (See Example 4.2.3).

Part B: Intermediate Problems

3. Sketch three cycles of the following periodic functions, and determine

their periods (See Example 4.1.4).
−𝑡𝑡 for −1 < 𝑡𝑡 ≤ 0
(a) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
0 for 0 < 𝑡𝑡 < 1
0 for −𝜋𝜋 < 𝑡𝑡 ≤ 0
(b) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
2𝑡𝑡 for 0 < 𝑡𝑡 < 𝜋𝜋
1 for −1 < 𝑡𝑡 < 0
(c) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
1 − 𝑡𝑡 for 0 ≤ 𝑡𝑡 < 1
(d) 𝑓𝑓(𝑡𝑡) = 𝑡𝑡 + 1 for − 1 < 𝑡𝑡 < 1 and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
Fourier Series 96

4(a) to (d). Calculate the Fourier series of the functions in Problem 2(a) to (d),
respectively (See Example 4.2.4).

5. Sketch three cycles of the following periodic functions, and for each
function, determine whether it is even or odd function (See Example
4.3.2).
−3 if −1 < 𝑡𝑡 < 0
(a) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
3 if 0 < 𝑡𝑡 < 1
𝑡𝑡 if −𝜋𝜋 < 𝑡𝑡 ≤ 0
(b) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
−𝑡𝑡 if 0 < 𝑡𝑡 < 𝜋𝜋
1 + 𝑡𝑡 if −1 < 𝑡𝑡 ≤ 0
(c) 𝑓𝑓(𝑡𝑡) = � and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2)
1 − 𝑡𝑡 if 0 < 𝑡𝑡 < 1
(d) 𝑓𝑓(𝑡𝑡) = −2𝑡𝑡 for − 𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋 and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)

6(a) to (d). Calculate the Fourier series for the functions in Problem 5(a) to (d),
respectively (See Examples 4.3.7 – 4.3.9).

Part C: Application Problems

(See Section 4.4)

7. (a) Calculate the Fourier Series of 𝑓𝑓(𝑡𝑡) given by

𝑓𝑓(𝑡𝑡) = 𝑡𝑡 − 𝑡𝑡 2 for −𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋 and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
(b) Hence, calculate the exact value of the infinite series
1 1 1 1 1
− + − + − ⋯,
12 22 32 42 52
leaving the answer in terms of 𝜋𝜋.
Fourier Series 97

8. (a) Calculate the Fourier Series of 𝑓𝑓(𝑡𝑡) given by

𝑓𝑓(𝑡𝑡) = 𝑡𝑡 sin 𝑡𝑡 for −𝜋𝜋 < 𝑡𝑡 < 𝜋𝜋 and 𝑓𝑓(𝑡𝑡) = 𝑓𝑓(𝑡𝑡 + 2𝜋𝜋)
(b) Hence, calculate the exact value of the infinite series
1 1 1 1 1
− + − + − ⋯,
1 ⋅ 3 3 ⋅ 5 5 ⋅ 7 7 ⋅ 9 9 ⋅ 11
leaving the answer in terms of 𝜋𝜋.

Part D: Beyond the Concepts

9. Use the Fourier series of 𝑓𝑓(𝑡𝑡) in Problem 5(b) to calculate the exact value
of the following infinite series, leaving the answer in terms of 𝜋𝜋.
1 1 1 1
(a) 2
+ 2+ 2+ 2+⋯
1 3 5 7

1 1 1 1
(b) + + + +⋯ (rearrangement of terms is allowed here)
12 22 32 42

1 1 1 1
(c) 4
+ 4+ 4+ 4+⋯ by using the Parseval’s Identity
1 2 3 4
𝑇𝑇/2 ∞
2 2
𝑎𝑎02
� [𝑓𝑓(𝑡𝑡)] 𝑑𝑑𝑑𝑑 = + �(𝑎𝑎𝑛𝑛2 + 𝑏𝑏𝑛𝑛2 )
𝑇𝑇 2
−𝑇𝑇/2 𝑛𝑛=1

∞
1
Remark. The infinite series � is called the Riemann Zeta function.
𝑘𝑘 𝑠𝑠
𝑘𝑘=1

− END OF CHAPTER 4 −