lOMoARcPSD|42174153

Contents

1. Chapter 1: Basics of Differential equation and its application Page no.

(i) Introduction 5 5
(ii) Basic Definitions of differential equation; Differential equation, 6 6
Order of Differential equation, Degree of Differential equation
Types of Differential equation 7
(iii) Applications of Differential equation 7 7
2. Chapter 2: Methods of solving Ordinary Differential equation
(i) Solutions of Differential equation; General solution, Particular 13 13
solution, singular solution.
(ii) Methods of solving Ordinary differential equation of 1st order 13
14
Variable separable method, Reducible to separable, Linear
Differential equation, Reducible to linear, Exact Differential equation,
Linear Higher Order Differential equation with constant coefficients
(iii) Second order differential equation with variable coefficients 18
19
Variation of parameter, change of independent variable
(iv) Solving of Ordinary differential equation of first order 23
23
and first degree by numerical method
Laplace transform, Picard’s successive approximation,
Euler’s method, Euler’s modifies method
(v) Power series solutions, Frobenius method 26 27
3. Chapter 3: Functions of Differential equation
(i) Legendre’s Differential equation, Power series solution 35
31
of Legendre’s Differential equation, Legendre’s function,
Generating function of Legendre’s polynomial, Some important results,
orthogonality property, Elementary properties of Legendre’s polynomial
(ii) Hermite Differential equation, Power Series solution of Hermite 45 48
Differential equation, Hermite polynomial
(iii) Bessel equation and Bessel function 48 56
61

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

Chapter 1
Introduction
The subject of differential equation constitutes a large and very important branch of modern
mathematics. An equation involving derivatives of one or more dependent variable with respect to
one or more independent variable is called differential equation.
For example,
𝑑2𝑦 𝑑𝑦 2
+ 𝑥𝑦 ( ) = 0
𝑑𝑥 2 𝑑𝑥
Form the early days of the calculus the subject has been an area of great theoretical research and
practical applications and so it continues to be so in our days. The nature arises several questions.
These questions indicate three major accepts of the subject; theory, method and applications.
Differential equation occurs in connection with numerous problems that are encountered in the
various branches of science and engineering. We indicate a few such problems in the following
list:
(1) The problem of determining the motion of a projection, rocket, satellite or planet.
(2) The problem of determining the charge or current in a electric circuit.
(3) The problem of the conduction of heat in a rod or in a slab.
(4) The problem of determining the vibrations of a wire.
(5) The study of rate of decomposition of a radioactive substance or the rate of growth of
population.
(6) The study of the reaction of chemicals.
The mathematical formulation of such problems give rise to differential equations. The situation
in each of the above problems, the objects involved obeys certain scientific law. These laws
involve various rate of change of one or more quantities with respect to other quantities.
Let us recall that such rates of changes are expressed mathematically by derivatives. In the
mathematical formulation of each of the above situation, the various rates of change are thus
expressed by various derivatives and the scientific laws are thus expressed themselves become
mathematical equations involving derivatives, that is differential equation. In this process of
mathematical formulation, certain simplifying assumptions generally have to be made in order that
the resulting differential equation be tractable.
But here the question arise that how does one obtain useful information from the differential
equation. The answer is essentially that if it is possible to so, one solves the differential equation
to obtain a solution. If this is not possible one uses the theory of differential equation to obtain
information about the solution.

Some important basic definition of the differential equations.

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

Differential equation: A differential equation is a mathematical equation that relates some

function with its derivatives.
or, it may also be said that:
An equation which contains derivatives of one or more dependent variable with respect to one or
more independent variable is called differential equation.
e.g.,
𝑑𝑦
= 2𝑥 + 2 ,
𝑑𝑥
𝑑𝑦 = 𝑐𝑜𝑠𝑥 𝑑𝑥
are differential equation.
Order of a differential equation: - The order of a differential equation is the order of the highest
differential coefficient occurring in it.
e.g.,
𝑑𝑦 2
𝑜𝑟𝑑𝑒𝑟 𝑜𝑓 ( ) + 7𝑥 + 5 = 0 𝑖𝑠 1
𝑑𝑥
𝑑2 𝑦 𝑑𝑦
𝑜𝑟𝑑𝑒𝑟 𝑜𝑓 2 + 𝑥 + 2 = 0 𝑖𝑠 2
𝑑𝑥 𝑑𝑥
Degree of a differential equation: - The degree of a differential equation is the degree of the
highest differential coefficient when the equation has been made free from fractions and radicals
as for as the differential coefficient is concerned.
e.g., Consider
𝑑2𝑦 𝑑𝑦
2
+7 +2=0
𝑑𝑥 𝑑𝑥
It is the equation with degree 1 and order 2 as the degree of the highest differential coefficient is
1
Consider
5
𝑑𝑦 2
[1 + ( ) ]
𝑑𝑥
=2
𝑑2𝑦
𝑑𝑥 2

This can be written as

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

5
𝑑𝑦 2 𝑑2𝑦
[1 + ( ) ] = 2 2
𝑑𝑥 𝑑𝑥

Degree of differential equation is 1 and order is 2

Types of differential equation: There are two of differential equation
(i) Ordinary differential equation: -

An ordinary differential equation is a differential equation that depends on only one

independent variable
e.g.
𝑑𝑦
+ 𝑥 + 1 = 0 𝑖𝑠 𝑎𝑛 𝑜𝑟𝑑𝑖𝑛𝑎𝑟𝑦 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑡𝑖𝑎𝑙 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛
𝑑𝑥
(ii) Partial differential equation: -
A partial differential equation is a differential equation in which the dependent variable
depends on more than two independent variables.
e.g.
𝜕𝑧 𝜕𝑧
−5 = 2 𝑖𝑠 𝑎 𝑝𝑎𝑟𝑡𝑖𝑎𝑙 𝑑𝑖𝑓𝑓𝑒𝑛𝑡𝑖𝑎𝑙 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛
𝜕𝑥 𝜕𝑦

Application of Differential Equation:

The study of Differential Equation is a wide field in pure and applied mathematics, physics and
engineering. All of these disciplines are concerned with the properties of Differential Equations
of various types.
Pure mathematics focuses on the existence and uniqueness of solutions, while applied
mathematics emphasizes the rigorous justification of the methods for approximating solutions.
Differential Equations plays an important role in modelling virtually every physical, technical, or
biological process, from celestial motion, to bridge design, to interactions between neurons.
Differential Equation such as those used to solve real life problems may not necessarily be directly
solvable i.e., do not have closed form solutions. Instead, solutions can be approximated using
numerical methods.

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

Many fundamentals laws of physics and chemistry can be formulated as Differential Equations.
Because laws are a set of rules followed by a system and such rules are used to explain a
phenomenon exhibited by a system and are usually expressed as a mathematical relation. Such
rules are observed by repeated experimentation. For example, Newton’s law of cooling is the first
order ordinary differential equation and is stated as the temperature of a body changes at a rate
proportional to the difference in temperature between its own temperature and the temperature of
its surroundings.
The differential equation of Newton law of cooling is given as:
𝑑𝑇
= −𝑘(𝑇 − 𝑇𝑠 )
𝑑𝑡
Where,
T = temperature of a body at any time, t
𝑇𝑠 = temperature of the surroundings
K = constant of proportionality
In biology and economics, differential equation is used to model the behavior of complex systems.
The mathematical theory of Differential Equation first developed together with the sciences where
the equations had originated and where the results found applications.
However, diverse problems, sometimes originating in quite distinct scientific fields, may give rise
to identical differential equations can be viewed as a unifying principle behind diverse phenomena.
As an example, Consider the propagation of light and sound in the atmosphere and of waves of a
pond. All of them may be described by the same second- order P.D.E. The wave equation which
allows us to think of light and sound as form of waves, much like familiar waves in the water.

Physics:
Differential Equation is very important in the field of Physics because many of the important
laws, equations in physics are in the form of differential equations.
The wave equation is a hyperbolic partial differential equation. It typically concerns a time
variable ‘t’, one or more spatial variables 𝑥1 , 𝑥2 . . . . and a scalar function 𝑢 =
𝑢(𝑥1 , 𝑥2 . . . . 𝑥𝑛 ; 𝑡) whose value could model, for examples the mechanical displacement of a
wave, The wave equation for u is,
𝜕 2𝑢
= 𝑐 2 ∇2 𝑢
𝜕𝑡 2
Where,
𝜕2
∇2 = , 𝑖𝑠 𝑡ℎ𝑒 (𝑠𝑝𝑎𝑡𝑖𝑎𝑙) 𝐿𝑎𝑝𝑙𝑖𝑐𝑖𝑎𝑛 𝑎𝑛𝑑 𝑐 𝑖𝑠 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
𝜕𝑥 2

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

Classical Mechanics:
As the force acting on a particle is known, Newton’s second law i.e.
𝑑𝑝 𝑑(𝑚𝑣)
𝐹= =
𝑑𝑡 𝑑𝑡
is sufficient to describe the motion of a particle. Once independent relations for each force acting
on a particle are available, they can be substituted into Newton’s second law to obtain an ODE,
which is called the equation of motion.
Electrodynamics:
Maxwell’s equation is a set of partial differential equation that, together with the Lorentz force
law, form the foundation of classical electrodynamics, classical optics, and electric circuits, these
fields in turn underlie modern electrical and communication technologies.
Maxwell’s equation describe how electric and magnetic fields are generated and altered by each
other and by charges and currents. They are named after the Scottish Physicist and mathematician
James Clerk Maxwell’s, who published an early form of those equations between 1861 and 1862.
General relativity:
The Einstein field equations also known as Einstein equations are a set of ten partial differential
equations in Albert Einstein’s general theory of relativity which describe the fundamental
interaction of gravitation as a result of space time being curved by matter and energy.

Quantum Mechanics:
In quantum Mechanics, the analogue of Newton’s law is Schrodinger’s equation (a partial
differential equation) for a quantum system (usually atoms, molecules and subs-atomic particles
whether free bound, or localized). It is not a simple algebraic equation, but in general a linear
partial differential equation, describing the time evolution of the system’s wave function also
called a state function.

Biology:
Predator prey equations:
The Lotka - Volterra equation also known as the predator-prey equations, are a pair of first-order,
non-linear, differential equations is frequently used to describe the population dynamics of two
species that interact, one as a predator and the other as prey.
The population change through time according to the pair of equation:
𝑑𝑥 𝑑𝑦
= 𝛼𝑥 − 𝛽𝑥𝑦 , = 𝛿𝑥𝑦 − 𝛾𝑦
𝑑𝑡 𝑑𝑡
where, ‘x’ is the number of prey

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

10

‘y’ is the number of predator

𝑑𝑦 𝑑𝑥
, represent the instantaneous growth rate
𝑑𝑡 𝑑𝑡

𝛼 , 𝛽 , 𝛾 , 𝛿 are positive real parameters describing the interactions between two

species.
Physical meaning of the equation
The Lotka-Volterra model males a number of assumptions about the environment of the predator
and prey populations:
1. The prey population finds ample food at all times.
2. The food supply of the predator population depends entirely on the species of the prey
population.
3. The rate of change of population is proportional to its size.
4. During the process, the environment does not change in favor of one species and genetic
adaptation is inconsequential.
5. Predators have limitless appetite.
As differential equation are used, the solution is deterministic and continuous. This in turn implies
that the generations of both the predator and prey are continually overlapping.

Chemistry:
Chemical reactions kinetics is the study of rates of chemical processes (reactions). The goal is to
find the relations between the concentrations ‘c’ of products of a chemical reaction (as depending
variable) and the time t (as independent variable)
In general, all chemical reactions can be described mathematically by first order differential
equation.
Simple reaction like the transformation of A to B (A→B) can be describe by the differential
equation.
−𝑑𝑐𝐴 = 𝑘. 𝑐𝐴 . 𝑑𝑡
The first order differential equation describes also a first order reaction in chemical kinetics, due
to exponent 1 of the concentration 𝑐𝐴 .
Following a separable of variables, the integration result is
𝐶𝐴𝑡 𝑡
𝑑𝑐𝐴
∫ = −𝑘 ∫ 𝑑𝑡
𝐴0 𝑐𝐴 0

𝐶𝐴𝑡
= ln = −𝑘𝑡
𝐶𝐴0

Can also be written in the exponent form

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

11

𝐶𝐴𝑡 = 𝐶𝐴0 𝑒 −𝑘𝑡

In majority of chemical reactions, however more than one product involves

e.g. A+B→C+D
(1)
These are second order reactions in chemical kinetics, because the exponent of concentrations of
A and B in the rate of law is two.
In case of a simple reaction, first order differential equation are resulting for the math description
𝑑𝐶𝐴
− = 𝑘. 𝐶𝐴 . 𝐶𝐵
𝑑𝑡
For equal initial concentration of A and B, Reaction (1) can be written as
2A→C+D

With the differential equation

1 𝑑𝑐 𝐴
− = 𝑘. 𝐶𝐴2
2 𝑑𝑡
Economics:
(i) The key equation of the Solow-Swan model is

𝜕𝑘(𝑡)
= 𝑆[𝑘(𝑡)]𝛼 − 𝑆𝑘(𝑡)
𝜕𝑥
(ii) The Black-Scholes equation is the P.D.E governing the price evolution of European call or
European put under the Black-chokes model.

For a European call or put on an underlying stock paying no dividends, the eq. is
𝜕𝑉 1 2 2 𝜕 2 𝑉 𝜕𝑉
+ 𝜎 𝑆 + 𝑟𝑆 − 𝑟𝑉 = 0
𝜕𝑡 2 𝜕𝑆 2 𝜕𝑆
where ‘V’ is the price of option as a function of stock price ‘S’ and time ‘t’
‘r’ is the risk-free interest rate
′𝜎′ is the volatility of the stock

(iii) The Sethi advertising model or simply the Sethi model provides a sales advertising dynamics
in the form of the following stochastic differential equation

𝑑𝑋𝑡 = (𝑟𝑈𝑡 √1 − 𝑋𝑡 − 𝛿𝑋𝑡 ) 𝑑𝑡 + 𝜎 (𝑋𝑡 )𝑑𝑧𝑡 , 𝑋0 = 𝑥

where,

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

12

′𝑋𝑡 ’ is the market share at time ‘t’

′𝑈𝑡 ′ is the rate of advertising at time ‘t’
‘r’ is the coefficient of the effectiveness of advertising
′𝛿′ is the decay constant
𝜎(𝑋𝑡 ) is the diffusion coefficient
′𝑧𝑡 ′ is the wienes process (Standard Brownian motion)
′𝑑𝑧𝑡 ′ is known as White noise

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

13

Chapter 2

Solution of a differential equation

A solution of a differential equation is a relation between the variables such that this relation and
the derivatives obtained from this relation satisfy the given differential equation.
General solution: - The solution of a differential equation which involves as many arbitrary
constants as the order of the differential equation, is called general solution.
It is also called complete solution.
Particular solution: - A particular solution of a differential equation is that which contain no
arbitrary constant and is obtained from the general solution by giving particular values to the
arbitrary constants.
Singular solution; - A singular solution of a differential equation is that which contains no
arbitrary constant and cannot be obtained from the general solution by giving particular values to
the arbitrary constants.

Basic methods of solving differential equations

● Differential equations of 1St order
(1) Variable separable method: -

The equation of the form f(x) dx + f(y) dy =0 where we can separate the variables easily, the
method we generally use to solve the differential equation is variable of separable method.
e.g.
𝑥(1 + 𝑦 2 )𝑑𝑥 + 𝑦(1 + 𝑥 2 )𝑑𝑦 = 0
𝑦(1 + 𝑥 2 )𝑑𝑦 = −𝑥(1 + 𝑦 2 )𝑑𝑥

𝑦 𝑥
2
𝑑𝑦 = − 𝑑𝑥
1+𝑦 1 + 𝑥2

Integrating
𝑦 𝑥
∫ 2
𝑑𝑦 = −∫ 𝑑𝑥
1+𝑦 1 + 𝑥2
𝑦 𝑥
∫ 2
𝑑𝑦 + ∫ 𝑑𝑥 = 0
1+𝑦 1 + 𝑥2
log(1 + 𝑦 2 ) + log (1 + 𝑥 2 ) = 𝑙𝑜𝑔𝑐
log(1 + 𝑦 2 ) (1 + 𝑥 2 ) = 𝑙𝑜𝑔𝑐 ⇒ (1 + 𝑥 2 )(1 + 𝑦 2 ) = 𝑐 , 𝑤ℎ𝑖𝑐ℎ 𝑖𝑠 𝑡ℎ𝑒 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 𝑠𝑜𝑙.

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

14

(2) Reducible to separable: -

Let us consider the equation
𝑑𝑦
= (4𝑥 + 𝑦 + 1) (1)
𝑑𝑥
In the above equation (1), we can’t use variable separable method because x and y do not
separate
So, Put ,
(4𝑥 + 𝑦 + 1) = 𝑡
𝑑𝑦 𝑑𝑡
4+ =
𝑑𝑥 𝑑𝑥
𝑑𝑦 𝑑𝑡
= −4
𝑑𝑥 𝑑𝑥
Therefore eq. (1) becomes,
𝑑𝑡
− 4 = 𝑡2
𝑑𝑥
𝑑𝑡
𝑜𝑟 , = 𝑡2 + 4
𝑑𝑥
Separating the variables , we get
𝑑𝑡
= 𝑑𝑥
𝑡2 + 4
Integrating both sides,
𝑑𝑡
∫ = ∫ 𝑑𝑥
𝑡2 + (2)2
1 𝑡
𝑜𝑟, tan−1 = 𝑥 + 𝑐
2 2
𝑡
𝑜𝑟, tan−1 = 2(𝑥 + 𝑐),
2
𝑡
𝑜𝑟, = 2 tan[2(𝑥 + 𝑐)]
2
4𝑥 + 𝑦 + 1 = 2 tan[2 (𝑥 + 𝑐)]
Which is the required solution

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

15

(3) Linear differential equation: -

There are two types of linear, x - linear and y – linear
𝑑𝑥
x—linear :- The differential equation of the form + 𝑃(𝑦)𝑥 = 𝑄(𝑦) is x- linear and its
𝑑𝑦
integrating factor is given by 𝑒 ∫ 𝑃(𝑦)𝑑𝑦
𝑑𝑦
y linear :- The differential equation which is of the form + 𝑃(𝑥)𝑦 = 𝑄(𝑥) is y- linear and
𝑑𝑥
its integration factor is given by 𝑒 ∫ 𝑃(𝑥)𝑑𝑥
e.g.
𝑑𝑦
(1 + 𝑦 2 ) + (𝑥 − tan−1 𝑦) =0
𝑑𝑥
𝑑𝑥
(1 + 𝑦 2 ) + 𝑥— tan−1 𝑦 = 0
𝑑𝑦
𝑑𝑥
(1 + 𝑦 2 ) + 𝑥 = tan−1 𝑦
𝑑𝑦
𝑑𝑥 𝑥 𝑡𝑎𝑛−1 𝑦
+ =
𝑑𝑦 1 + 𝑦 2 1 + 𝑦 2
1 𝑡𝑎𝑛−1 𝑦
𝑃= , 𝑄 =
(1 + 𝑦 2 ) 1 + 𝑦2

1
∫ −1 𝑦
𝐼. 𝑓 = 𝑒 ∫ 𝑃(𝑦)𝑑𝑦 = 𝑒 1+𝑦 2 = 𝑒 tan
Therefore, 𝑥 × 𝐼. 𝑓 = ∫ 𝐼. 𝑓 × 𝑄 𝑑𝑦 + 𝑐

−1 𝑦 −1 𝑦 𝑡𝑎𝑛−1 𝑦
𝑥. 𝑒 tan = ∫ 𝑒 tan 𝑑𝑦 + 𝑐
1 + 𝑦2
Let tan−1 𝑦 = 𝑡
1
1+𝑦 2
𝑑𝑦 = 𝑑𝑡

𝑥𝑒 𝑡 = ∫ 𝑒 𝑡 𝑡
𝑥𝑒 𝑡 = 𝑡𝑒 𝑡 − 𝑒 𝑡 + 𝑐
−1 𝑦 −1 𝑦 −1 𝑦
𝑥𝑒 tan = 𝑡𝑎𝑛−1 𝑦 𝑒 𝑡𝑎𝑛 − 𝑒 tan +𝑐
−1 𝑦 −1 𝑦
𝑥𝑒 𝑡𝑎𝑛 = 𝑒 tan (tan−1 𝑦 − 1) + 𝑐
which is the required solution.

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

16

(4) Reducible to linear differential equation:-

𝑑𝑦
The equation of the form + 𝑃(𝑥)𝑦 = 𝑄(𝑥)𝑦 𝑛 are not solvable by the Linear differential
𝑑𝑥
method. The method we use to solve these types of diff. eq. is Reducible to linear . The above eq.
is also known as Bernoulli diff. eq.
e.g.
𝑑𝑦
− 𝑦𝑡𝑎𝑛𝑥 = −𝑦 2 sec 𝑥
𝑑𝑥
1 𝑑𝑦 𝑡𝑎𝑛 𝑥
− = − sec 𝑥
𝑦 2 𝑑𝑥 𝑦
1 1 𝑑𝑦 𝑑𝑡
𝑙𝑒𝑡 =𝑡 ⇒ − 2 =
𝑦 𝑦 𝑑𝑥 𝑑𝑥
𝑑𝑡
− − 𝑡𝑎𝑛𝑥 𝑡 = − sec 𝑥
𝑑𝑥
𝑑𝑡
+ tan 𝑥 𝑡 = sec 𝑥
𝑑𝑥
P = tan x , Q = sec x
I.F =
𝑒 ∫ 𝑡𝑎𝑛 𝑥 𝑑𝑥
= 𝑒 log sec 𝑥 = sec 𝑥
Therefore the solution of the given diff . is
𝑡. sec 𝑥 = ∫ sec 2 𝑥 𝑑𝑥 + 𝑐
𝑡 𝑠𝑒𝑐 𝑥 = 𝑡𝑎𝑛 𝑥 + 𝑐
𝑠𝑒𝑐
= 𝑡𝑎𝑛 𝑥 + 𝑐
𝑦
(5) Exact differential equation: - The equation of the form

𝜕𝑀 𝜕𝑁
𝑀 (𝑥, 𝑦)𝑑𝑥 + 𝑁 (𝑥, 𝑦 ) 𝑑𝑦 = 0 𝑤ℎ𝑒𝑟𝑒 =
𝜕𝑦 𝜕𝑥
is called the exact differential equation
e.g.
(𝑦 3 − 2𝑦)𝑑𝑥 + (3𝑥𝑦 2 + 4𝑦 3 − 2𝑥)𝑑𝑦 = 0 (1)
Comparing it with M dx + N dy = 0
𝑀 = 𝑦 3 − 2𝑦 , 𝑁 = 3𝑥𝑦 2 + 4𝑦 3 − 2𝑥

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

17

Now
𝜕𝑀 𝜕𝑁
= 3𝑦 2 − 2 , = 3𝑦 2 − 2
𝜕𝑦 𝜕𝑥
Therefore
𝜕𝑀 𝜕𝑁
=
𝜕𝑦 𝜕𝑥
So the solutions of (1) is

∫ (𝑦 3 − 2𝑦) 𝑑𝑥 + ∫ ( 𝑡𝑒𝑟𝑚𝑠 𝑖𝑛 𝑁 𝑛𝑜𝑡 𝑐𝑜𝑛𝑡𝑎𝑖𝑛𝑖𝑛𝑔 𝑥 ) 𝑑𝑥 = 𝑐

𝑦 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡

or,
(𝑦 3 − 2𝑦) ∫ 1 𝑑𝑥 + ∫ 4 𝑦 3 𝑑𝑦 = 𝑐

or,
4𝑦 3
(𝑦 3 − 2𝑦) (𝑥) + =𝑐
4
or
𝑥(𝑦 3 − 2𝑦) + 𝑦 4 = 𝑐
(6) Linear Higher order differential equation with constant coefficients:-
e.g. Consider the differential equation
𝑑2𝑦 𝑑𝑦
+ 3 + 2𝑦 = 𝑥 𝑒 𝑥 sin 𝑥 (1)
𝑑𝑥 2 𝑑𝑥
The symbolic form of (1) is
(𝐷 2 + 3𝐷 + 2 )𝑦 = 𝑥 𝑒 𝑥 sin 𝑥 (2)
The auxiliary equation of (1) is
𝑚2 + 3𝑚 + 2 = 0
⇒ 𝑚2 + 2𝑚 + 𝑚 + 2 = 0
⇒ 𝑚(𝑚 + 2) + 1(𝑚 + 2) + 0
⇒ (𝑚 + 1)(𝑚 + 2) = 0
⇒ 𝑚 = −1 , −2
The complementary function of (1) is
𝑦𝐶 = 𝐶1 𝑒 −𝑥 + 𝐶2 𝑒 −2𝑥
Now ,

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

18

1
𝑦𝑃 = 𝑥 𝑒 𝑥 𝑆𝑖𝑛 𝑥
𝐷2 + 3𝐷 + 2
1
= 𝑒𝑥 𝑥 sin 𝑥
(𝐷 + 1)2 + 3 (𝐷 + 1) + 2
1
= 𝑒𝑥 𝑥 sin 𝑥
𝐷2 + 1 + 2𝐷 + 3𝐷 + 3 + 2
1
= 𝑒𝑥 [ 𝑥 sin 𝑥]
𝐷2 + 5𝐷 + 6
1 (2𝐷 + 5)
= 𝑒 𝑥 [ 𝑥. sin 𝑥 − sin 𝑥]
𝐷2 + 5𝐷 + 6 (𝐷2 + 5𝐷 + 6)2

1 (2𝐷 + 5)
= 𝑒 𝑥 [ 𝑥. sin 𝑥 − 𝑠𝑖𝑛 𝑥]
−12 + 5𝐷 + 6 (−12 + 5𝐷 + 6)2

1 (2𝐷 + 5)
= 𝑒 𝑥 [ 𝑥. sin 𝑥 − 𝑠𝑖𝑛 𝑥]
5𝐷 + 5 (5𝐷 + 5)2
𝑥 𝐷−1 1 2𝐷 + 5
= 𝑒𝑥 [ sin 𝑥 − sin 𝑥]
5 (𝐷 + 1)(𝐷 − 1) 25 (𝐷 + 1)2
𝑥 𝐷−1 1 2𝐷 + 5
= 𝑒𝑥 [ 2
𝑠𝑖𝑛 𝑥 − 2
𝑠𝑖𝑛 𝑥]
5 𝐷 −1 25 𝐷 + 1 + 2𝐷
𝑥 𝐷−1 1 2𝐷 + 5
= 𝑒𝑥 [ 2
𝑠𝑖𝑛 𝑥 − 2
𝑠𝑖𝑛 𝑥]
5 −1 − 1 25 −1 + 1 + 2𝐷

𝑥 1 2𝐷 + 5
= 𝑒𝑥 [ − (𝐷 𝑠𝑖𝑛𝑥 − 𝑠𝑖𝑛𝑥) − sin 𝑥]
10 25 2𝐷

𝑥 1 1
= 𝑒𝑥 [ − (cos 𝑥 − 𝑠𝑖𝑛𝑥) − (2 + 5 ) 𝑠𝑖𝑛 𝑥]
10 50 𝐷

𝑥 1
= 𝑒𝑥 [ − (𝑐𝑜𝑠 𝑥 − 𝑠𝑖𝑛𝑥) − (2 sin 𝑥 + 5 cos 𝑥) ]
10 50
Thus the complete solution of (1) is 𝑦 = 𝑦𝑐 + 𝑦𝑃

● Important Methods to solve the Second order differential equations with

variable coefficient: -

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

19

(1) Change of dependent variable method

(2) Variation of parameter
(3) Change of independent variable

Example: -. Solve the differential equation

𝑑2𝑦
+ 𝑦 = 𝑡𝑎𝑛𝑥
𝑑𝑥 2
by using the method of variation of parameters.
Sol: - Given differential is
𝑑2𝑦
+ 𝑦 = 𝑡𝑎𝑛𝑥 (1)
𝑑𝑥 2
Now complimentary function of eq. (1) is
𝑦𝑐 = 𝐶1 𝑆𝑖𝑛𝑥 + 𝐶2 𝐶𝑜𝑠 𝑥
We assume 𝑦𝑃 (𝑥) = 𝑣1 (𝑥)𝑠𝑖𝑛𝑥 + 𝑣2 (𝑥)𝑐𝑜𝑠 𝑥 (2)
Now,
Differentiating eq. (2) w.r.t ‘x’, we get
𝑦𝑝′ (𝑥) = 𝑣1 (𝑥) cos 𝑥 − 𝑣2 (𝑥)𝑠𝑖𝑛𝑥 + 𝑣1′ (𝑥)𝑠𝑖𝑛𝑥 + 𝑣2 ′(𝑥) 𝑐𝑜𝑠𝑥

Now , we impose the condition 𝑣1 ′ (𝑥)𝑠𝑖𝑛𝑥 + 𝑣2 ′(𝑥)𝑐𝑜𝑠 𝑥 = 0 (3)

Therefore , 𝑦𝑝′ (𝑥) = 𝑣1 (𝑥) 𝑐𝑜𝑠 𝑥 – 𝑣2 (𝑥)𝑠𝑖𝑛𝑥 (4)

Again, differentiating eq. (4) w.r. t ‘x’ ,we get

𝑦𝑝′′ (𝑥) = −𝑣1 (𝑥) 𝑠𝑖𝑛𝑥 + cos 𝑥 𝑣1′ (𝑥) − 𝑣2 (𝑥) cos 𝑥 − 𝑠𝑖𝑛𝑥 𝑣2′ (𝑥) (5)

Now substituting the value of eq. (2) and eq. (5) in eq. (1), we get
−𝑣1 (𝑥) 𝑠𝑖𝑛𝑥 + cos 𝑥 𝑣1′ (𝑥) − 𝑣2 (𝑥) cos 𝑥 − 𝑠𝑖𝑛𝑥 𝑣2′ (𝑥) + 𝑣1 (𝑥) sin 𝑥 + 𝑣2 (𝑥) cos 𝑥
= 𝑡𝑎𝑛𝑥

⇒ 𝑣1′ (𝑥)𝑐𝑜𝑠𝑥 − 𝑣2′ (𝑥)𝑠𝑖𝑛𝑥 = 𝑡𝑎𝑛𝑥 (6)

Now, we have
𝑣1 ′ (𝑥)𝑠𝑖𝑛𝑥 + 𝑣2 ′(𝑥)𝑐𝑜𝑠 𝑥 = 0
and 𝑣1′ (𝑥)𝑐𝑜𝑠𝑥 − 𝑣2′ (𝑥)𝑠𝑖𝑛𝑥 = 𝑡𝑎𝑛𝑥

(7)
Solving these two equation for 𝑣1′ (𝑥)𝑎𝑛𝑑 𝑣2′ (𝑥)

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

20

Therefore ,
0 𝑐𝑜𝑠𝑥
| |
𝑣1′ (𝑥) = 𝑡𝑎𝑛𝑥 −𝑠𝑖𝑛𝑥
𝑠𝑖𝑛𝑥 𝑐𝑜𝑠𝑥
| |
𝑐𝑜𝑠𝑥 −𝑠𝑖𝑛𝑥

0 − 𝑐𝑜𝑠𝑥 𝑡𝑎𝑛𝑥
=
− sin2 𝑥 − cos2 𝑥
𝑠𝑖𝑛𝑥
−𝑐𝑜𝑠𝑥
= 𝑐𝑜𝑠𝑥
−(sin2 𝑥 + cos 2 𝑥)
−𝑠𝑖𝑛𝑥
=
−(1)
= 𝑠𝑖𝑛𝑥
Similarly
𝑠𝑖𝑛𝑥 0
| |
𝑣2′ (𝑥) = 𝑐𝑜𝑠𝑥 𝑡𝑎𝑛𝑥
𝑠𝑖𝑛𝑥 𝑐𝑜𝑠𝑥
| |
𝑐𝑜𝑠𝑥 −𝑠𝑖𝑛𝑥

𝑠𝑖𝑛𝑥 𝑡𝑎𝑛𝑥 − 0
=
− sin2 𝑥 − cos 2 𝑥

𝑠𝑖𝑛𝑥
𝑠𝑖𝑛𝑥
= 𝑐𝑜𝑠𝑥
−(sin2 𝑥 + cos 2 𝑥)
sin2 𝑥
= 𝑐𝑜𝑠𝑥
−(1)

1 − cos2 𝑥
= −
cos 𝑥

1 cos 2 𝑥
=− +
cos 𝑥 𝑐𝑜𝑠𝑥
= 𝑐𝑜𝑠𝑥 − 𝑠𝑒𝑐𝑥
therefore,
𝑣1 (𝑥) = ∫ 𝑠𝑖𝑛𝑥. 𝑑𝑥 = −𝑐𝑜𝑠𝑥 + 𝐶3

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

21

𝑎𝑛𝑑 𝑣2 (𝑥) = ∫ (𝑐𝑜𝑠𝑥 − 𝑠𝑒𝑐𝑥)𝑑𝑥 = 𝑠𝑖𝑛𝑥 − log|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥| + 𝐶4

Substituting the values of 𝑣1 (𝑥) 𝑎𝑛𝑑 𝑣2 (𝑥) 𝑖𝑛 𝑒𝑞. (2)
𝑦𝑝 (𝑥) = (−𝑐𝑜𝑠𝑥 + 𝐶3 )𝑠𝑖𝑛𝑥 + (𝑠𝑖𝑛𝑥 − log|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥| + 𝐶4 )𝑐𝑜𝑠𝑥

= −𝑠𝑖𝑛𝑥 𝑐𝑜𝑠𝑥 + 𝐶3 𝑠𝑖𝑛𝑥 + 𝑠𝑖𝑛𝑥 𝑐𝑜𝑠𝑥 − 𝑐𝑜𝑠𝑥 log | 𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥| + 𝐶4 𝑐𝑜𝑠𝑥
= 𝐶3 𝑠𝑖𝑛𝑥 + 𝐶4 𝑐𝑜𝑠𝑥 − 𝑐𝑜𝑠𝑥 log|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥|
Since a particular integral is a solution free of arbitrary constants , we may assume any particular
values A and B to 𝐶3 𝑎𝑛𝑑 𝐶4 respectively.
Then the particular integral
𝑦𝑝 = 𝐴𝑠𝑖𝑛𝑥 + 𝐵𝑐𝑜𝑠𝑥 − 𝑐𝑜𝑠𝑥 log|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥|

Thus, 𝑦 = 𝑦𝑐 + 𝑦𝑝

= 𝐶1 𝑆𝑖𝑛𝑥 + 𝐶2 𝐶𝑜𝑠 𝑥 + 𝐴𝑠𝑖𝑛𝑥 + 𝐵𝑐𝑜𝑠𝑥 − 𝑐𝑜𝑠𝑥 log|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥|

= (𝐶1 + 𝐴)𝑠𝑖𝑛𝑥 + (𝐶2 + 𝐵)𝑐𝑜𝑠𝑥 − 𝑐𝑜𝑠𝑥(log|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥|)
= 𝐶1′ 𝑠𝑖𝑛𝑥 + 𝐶2′ cos 𝑥 − cos(𝑙𝑜𝑔|𝑠𝑒𝑐𝑥 + 𝑡𝑎𝑛𝑥|)
𝑤ℎ𝑒𝑟𝑒 𝐶1′ = 𝐶1 + 𝐴 𝑎𝑛𝑑 𝐶2′ = 𝐶2 + 𝐵

Change of independent variable: -

Consider the differential eq.
𝑑2𝑦 𝑑𝑦
+ 𝑃 + 𝑄𝑦 = 𝑅 (1)
𝑑𝑥 2 𝑑𝑥
In the above eq. (1) we can change independent variable x to z by relation z= f(x)
So,
𝑑2𝑦 𝑑𝑦
+ 𝑃1 + 𝑄1 𝑦 = 𝑅1
𝑑𝑧 2 𝑑𝑧
where
𝑑2𝑧 𝑑𝑧
𝑑𝑥 2 + 𝑃 𝑑𝑥 𝑄 𝑅
𝑃1 = , 𝑄1 = , 𝑅1 =
𝑑𝑧 2 𝑑𝑧 2
𝑑𝑧 2
( ) ( ) ( )
𝑑𝑥 𝑑𝑥 𝑑𝑥

e.g.

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

22

𝑑2 𝑦 𝑑𝑦
𝑥 2− − 4𝑥 3 𝑦 = 4𝑥 3 𝑠𝑖𝑛𝑥 2 (1)
𝑑𝑥 𝑑𝑥
𝑑2 𝑦 1 𝑑𝑦
− − 4𝑥 2 𝑦 = 4𝑥 2 𝑠𝑖𝑛𝑥 2
𝑑𝑥 2 𝑥 𝑑𝑥
Where
1
𝑃= − , 𝑄 = −4𝑥 2 , 𝑅 = 4𝑥 2 𝑠𝑖𝑛𝑥 2
𝑥
Change the independent variable x to z
𝑑2𝑦 𝑑𝑦
2
+ 𝑃1 + 𝑄1 𝑦 = 𝑅1 (2)
𝑑𝑧 𝑑𝑧
Let
𝑄 𝑑𝑧 2 𝑑𝑧
2
𝑥2
𝑄1 = = −4 ⇒ −4𝑥 = 4 ( ) ⇒ = 𝑥 ⇒ 𝑑𝑧 = 𝑥𝑑𝑥 ⇒ 𝑧 =
𝑑𝑧 2 𝑑𝑥 𝑑𝑥 2
( )
𝑑𝑥
In this we generally assume the value of 𝑄1 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 𝑟𝑒𝑙𝑎𝑡𝑒𝑑 𝑡𝑜 𝑄
Now
1
1 + (− ) 𝑥 4𝑥 2 𝑠𝑖𝑛𝑥 2
𝑃1 = 𝑥 = 0 , 𝑄 = −4 , 𝑅 = = 4 sin 2𝑧
1 1
𝑥2 𝑥2
Therefore eq. (2) becomes
𝑑2𝑦
− 4𝑦 = 4 𝑠𝑖𝑛2𝑧
𝑑𝑧 2
(𝐷2 − 4)𝑦 = 4𝑠𝑖𝑛2𝑧
The auxiliary eq. is
𝑚2 − 4 = 0
𝑚 = 2 , −2
Therefore, 𝐶. 𝐹 = 𝐶1 𝑒 2𝑧 + 𝐶2 𝑒 −2𝑧
𝑠𝑖𝑛𝑧 4𝑠𝑖𝑛2𝑧 4𝑠𝑖𝑛2𝑧 𝑠𝑖𝑛2𝑧
𝑃. 𝐼 = 𝑦 = 4 = = =
𝐷2 − 4 −4 − 4 −8 −2
Therefore
𝑌 = 𝐶. 𝐹 + 𝑃. 𝐼

2𝑍 −2𝑧
𝑠𝑖𝑛2𝑧 𝑥2 −𝑥 2
𝑠𝑖𝑛𝑥 2
𝑌 = 𝐶1 𝑒 + 𝐶2 𝑒 − = 𝐶1 𝑒 + 𝐶2 𝑒 −
2 2

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

23

Solutions of Ordinary Differential Equation of First order and First degree by

Numerical analysis

(i) Application of Laplace transformation to the ordinary differential equation with

constant coefficient.
(ii) Solution of ordinary differential equation of first order by Picard’s method of
successive approximations.
(iii) Euler method or Runge kutta method of 1st order.
(iv) Euler modified method or Runge kutta method of 2nd order.
(v) Milne predicator and corrector method.
(vi) Adam Bashforrth Predicator and Corrector method.
Examples
𝑑2 𝑦 𝑑𝑦
(1) Using Laplace transform, determine the solution of ( ) + 3 ( ) + 2𝑦 = 𝑒 −𝑡
𝑑𝑡 2 𝑑𝑡

𝑦(0) = 𝑦 ′ (0) = 0
Sol: - Re-writing the given differential equation and condition, we have
𝑦 ′′ + 3𝑦 ′ + 2𝑦 = 𝑒 −𝑡 (1)
with initial conditions: 𝑦(0) = 0 𝑎𝑛𝑑 𝑦 ′ (0) = 0 (2)
Taking Laplace transform of both sides of (1), we have
𝐿 {𝑦 ′′ } + 3𝐿 {𝑦 ′ } + 2𝐿 {𝑦} = 𝐿{𝑒 −𝑡 }
1
𝑠 2 𝐿 {𝑦} − 𝑠𝑦(0) − 𝑦 ′ (0) + 3 {𝑠𝐿 {𝑦} − 𝑦(0)} + 2𝐿{𝑦} =
𝑠+1
1
(𝑠 2 + 3𝑠 + 2) 𝐿{𝑦} = , 𝑢𝑠𝑖𝑛𝑔 (2)
𝑠+1
or,
1 1 1 1 1
𝐿 {𝑦} = = = − + (3)
(𝑠 2 + 3𝑠 + 2)(𝑠 + 1) (𝑠 + 2)(𝑠 + 1)2 𝑠 + 2 𝑠 + 1 (𝑠 + 1)2
[On resolving into partial fraction]
Taking inverse transform of both sides of (3) , we get
1 1 1
𝑦 = 𝐿−1 { } − 𝐿−1 { } + 𝐿−1 { }
𝑠+2 𝑠+1 (𝑠 + 1)2
1
= 𝑒 −2𝑡 − 𝑒 −𝑡 − 𝑒 −𝑡 𝐿−1 { } = 𝑒 −2𝑡 − 𝑒 −𝑡 − 𝑒 −𝑡 × 𝑡
𝑠2
(2) Find the third approximation in the following differential equation

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

24

𝑑𝑦
= 1 + 𝑦 2 , 𝑦(0) = 0 𝑥0 = 0 , 𝑦0 = 0
𝑑𝑥
Also , find the exact solution
Sol :- Let us choose zeroth approximation
𝜑0 = 0 , 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥
Also ,
𝑑𝑦
= 𝑓(𝑥, 𝑦) = 1 − 𝑦 2
𝑑𝑥
Now , The 𝑛𝑡ℎ approximation is given 𝜑𝑛 , 𝑛 ≥ 1 is given by the formula
𝑥
𝜑𝑛 (𝑥) = 𝑦0 + ∫ 𝑓 [𝑡 , 𝜑𝑛−1 (𝑡) ] 𝑑𝑡
𝑥0

Therefore ,
𝑥
𝜑1 (𝑥) = 0 + ∫ 𝑓 [𝑡 , 𝜑0 (𝑡) ] 𝑑𝑡
𝑥0
𝑥
= ∫ 𝑓 [𝑡 ,0 ] 𝑑𝑡
0
𝑥
= ∫ 𝑓 [1 + 02 ] 𝑑𝑡
0
𝑥
= ∫ 1. . 𝑑𝑡
0

= [𝑡]0𝑥
=𝑥
𝑥
𝜑2 (𝑥) = 0 + ∫ 𝑓[𝑡 , 𝜑1 (𝑡) ] 𝑑𝑡
𝑥0
𝑥
= ∫ 𝑓 [𝑡 , 𝑡 ] 𝑑𝑡
0
𝑥
= ∫ 𝑓 [1 + 𝑡 2 ] 𝑑𝑡
0
𝑥
𝑡3
=[𝑡+ ]
3 0

𝑥3
=𝑥+
3

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

25

𝑥
𝜑3 (𝑥) = 0 + ∫ 𝑓[𝑡 , 𝜑2 (𝑡) ] 𝑑𝑡
𝑥0

𝑥
𝑡3
= ∫ 𝑓 [ 𝑡, 𝑡 + ] 𝑑𝑡
0 3
𝑥 2
𝑡3
= ∫ [1 + (𝑡 + ) ] 𝑑𝑡
0 3
𝑥
𝑡 6 2𝑡 9
= ∫ [1+ 𝑡 2 + + ] 𝑑𝑡
0 9 3
𝑥
𝑡 3 𝑡 7 2𝑡 5
= [𝑡 + + + ]
3 9.7 3.5 0

𝑥 3 𝑥 7 2𝑥 5
=𝑥+ + +
3 63 15
Now exact solution,
𝑑𝑦
= 1 + 𝑦2
𝑑𝑥
𝑑𝑦
= 𝑑𝑥
1 + 𝑦2
On integrating both sides, we get
𝑑𝑦
∫ = ∫ 𝑑𝑥
1 + 𝑦2

tan−1 𝑦 = 𝑥 + 𝑐
Where c is an arbitrary constant to be determined
Using the initial condition y (0) =0, we have
C=0
Therefore, tan−1 𝑦 = 𝑥 ⇒ 𝑦 = 𝑡𝑎𝑛𝑥
But we know that the series for tan x is
𝑥3 2 5
𝑡𝑎𝑛𝑥 = 𝑥 + + 𝑥 + . . . . . . . . .
3 15
𝑑𝑦
=𝑥+𝑦
𝑑𝑥

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

26

ℎ
𝑌1 = 𝑌0 + [ 𝑓(𝑥0 , 𝑦0 ) + 𝑓(𝑥1 , 𝑌1∗ )]
2
ℎ
𝑌1 = 𝑌0 + [ (𝑥02 + 𝑦0 ) + (𝑥12 + 𝑌1∗ )]
2
02
0.02
𝑌1 = 1 + [ (02 + 1) + (0.02)2 + 1.02]
2
𝑌1 = 1.0202
Put n = 1 in eq. (2)
𝑌2∗ = 𝑌1 + ℎ𝑓(𝑥1 , 𝑦1 )
𝑌2∗ = 𝑌1 + ℎ(𝑥12 + 𝑦1 ) = 1.0202 + 0.02[ (0.02)2 + 1.0202] = 1.0406
Again Put n = 1 in eq. (1)
ℎ
𝑌2 = 𝑌1 + [ 𝑓(𝑥1 , 𝑦1 ) + 𝑓(𝑥2 , 𝑌2∗ )]
2
ℎ
𝑌2 = 𝑌1 + [ (𝑥12 + 𝑦1 ) + (𝑥22 + 𝑌2∗ )]
2
0.02
𝑌2 = 1.0202 + (0.02)2 + 1.0202 + (0.04)2 + 1.0406
2
𝑌2 = 1.0408

Power Series:
Differential equation possesses solutions expressible in terms of elementary functions such as
polynomials, exponential, cosine, sine. There are many methods existing for finding such solutions
that we discussed above also, for example the linear first order differential equation y ‘=
y/x is easily solved by the way of “separation of variable “. This method yields the family of
elementary solutions 𝑦 = 𝑐𝑥, c is arbitrary constant. In general, however, higher order linear
equation has no solutions that can be expressed in such a simpler manner. Thus, we must seek
other means of expression for the solutions of these equations. One such means of expression is
furnished by infinite series representation.
Consider the second-order homogenous linear differential equation
𝑑2 𝑦 𝑑𝑦
𝑎0 (𝑥) 2
+ 𝑎1 (𝑥) + 𝑎2 (𝑥)𝑦 = 0 (1)
𝑑𝑥 𝑑𝑥

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

27

And suppose that this equation has no solution that is expressible as finite linear combination of
known elementary functions. Let us assume, however, that it does have a solution that in the form
of an infinite series. Specially, we assume that it has a solution expressible in the form
∞
)2
𝑐0 + 𝑐1 (𝑥 − 𝑥0 ) + 𝑐2 (𝑥 − 𝑥0 + . . . .. . . . . . 𝑓(𝑧) = ∑(𝑥 − 𝑥0 )𝑛 (2)
𝑛=0
where , 𝑐0 , 𝑐1 , 𝑐3 . . . . . . . .. are constants. An expression of the form in
equation (2) is called a power series in 𝑥 − 𝑥0 . We have thus assumed that the differential equation
(1) has a solution called power series of the form (2). Assuming that this assumptions is valid, we
can proceed to determine the coefficient 𝑐0 , 𝑐1 , 𝑐2 . . . . . ..in (2) in such a manner that the expression
does indeed satisfy the equation(1).
We generally have three following three types of power series solution.
(i) Taylor series Method. With the help of initial condition and the given differential
equations, we obtain the Taylor series containing only a few terms as the power series
solutions. In this method, we cannot obtain either the recurrence relation among the
coefficients or the general formula for the n-th term of the solution.
(ii) Power series Method. We assume the solution of the given equation as a power series
and express the solution in terms of an arbitrary constant which may be evaluated by
using the initial conditions. This method is applicable when we solve the equation at
an ordinary point. We shall illustrate this methods by using Hermite equation x’’ -2tx’
+ 2x = 0 at t=0 and the Legendre equation (1 − 𝑡 2 )x’’ – 2tx’ + p(p+1) x =0, where p is
real number.
(iii) Frobenius Method.The Frobenius method gives the power series solutions of
differential equation at regular singular points. In this method we obtain an equation
called indicial equation. Depending upon the roots of the indicial equation, we obtain
the general solution. This method is best illustrated with the Bessel equation of order
p given by 𝑡 2 𝑥 ′′ + 𝑡𝑥 ′ + (𝑡 2 − 𝑝2 )𝑥 = 0 where p is a positive constant.

Example 1: -Find the Power series solution of the differential equation

𝑑2 𝑦 𝑑𝑦
(𝑥 2 − 1) 2
+ 3𝑥 + 𝑥𝑦 = 0 , 𝑦(0) = 4 𝑎𝑛𝑑 𝑦 ′ (0) = 6
𝑑𝑥 𝑑𝑥
Sol:- Let
𝑑2𝑦 𝑑𝑦
(𝑥 2 − 1) 2
+ 3𝑥 + 𝑥𝑦 = 0 (1)
𝑑𝑥 𝑑𝑥
𝑦(0) = 4 (2)
𝑦 ′ (0) = 6 (3)
We first observe that all the points except 𝑥 = ±1 are the ordinary points of the differential
equation (1).

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

28

We shall choose 𝑥0 = 0 and seek solution in the powers of x

Thus , we, may assume
∞

𝑦 = ∑ 𝐶𝑛 𝑥 𝑛 (4)
𝑛=0

Diff. eq. (4) w.r.t ‘x’ term by term , we have

∞
𝑑𝑦
= ∑ 𝑛 𝐶𝑛 𝑥 𝑛−1 (5)
𝑑𝑥
𝑛=1

Also,
∞
𝑑2𝑦
2
= ∑ 𝑛 (𝑛 − 1)𝐶𝑛 𝑥 𝑛−2 (6)
𝑑𝑥
𝑛=2

Using eq. (4) , (5) , (6) in eq. (1) , we have

∞ ∞ ∞

(𝑥 2 − 1) ∑ 𝑛 (𝑛 − 1)𝐶𝑛 𝑥 𝑛−2
+ 3𝑥 ∑ 𝑛 𝐶𝑛 𝑥 𝑛−1
+ 𝑥 ∑ 𝐶𝑛 𝑥 𝑛 = 0
𝑛=2 𝑛=1 𝑛=0
∞ ∞ ∞ ∞
𝑛 𝑛−2
⇒ ∑ 𝑛 (𝑛 − 1)𝐶𝑛 𝑥 − ∑ 𝑛 (𝑛 − 1)𝐶𝑛 𝑥 + 3 ∑ 𝑛 𝐶𝑛 𝑥 + ∑ 𝐶𝑛 𝑥 𝑛+1 = 0
𝑛

𝑛=2 𝑛=2 𝑛=1 𝑛=0

∞ ∞ ∞ ∞

⇒ ∑ 𝑛 (𝑛 − 1)𝐶𝑛 𝑥 − ∑(𝑛 + 2)(𝑛 + 1)𝐶𝑛+2 𝑥 + 3 ∑ 𝑛 𝐶𝑛 𝑥 + ∑ 𝐶𝑛−1 𝑥 𝑛 = 0

𝑛 𝑛 𝑛

𝑛=2 𝑛=0 𝑛=1 𝑛=1

∞ ∞ ∞

⇒ ∑ 𝑛 (𝑛 − 1)𝐶𝑛 𝑥 − 2𝐶2 − 6𝐶3 𝑥 − ∑(𝑛 + 2) (𝑛 + 1)𝐶𝑛+2 𝑥 + 3𝐶1 𝑥 + ∑ 3𝑛 𝐶𝑛 𝑥 𝑛

𝑛 𝑛

𝑛=2 𝑛=2 𝑛=2

∞

+ 𝐶0 𝑥 + ∑ 𝐶𝑛−1 𝑥 𝑛 = 0
𝑛=2
∞

⇒ −2𝐶2 + ( 𝐶0 + 3 𝐶1 − 6 𝐶3 ) 𝑥 + ∑[ 𝑛(𝑛 + 2)𝐶𝑛 − (𝑛 + 2)(𝑛 + 1)𝐶𝑛+2 + 𝐶𝑛−1 ]𝑥 𝑛 = 0

𝑛=2
. . . . . . . . . . (7)
Equating to zero the coefficient of each power of x in eq. (7) , w get
−2𝐶2 = 0 (8)
𝐶0 + 3𝐶1 − 6𝐶3 = 0 (9)
𝑎𝑛𝑑 𝑛(𝑛 + 2)𝐶𝑛 − (𝑛 + 2)(𝑛 + 1)𝐶𝑛+2 + 𝐶𝑛−1 = 0 , 𝑛 ≥ 2 (10)

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

29

From eq.(8) , we have

𝐶2 = 0
From eq. (9) , we have
𝐶0 + 3𝐶1 = 6𝐶3
1 1
𝐶3 = 𝐶0 + 𝐶1
6 2
From eq. (10) , we have
𝑛(𝑛 + 2)𝐶𝑛 + 𝐶𝑛−1 = (𝑛 + 2)(𝑛 + 1)𝐶𝑛+2 , 𝑛 ≥ 2
𝑛(𝑛 + 2)𝐶𝑛 + 𝐶𝑛−1
𝐶𝑛+2 = , 𝑛≥2
(𝑛 + 2)(𝑛 + 1)
Using this, we find successively,
2 × 4𝐶2 + 𝐶1
𝐶4 =
4×3
8𝐶2 + 𝐶1
=
12
1
= 𝐶
12 1
And , Similarly
(3 × 5𝐶3 + 𝐶2 ) 15𝐶3 + 𝐶2
𝐶5 = =
5×4 20
3 3 1 1
= 𝐶3 = [ 𝐶0 + 𝐶1 ]
4 4 6 2
1 3
= 𝐶0 + 𝐶1 𝑎𝑛𝑑 𝑠𝑜 𝑜𝑛
8 8
Substituting these values of 𝐶2 , 𝐶3 , 𝐶4 , 𝐶5 . . . . in eq. (4) , we get
𝑦 = 𝐶0 + 𝐶1 𝑥 + 𝐶2 𝑥 2 + 𝐶3 𝑥 3 + 𝐶4 𝑥 4 + 𝐶5 𝑥 5 + . . . . . .
1 1 1 1 3
= 𝐶0 + 𝐶1 𝑥 + ( 𝐶0 + 𝐶1 ) 𝑥 3 + 𝐶1 𝑥 4 + ( 𝐶0 + 𝐶1 ) 𝑥 5 + . . . .
6 2 12 8 8
1 1 1 1 4 3 5
= 𝐶0 ( 1 + 𝑥 3 + 𝑥 5 + . . ) + 𝐶1 ( 𝑥 + 𝑥 3 + 𝑥 + 𝑥 + . .) (11)
6 8 2 12 8
Now, Applying initial condition (2) in eq. (11) , we get
C0 = 4
Diff. eq. (11) w.r.t. ‘x’ , we have

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

30

1 1 1
𝑦 ′ (𝑥) = 𝐶0 ( × 3𝑥 2 + × 5𝑥 4 + . . . ) + 𝐶1 ( 1 + × 3𝑥 2 + . . . ) (12)
6 8 2
Applying the initial condition 𝑦 ′ (0) = 6 in eq. (12), we get
𝐶1 = 6
Thus the solution of the given initial value problem is given as
1 1 1 1 4 3 5
𝑦(𝑥) = 4 ( 1 + 𝑥 3 + 𝑥 5 + . . ) + 6 ( 𝑥 + 𝑥 3 + 𝑥 + 𝑥 + . .)
6 8 2 12 8
11 3 1 4 11 5
= 4 + 6𝑥 + 𝑥 + 𝑥 + 𝑥 + . . . . .
3 2 5

Example 2: - Use the method of Frobenius to find the solution of the differential equation
𝑑2 𝑦 𝑑𝑦
2𝑥 2 +𝑥 + (𝑥 2 − 3)𝑦 = 0 𝑖𝑛 𝑠𝑜𝑚𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 0 < 𝑥 < 𝑅 (1)
𝑑𝑥 2 𝑑𝑥

Sol:- Here 𝑥0 = 0 is a regular point of the differential equation (1)

So, by the method of Frobenius,
We assume the solution of eq. (1) to be of the form
∞

𝑦 = ∑ 𝐶𝑛 (𝑥)𝑛+𝑟 , 𝐶0 ≠ 0 (2)
𝑛=0

Differentiate eq. (2) w.r.t. ‘x’ , we get

∞
𝑑𝑦
= ∑(𝑛 + 𝑟) 𝐶𝑛 (𝑥)𝑛+𝑟−1 (3)
𝑑𝑥
𝑛=0

Again diff. the above , we get

∞
𝑑2 𝑦
= ∑(𝑛 + 𝑟) (𝑛 + 𝑟 − 1)𝐶𝑛 (𝑥)𝑛+𝑟−2 (4)
𝑑𝑥 2
𝑛=0

𝑑𝑦 𝑑2𝑦
Substituting the values of 𝑦, , in eq. (1) , we get
𝑑𝑥 𝑑𝑥 2
∞ ∞ ∞
2 (𝑥)𝑛+𝑟−2 (𝑥)𝑛+𝑟−1 (𝑥 2
2𝑥 ∑(𝑛 + 𝑟) (𝑛 + 𝑟 − 1)𝐶𝑛 + 𝑥 ∑(𝑛 + 𝑟) 𝐶𝑛 + − 3) ∑ 𝐶𝑛 (𝑥)𝑛+𝑟 = 0
𝑛=0 𝑛=0 𝑛=0

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

31

Chapter 3

While obtaining the power series solutions of some special type of differential equation, we come
across functions of Legendre, Hermite, Laguerre, Bessel. We call these functions as functions of
differential equations. Sometimes these are also known as special functions. The special functions
which arise as the solution of the second order linear differential equations have considerable
degree of unity in the sense that they have properties similar to each other and their wide
applications are more or less same.

Legendre’s Differential equation:

Legendre differential equation is the second order ordinary equation which can be written as
𝑑2 𝑦 𝑑𝑦
(1 − 𝑥 2 ) 2
− 2𝑥 + 𝑛(𝑛 + 1)𝑦 = 0
𝑑𝑥 𝑑𝑥
where n is constant
or, it may also be written as

𝑑 𝑑𝑦
[1 − 𝑥 2 ) ] + 𝑛(𝑛 + 1) = 0
𝑑𝑥 𝑑𝑥

Power series solution of Legendre’s equation

The Legendre’s differential equation is
𝑑2 𝑦 𝑑𝑦
(1 − 𝑥 2 ) 2
− 2𝑥 + 𝑛(𝑛 + 1)𝑦 = 0 (1)
𝑑𝑥 𝑑𝑥
where n is constant
The equivalent normalized form of equation (1) is
𝑑2 𝑦 𝑑𝑦
2
+ 𝑃𝑥 + 𝑄(𝑥)𝑦 = 0 (2)
𝑑𝑥 𝑑𝑥
Where
2𝑥
𝑃(𝑥) = −
1 − 𝑥2

𝑛(𝑛 + 1)
𝑄(𝑥) = (3)
1 − 𝑥2

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

32

Here both P(x) & Q(x) are analytic at 𝑥0 = 0. The origin is therefore an arbitrary point. So, we
can assume the solution of equation to be of the form
∞

𝑦 = ∑ 𝐶𝑟 𝑥 𝑟 (4)
𝑟=0

Diff. equation (4) term by term w.r.t. ‘x’ . we get

∞
𝑑𝑦
= ∑ 𝑟𝐶𝑟 𝑥 𝑟−1
𝑑𝑥
𝑟=1
∞
𝑑2 𝑦
= ∑ 𝑟(𝑟 − 1)𝐶𝑟 𝑥 𝑟−2
𝑑𝑥 2
𝑟=2

𝑑𝑦 𝑑2 𝑦
Substituting the values of 𝑦, , 𝑖𝑛 𝑒𝑞, (1), we get
𝑑𝑥 𝑑𝑥 2
∞ ∞ ∞
2) 𝑟−2 𝑟−1
(1 − 𝑥 ∑ 𝑟(𝑟 − 1) 𝐶𝑟 𝑥 − 2𝑥 ∑ 𝑟 𝐶𝑟 𝑥 + 𝑛(𝑛 + 1) ∑ 𝐶𝑟 𝑥 𝑟 = 0
𝑟=2 𝑟=1 𝑟=0
∞ ∞ ∞ ∞
𝑟−2
⇒ ∑ 𝑟(𝑟 − 1) 𝐶𝑟 𝑥 𝑟
− ∑ 𝑟(𝑟 − 1)𝐶𝑟 𝑥 − 2 ∑ 𝑟 𝐶𝑟 𝑥 + 𝑛(𝑛 + 1) ∑ 𝐶𝑟 𝑥 𝑟 = 0 𝑟

𝑟=2 𝑟=2 𝑟=1 𝑟=0

∞ ∞ ∞ ∞

⇒ 𝑟
∑(𝑟 + 2)(𝑟 + 1)) 𝐶𝑟+2 𝑥 − ∑ 𝑟(𝑟 − 1)𝐶𝑟 𝑥 − 2 ∑ 𝑟 𝐶𝑟 𝑥 + 𝑛(𝑛 + 1) ∑ 𝐶𝑟 𝑥 𝑟 = 0 𝑟 𝑟

𝑟=0 𝑟=2 𝑟=1 𝑟=0

∞

⇒ 2𝐶2 + 6𝐶3 𝑥 + ∑(𝑟 + 2)(𝑟 + 1)) 𝐶𝑟+2 𝑥 𝑟

𝑟=2
∞ ∞ ∞

− ∑(𝑟 2 − 𝑟)𝐶𝑟 𝑥𝑟 − 2𝐶1 𝑥 − 2 ∑ 𝑟 𝐶𝑟 𝑥𝑟 + 𝑛(𝑛 + 1)𝐶0 + 𝑛(𝑛 + 1)𝐶1 𝑥 + 𝑛(𝑛 + 1) ∑ 𝐶𝑟 𝑥 𝑟 = 0

𝑟=2 𝑟=2 𝑟=2

⇒ 2𝐶2 + 𝑛(𝑛 + 1)𝐶0 + [ 6𝐶3 + {𝑛(𝑛 + 1) − 2}𝐶1 ]𝑥

∞

+ ∑{(𝑟 + 2)(𝑟 + 1)𝐶𝑟+2 + 𝑛(𝑛 + 1) − 2𝑟 − (𝑟 2 − 𝑟)]𝐶𝑟 }𝑥 𝑟 = 0 (5)

𝑟=2

Equating like power, we get

2𝐶2 + 𝑛(𝑛 + 1)𝐶0 = 0 (6)
6𝐶3 + [ 𝑛(𝑛 + 1) − 2 ]𝐶1 = 0 (7)

𝑎𝑛𝑑 (𝑟 + 2)(𝑟 + 1)𝐶𝑟+2 = (𝑟 2 − 𝑛2 + 𝑟 − 𝑛 )𝐶𝑟 , 𝑟 ≥ 2

(𝑟 − 𝑛)(𝑟 + 𝑛 + 1)
𝐶𝑟+2 = 𝐶𝑟 𝑟≥2 (8)
(𝑟 + 2)(𝑟 + 1)

From eq. (6) , we get

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

33

𝑛(𝑛 + 1)
2𝐶2 = −𝑛(𝑛 + 1)𝐶0 ⇒ 𝐶2 = − 𝐶0
2!
From eq. (7) we get
6𝐶3 = −[𝑛(𝑛 + 1) − 2]𝐶1
(𝑛 − 1)(𝑛 + 2)
𝐶3 = − 𝐶1 (10)
3!
Putting r = 2,3,4 . . . . . . in eq. (8) , we get
(2 − 𝑛)(𝑛 + 3)
𝐶4 = 𝐶1
4.3
𝑛(𝑛 + 1)(2 − 𝑛)(𝑛 + 3)
=− 𝐶0
4!
𝑛(𝑛 + 1)(𝑛 − 2)(𝑛 + 3)
𝐶0
4!
(𝑛 − 1)(𝑛 + 2)(𝑛 − 3)(𝑛 + 4)
𝐶5 = 𝐶1
5!
By inserting these coefficient into the assume solution (4) , we get
𝑦 = 𝐶0 + 𝐶1 𝑥 + 𝐶2 𝑥 2 + 𝐶3 𝑥 3 + 𝐶4 𝑥 4 + . . . . . . . . . . . . . . ..
𝑛(𝑛 + 1) (𝑛 − 1)(𝑛 + 2) 𝑛(𝑛 + 1)(𝑛 − 2)(𝑛 + 3)
𝑦 = 𝐶0 + 𝐶1 𝑥 − 𝐶0 𝑥 2 − 𝐶1 𝑥 3 + 𝐶0 𝑥 4
2! 3! 4!
(𝑛 − 1)(𝑛 + 2)(𝑛 − 3)(𝑛 + 4)
+ 𝐶1 𝑥 5 + . . . . . . . . ..
5!
𝑛(𝑛 + 1) 2 𝑛(𝑛 + 1)(𝑛 − 2)(𝑛 + 3) 4
𝑦 = 𝐶0 [ 1 − 𝑥 + 𝑥 . . . .]
2! 4!

𝑛(𝑛 − 1)(𝑛 + 2) 3 (𝑛 − 1)(𝑛 + 2)(𝑛 − 3)(𝑛 + 4) 5

+ 𝐶1 [𝑥 − 𝑥 + 𝑥 + .........]
3! 5!

which gives the solution of the differential eq. (1).

Legendre’s Function
∞

If ∑ ak 𝑥 2 is a power series solution of the legendre equation

(k=0)

(1 − 𝑡 2 )𝑥 ′′ − 2𝑡𝑥 ′ + 𝑝(𝑝 + 1)𝑥 =

0, where p is a real constant (1)

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

34

Then, we have the following recurrence formula;

(p − k)(p + k + 1)
ak+2 = − ak (2)
(k + 1)(k + 2)
and the solution 𝑥(𝑡) = 𝑎0 𝑥1 (𝑡) + 𝑎1 𝑥2 (𝑡)
p(p+1) 2 p(p+1)(P−2)(p+3) 4 p(p+1)(p−2)(p+3)(p−4)(p+5) 6
where, x1 (t) = a0 [1 − t + t − t . .] (3)
2! 4! 6!
(p−1)(p+2) 3 (𝑝−1)(𝑝+2)(𝑝−3)(𝑝+4) 5 (𝑝−1)(𝑝+2)(𝑝−3)(𝑝+4)(𝑝−5)(𝑝+6)
𝑥2 (t) = 𝑎1 [𝑡 − t + 𝑡 − 𝑡 7 . ] (4)
3! 5! 7!

If p is a non- negative integer, then one of the series terminate so that it is a polynomial in ‘t ’and
the other remains as a series so that the solution is the sum of a polynomial and a series.
Let us first consider a few polynomials for:(i) p=0 ,2 ,4 (ii) p = 1 , 3 ,5
(i) When p=0 then 𝑥1 (𝑡) = 𝑎0
2.3 2
P=2 then 𝑥1 (𝑡) = 𝑎0 [ 1 − 𝑡 ]
2!
4.5 2 4.2.5.7 4
P=4 then 𝑥1 (𝑡) = 𝑎0 [1 − 𝑡 + 𝑡 ]
2! 4!

In all the above, when p is an even integer, 𝑥1 (𝑡) resuces to an even polynomial in t and
𝑥2 (𝑡)remains as an infinite series (4)
(ii) When p=1 then 𝑥2 (𝑡) = 𝑎1 (𝑡)
2.5 3
P=3 then 𝑥2 (𝑡) = 𝑎1 [𝑡 − 𝑡 ]
3!
4.7 4.2.7.9 5
P=5 then 𝑥2 (𝑡) = 𝑎1 [𝑡 − + 𝑡 ]
3! 5!

In all the above cases 𝑥2 (𝑡) reduces to an odd polynomial and 𝑥1 (𝑡) remains as an infinite series
(3). The particular solutions corresponding to positive integral values of p are called Legendre’s
polynomial.
Definition; When p=n , the polynomial solution 𝑃𝑛 of degree n of
(1 − 𝑡 2 )𝑥 ′′ − 2𝑡𝑥 ′ + 𝑛(𝑛 + 1)𝑥 = 0

Satisfy 𝑝𝑛 (1) = 1 is called 𝑛𝑡ℎ 𝑙𝑒𝑔𝑒𝑛𝑑𝑟𝑒 ′ 𝑠 𝑝𝑜𝑙𝑛𝑜𝑚𝑖𝑎𝑙

Theorem: (Rodrigues’s Formula)

The Legendre polynomial 𝑃𝑛 (𝑥) is given as
1 dn 2
𝑃𝑛 (𝑥) = (𝑥 − 1)𝑛
2𝑛 𝑛! dx n
Proof: Let
𝑦 = (𝑥 2 − 1)𝑛

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

35

Therefore,
𝑑𝑦
= 𝑛(𝑥 2 − 1)𝑛−1 (2𝑥)
𝑑𝑥
𝑑𝑦
= 2𝑛𝑥 (𝑥 2 − 1)𝑛 (𝑥 2 − 1)−1
𝑑𝑥
𝑑𝑦 2𝑛𝑥(𝑥 2 − 1)𝑛
=
𝑑𝑥 𝑥2 − 1

𝑑𝑦
⇒ (𝑥 2 − 1) = 2𝑛𝑥𝑦 . . . . . . . . . .(2) [ .̈ 𝑦 = (𝑥 2 − 1)𝑛 ]
𝑑𝑥

Differentiate b/s of equation (2) w.r.t x (n+1) times by Leibnitz theorem, we get
[ .̈ 𝐷𝑛 (𝑢𝑣) = (𝐷 𝑛 𝑢)𝑣+ 𝑛
𝐶1 (𝐷𝑛−1 𝑢)(𝐷𝑣)+ 𝑛 𝐶2 (𝐷𝑛−2 𝑢)(𝐷2 𝑣)+. . . . . ]
𝑑𝑛+2 𝑦 𝑑𝑛+1 𝑦 𝑑𝑛 𝑦 𝑑𝑛+1 𝑦 𝑑𝑛 𝑦
(𝑥 2 − 1) +(𝑛+1) 𝐶1 (2𝑥) +(𝑛+1) 𝐶2 (2) = 2𝑛 [𝑥 +(𝑛+1) 𝐶1 ]
𝑑𝑥 𝑛+2 𝑑𝑥 𝑛+1 𝑑𝑥 𝑛 𝑑𝑥 𝑛+1 𝑑𝑥 𝑛

𝑑𝑛+2 𝑦 𝑑𝑛+1 𝑦 𝑛(𝑛+1) 𝑑𝑛 𝑦 𝑑𝑛+1 𝑦 𝑑𝑛 𝑦

or, (𝑥 2 − 1) + (𝑛 + 1)2𝑥 + 𝑋 2 = 2𝑛𝑥 + (𝑛 + 1)2𝑛
𝑑𝑥 𝑛+2 𝑑𝑥 𝑛+1 2! 𝑑𝑥 𝑛 𝑑𝑥 𝑛+1 𝑑𝑥 𝑛

𝑑𝑛+2 𝑦 𝑑𝑛+1 𝑦 𝑑𝑛 𝑦
or, (𝑥 2 − 1) + (𝑛 + 1 − 𝑛)2𝑥 − 𝑛(𝑛 + 1) =0
𝑑𝑥 𝑛+2 𝑑𝑥 𝑛+1 𝑑𝑥 𝑛

𝑑𝑛+2 𝑦 𝑑𝑛+1 𝑦 𝑑𝑛 𝑦
or, (𝑥 2 − 1) 𝑛+2
+ 2𝑥 𝑛+1 − 𝑛(𝑛 + 1) 𝑑𝑥 𝑛
=0
𝑑𝑥 𝑑𝑥

𝑑𝑛+2 𝑦 𝑑𝑛+1 𝑦 𝑑𝑛 𝑦
or, (1 − 𝑥 2 ) − 2𝑥 + 𝑛(𝑛 + 1) = 0 (3)
𝑑𝑥 𝑛+2 𝑑𝑥 𝑛+1 𝑑𝑥 𝑛

𝑑𝑛
Putting 𝑑𝑥 𝑛 𝑦 = 𝑣 in eq. (3), we get

𝑑2 𝑣 𝑑𝑣
(1 − 𝑥 2 ) − 2𝑥 + 𝑛(𝑛 + 1)𝑣 = 0 (4)
𝑑𝑥 2 𝑑𝑥
Now, eq. (4) is the Legendre’s diff. equation and shows that ‘v’ is a solution of this eq.

Hence ,
Pn (x) = cv
𝑑𝑛
=𝑐 𝑦 (5)
𝑑𝑥 𝑛
where, c is a constant.
To find c, put x=1 in both sides of eq. (5), we get

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

36

𝑑𝑛
𝐶 (𝑑𝑥 𝑛 𝑦) = Pn (1) = 1 (6)
Again, 𝑦 = (𝑥 2 − 1)𝑛 = (𝑥 − 1)𝑛 . (𝑥 + 1)𝑛
Diff. both sides ‘n’ times by Leibnitz theorem, we get
dn dn 𝑑𝑛−1 dn
𝑑𝑥 𝑛
𝑦 = (𝑥 − 1)𝑛 𝑛
(𝑥 + 1)𝑛 + 𝑛 𝑛−1
(𝑥 + 1)𝑛 {𝑛(𝑥 − 1)𝑛−1 }+ . . . +(𝑥 + 1)𝑛 (𝑥 − 1)𝑛
𝑑𝑥 𝑑𝑥 𝑑𝑥 𝑛

. . . . . . .(7)
Now putting x =1 in both sides of eq. (7) , all the terms in R.H.S except the last vanish.
Since each term contains the factor (x -1)
Also,
dn (x − 1)n
= 𝑛!
𝑑𝑥 𝑛
Therefore,
dn
[ y ] = 2n . n! (8)
dx n x=1

Hence, from eq. (6) and (7), we get

C(2n . n!) = 1
1
C=
2𝑛 . 𝑛!
Now, from eq. (5), we get
1 𝑑𝑛
𝑃𝑛 (𝑥) = 𝑦
2𝑛 . 𝑛! 𝑑𝑥 𝑛
1 𝑑𝑛
= 𝑛 𝑛
(𝑥 2 − 1)𝑛
2 . 𝑛! 𝑑𝑥

Generating function of Legendre’s polynomial:

We show that
∞
1
2 ) −2
(1 − 2𝑥𝑡 + 𝑡 = ∑ 𝑃𝑛 (𝑥)𝑡 𝑛 , | 𝑡 | < 1 , −1 ≤ 𝑥 ≤ 1
𝑛=0

where, 𝑝𝑛 (𝑥) is given by the Rodrigues’s formula as :

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

37

1 dn 2
𝑃𝑛 (𝑥) = (𝑥 − 1)𝑛
2𝑛 𝑛! dx n
1 1
Proof : Consider (1 − 2𝑥𝑡 + 𝑡 2 )−2 = [ 1 + ( −2𝑥𝑡 + 𝑡 2 ) ]−2
Using binomial theorem, we have
1 1 1
1 (− )(− −1)
(1 − 2xt + t 2 )−2 = [ 1 + (− ) (−2xt + t 2 ) + 2 2
(−2xt + t 2 )2 +
2 2!
1 1 1 1 1
(− )(− −1)(− −2) (− )(− −1) . . .n−factors .
2 2 2
(−2xt + t 2 )3 + . . . + 2 2
(−2xt + t 2 )n ]
3! n!

1 3 1 3 5
1 (2) ( ) (2) ( ) ( )
2
= 1 + (2xt − t ) + 2 2 2
(2xt − t ) + 2 2 (2xt − t 2 )3 + . . . + 1.3.5 . . . (2n − 1) )
2 2! 3! 2n n!
(2xt − t 2 )n

1 1.3 1.3.5
= 1 + (2xt − t2 ) + 2 (4x 2 t 2 + t 4 − 4xt 3 ) + 3 (8x 3 t 3 − t 6 − 12x 2 t 4
2 2 2! 2 3!
1.3.5 … … … . (2n − 1)
+ 6xt 5 )+ . . . + (2xt − t 2 )n
2n n!
1 1
= 1 + 𝑥𝑡 + (3𝑥 2 − 1)𝑡 2 + (5𝑥 3 − 3𝑥)𝑡 3
2 2
1.3.5 … … … . (2n − 1)
+ . . . . . + (2xt − t 2 )n 𝑡 𝑛
2n n!
@
1 1
∑ 𝑃𝑛 (𝑥)𝑡 𝑛 = 1 + 𝑥𝑡 + (3𝑥 2 − 1)𝑡 2 + (5𝑥 3 − 3𝑥)𝑡 3
2 2
𝑛=0
1.3.5 … … … . (2𝑛 − 1)
+ . . . . . (2𝑥𝑡 − 𝑡 2 )𝑛 𝑡 𝑛
2𝑛 𝑛!
Equating coefficient of 𝑡 0 , 𝑡1 , 𝑡 2 ,. . . . . . . 𝑡 𝑛 ,we get
𝑃0 (𝑥) = 1
𝑝1 (𝑥) = 𝑥
1
𝑝2 (𝑥) = ( 3𝑥 2 − 1)
2
1
𝑝3 (𝑥) = (5𝑥 3 − 3𝑥). . . . . . . . . .
2
1.3.5 … … … . (2𝑛 − 1)
𝑝𝑛 (𝑥) = (2𝑥)𝑛 (∗)
2𝑛 𝑛!
The sum on the R.H.S of the eq. (*) can be shown to be equal to

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

38

1 dn 2
𝑃𝑛 (𝑥) = (𝑥 − 1)𝑛
2𝑛 𝑛! dx n
1
on account of this property, the function of the type (1 − 2𝑥𝑡 + 𝑡 2 )−2 is called as the ‘generating
function’
Some important results;
(1) 𝑃𝑛 (1) = 1
(2) 𝑃𝑛 (−1) = (−1)𝑛
1
(3) 𝑃𝑛′ (1) = (𝑛2 + 𝑛)
2

Proof :-Since we know that

∞
1
2 )−2
(1 − 2𝑥𝑡 + 𝑡 = ∑ Pn (x)t n |t| < 1, −1 ≤ 𝑥 ≤ 1
n=0

(1) Put x=1 , we get

∞
1
∑ 𝑃𝑛 (1) 𝑡 𝑛 = ( 1 − 2𝑡 + 𝑡 2 )−2
𝑛=0
1
= [ (1 − 𝑡)2 ]−2

= (1 − 𝑡)−1
= 1 + 𝑡 + 𝑡 2 + . . . . . . . +𝑡 𝑛

Equating the coefficient of 𝑡 𝑛 on both side , we get

Pn (x) = 1 , for every non − negative integer .
(2) Put x= -1 , we get
∞
1
∑ 𝑃𝑛 (−1) 𝑡 𝑛 = ( 1 + 2𝑡 + 𝑡 2 )−2
𝑛=0
1
= [ (1 + 𝑡)2 ]−2

= (1 + 𝑡)−1

= 1 − 𝑡 + 𝑡 2 + . . . . . 𝑡 𝑛 (−1)𝑛

Equating coefficient of 𝑡 𝑛 on both side , we get

Pn (−1) = (−1)n , for every non − negative integer
(3) Since y = 𝑃𝑛 (𝑥) , 𝑛 = 0,1,2,3,. . . . satisfies Legendre’s equation of order n

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

39

2)
𝑑2 𝑦 𝑑𝑦
(1 − 𝑥 − 2𝑥 + 𝑛(𝑛 + 1)𝑦 = 0
𝑑𝑥 2 𝑑𝑥
(1 − 𝑥 2 ) 𝑃𝑛′′ (𝑥) − 2𝑥 𝑃𝑛′ (𝑥) + 𝑛(𝑛 + 1)𝑃𝑛 (𝑥) = 0
Put x=1, we get
0 − 2𝑃𝑛′ (1) + 𝑛(𝑛 + 1)𝑃𝑛 (1) = 0

−2𝑃𝑛′ (1) = −𝑛(𝑛 + 1)𝑃𝑛 (1)

𝑛(𝑛 + 1)
𝑃𝑛 (1) =
2
Theorem; - (Orthogonality property). If are the Legendre’s polynomial, then
0 𝑖𝑓 𝑚 ≠ 𝑛
1
∫ 𝑃𝑚 (𝑥)𝑃𝑛 (𝑥) 𝑑𝑥 = { 2
−1 𝑖𝑓 𝑚 = 𝑛
2𝑛 + 1
. . . . . . . (1)

Proof ;- Since 𝑃𝑚 (𝑥) 𝑎𝑛𝑑 𝑃𝑛 (𝑥) 𝑠𝑎𝑡𝑖𝑠𝑓𝑖𝑒𝑠 𝑡ℎ𝑒 𝐿𝑒𝑔𝑒𝑛𝑑𝑟𝑒 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛

𝑑2 𝑦 𝑑𝑦
(1 − 𝑥 2 ) 2
− 2𝑥 + 𝑛(𝑛 + 1)𝑦 = 0
𝑑𝑥 𝑑𝑥
Therefore
𝑑2 𝑑
(1 − 𝑥 2 ) 𝑃𝑚 (𝑥) − 2𝑥 𝑃 (𝑥) + 𝑚(𝑚 + 1)𝑃𝑚 (𝑥) = 0
𝑑𝑥 2 𝑑𝑥 𝑚
and,
𝑑2 𝑑
(1 − 𝑥 2 ) 𝑃𝑛 (𝑥) − 2𝑥 𝑃 (𝑥) + 𝑛(𝑛 + 1)𝑃𝑛 (𝑥) = 0
𝑑𝑥 2 𝑑𝑥 𝑛
We can rewrite the above equation as
𝑑 𝑑
[(1 − 𝑥 2 ) 𝑃𝑚 (𝑥)] + 𝑚(𝑚 + 1)𝑃𝑚 (𝑥) = 0 (2)
𝑑𝑥 𝑑𝑥
and,
𝑑 𝑑
[ (1 − 𝑥 2 ) 𝑃𝑛 (𝑥)] + 𝑛(𝑛 + 1)𝑃𝑛 (𝑥) = 0 (3)
𝑑𝑥 𝑑𝑥
Multiplying eq. (2) by 𝑃𝑛 (𝑥) and eq. (3) by 𝑃𝑚 (𝑥) ,we get
𝑑 𝑑
𝑃𝑛 (𝑥) [(1 − 𝑥 2 ) 𝑃𝑚 (𝑥)] + 𝑚(𝑚 + 1)𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥) = 0 (4)
𝑑𝑥 𝑑𝑥

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

40

𝑑 𝑑
𝑃𝑚 (𝑥) [ (1 − 𝑥 2 ) 𝑃𝑛 (𝑥)] + 𝑛(𝑛 + 1)𝑃𝑚 (𝑥)𝑃𝑛 (𝑥) = 0 (5)
𝑑𝑥 𝑑𝑥
Subtracting eq. (5) from eq (4), we have
𝑑 𝑑 𝑑 𝑑
𝑃𝑛 (𝑥) 𝑑𝑥 [(1 − 𝑥 2 ) 𝑑𝑥 𝑃𝑚 (𝑥)] − 𝑃𝑚 (𝑥) 𝑑𝑥 [ (1 − 𝑥 2 ) 𝑑𝑥 𝑃𝑛 (𝑥)] + (𝑚2 + 𝑚 − 𝑛2 − 𝑛)𝑃𝑚 (𝑥)𝑃𝑛 (𝑥) = 0

⇒ (𝑚 − 𝑛)(𝑚 + 𝑛 + 1) 𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥)

𝑑 𝑑 𝑑 𝑑
= 𝑃𝑚 (𝑥)
𝑑𝑥
[ (1 − 𝑥 2 ) 𝑑𝑥 𝑃𝑛 (𝑥) ] − 𝑃𝑛 (𝑥) 𝑑𝑥 [ (1 − 𝑥 2 ) 𝑑𝑥
𝑃𝑚 (𝑥) ]
(6)

Now using the relation

𝑑 𝑑𝑣 𝑑𝑢
(𝑢𝑣) = 𝑢 +𝑣
𝑑𝑥 𝑑𝑥 𝑑𝑥
Or
𝑑𝑣 𝑑 𝑑𝑢
𝑢 = (𝑢𝑣) − 𝑣
𝑑𝑥 𝑑𝑥 𝑑𝑥
Therefore
(m-n)(𝑚 + 𝑛 + 1)𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥) =
𝑑 𝑑 𝑑 𝑑 𝑑 𝑑
[ 𝑃 (𝑥)(1 − 𝑥 2 ) 𝑃 (𝑥)] − (1 − 𝑥 2 ) 𝑃 (𝑥) 𝑃 (𝑥) − 𝑃 (𝑥)(1 − 𝑥 2 ) 𝑃𝑚 |(𝑥)
𝑑𝑥 𝑚 𝑑𝑥 𝑛 𝑑𝑥 𝑛 𝑑𝑥 𝑚 𝑑𝑥 𝑛 𝑑𝑥
𝑑 𝑑
+ (1 − 𝑥 2 ) 𝑃 (𝑥) 𝑃 (𝑥
𝑑𝑥 𝑚 𝑑𝑥 𝑛
𝑑 𝑑 𝑑
= [ (1 − 𝑥 2 ) {𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥) − 𝑃𝑛 (𝑥) 𝑃 (𝑥)} ] (7)
𝑑𝑥 𝑑𝑥 𝑑𝑥 𝑚
Integrating b/s with respect to ‘x’ from limit -1 to 1, we get
1 1
2)
𝑑 𝑑
(𝑚 − 𝑛)(𝑚 + 𝑛 + 1) ∫ 𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥) = | (1 − 𝑥 {𝑃𝑚 (𝑥) 𝑃 (𝑥) − 𝑃𝑛 (𝑥) 𝑃 (𝑥)} |
−1 𝑑𝑥 𝑛 𝑑𝑥 𝑚 −1

The equation with in the bar being zero at the end point at x= ± 1 and consequently, we get
1
∫ 𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥) 𝑑𝑥 = 0 𝑓𝑜𝑟 𝑚 ≠ 𝑛
−1

Now, we know that

∞
1
∑ 𝑃𝑛 (𝑥) 𝑡 𝑛 = ( 1 − 2𝑡 + 𝑡 2 )−2
𝑛=0

Squaring b/s, we get

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

41

∞ 2

[∑ 𝑃𝑛 (𝑥)𝑡 𝑛 ] = ( 1 − 2𝑡 + 𝑡 2 )−1
𝑛=0

Or,
∞ ∞
1
[ ∑ 𝑃𝑚 (𝑥)𝑡 𝑚 ] [∑ 𝑃𝑛 (𝑥)𝑡 𝑛 ] =
1 − 2𝑥𝑡 + 𝑡 2
𝑚=0 𝑛=0

Integrating b/s from -1 to 1

∞ 1 1
𝑚+𝑛
1
∑ 𝑡 ∫ 𝑃𝑚 (𝑥) 𝑃𝑛 (𝑥) 𝑑𝑥 = ∫ 2
𝑑𝑥 (8)
𝑚,𝑛=0 −1 −1 1 − 2𝑥𝑡 + 𝑡

where the operation of integration and summation are interchanged. Noting that the L.H.S contains
the terms for which m=n, we have
∞ 1
2𝑛
1
∑𝑡 ∫ [𝑃𝑛 (𝑥)]2 𝑑𝑥 = − |log(1 − 2𝑥𝑡 + 𝑡 2 )|
−1 2𝑡
𝑛=0

1
=− [log(1 − 2𝑡 + 𝑡 2 ) − log(1 + 2𝑡 + 𝑡 2 )]
2𝑡
1
=− [log(1 − 𝑡)2 − log(1 + 𝑡)2 ]
2𝑡
1
=− [ 2log(1 − 𝑡) − 2log(1 + 𝑡)
2𝑡
1
= [log(1 + 𝑡) − log(1 − 𝑡)]
𝑡
1 𝑡2 𝑡3 𝑡2 𝑡3
= [(𝑡− + . . . . . . ) − ( −𝑡 − − − . . . . . . . )]
𝑡 2 3 2 3

1 2𝑡 3 2𝑡 5
= [ ( 2𝑡 + + + . . . . . . )]
𝑡 3 5

𝑡2 𝑡4
= 2 [1 + + + . . . . .]
3 5
∞
𝑡 2𝑛
=2∑
(2𝑛 + 1)
𝑛=0
∞
2
= ∑ 𝑡 2𝑛
2𝑛 + 1
𝑛=0
∞ 1 ∞
2𝑛
2
= ∑𝑡 ∫ [𝑃𝑛 (𝑥)]2 𝑑𝑥 = ∑ 𝑡 2𝑛 (9)
−1 2𝑛 + 1
𝑛=0 𝑛=0

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

42

Now, equating coefficient of 𝑡 2𝑛 on both sides of eq. (9), we get

1
2
∫ [𝑃𝑛 (𝑥)]2 𝑑𝑥 = ;𝑚 = 𝑛
−1 2𝑛 + 1

Some elementary properties of Legendre’s polynomial; -

Theorem:- If 𝑃𝑛 (𝑡) is the legendres polynomial then
𝑛
[ ] 1 𝑛! (−1)𝑟 (2𝑛−2𝑟)!
(i) 𝑃𝑛 (𝑡) = ∑𝑟=0 2
𝑡 𝑛−2𝑟
2𝑛 𝑛! 𝑟! 𝑛−𝑟 (𝑛−2𝑟)!

(2𝑛)!
(ii) The coefficient of 𝑡 𝑛 in 𝑃𝑛 (𝑡)𝑖𝑠 (2𝑛 .(𝑛!)2 )
(iii) 𝑃𝑛 (−𝑡) = (−1)𝑛 𝑃𝑛 (𝑡) 𝑎𝑛𝑑 ℎ𝑒𝑛𝑐𝑒 𝑃𝑛 (−1) = (−1)𝑛 𝑃𝑛 (1)

Proof; - From the Rodrigues formula, we have

1 dn 2
𝑃𝑛 (𝑡) = (𝑡 − 1)𝑛
2𝑛 𝑛! dx n
Expanding (𝑡 2 − 1) by Binomial theorem
𝑛 𝑛

(𝑡 2 𝑛
− 1) = ∑ 𝑛𝐶𝑟 (−1)𝑟 (𝑡 2 )𝑛−𝑟 = ∑ 𝑛𝐶𝑟 (−1)𝑟 𝑡 2𝑛−2𝑟
𝑟=0 𝑟=0

Hence
𝑑𝑛 2 𝑛 𝑟
𝑑 𝑛 2𝑛−2𝑟
(𝑡 − 1) = ∑ 𝑛𝐶𝑟 (−1) (𝑡 ) (1)
𝑑𝑡 𝑛 𝑑𝑡
𝑑𝑛
𝑊𝑒 𝑘𝑛𝑜𝑤 𝑡ℎ𝑎𝑡 𝑡𝑚 = 0 𝑖𝑓 𝑚 < 𝑛 𝑛𝐶𝑟
𝑑𝑡 𝑛

𝑑𝑛 𝑚 𝑚!
𝑡 = 𝑡 𝑚−𝑛 𝑖𝑓 𝑚 > 𝑛 (2)
𝑑𝑡 𝑛 (𝑚 − 𝑛)!
𝑑𝑛 𝑛 𝑛
Hence, 𝑡 2𝑛−2𝑟 = 0 if 2n-2r < n which gives n < 2r or < 𝑟 .So we have terms upto and
𝑑𝑡 𝑛 2 2
𝑛
all the subsequent terms in the summation are zero. Hence we replace ∑𝑛𝑟=0 𝑏𝑦 ∑𝑟=0 𝑖𝑓 n 2

𝑛−1
𝑛 𝑛 𝑛−1
is even and by ∑𝑟=0 𝑖𝑓 n ia odd . Let [ 2] denote 2 if n is even and
2
if n is odd. Then
2
we can rewrite
𝑛
[ ]
2
𝑛
1 𝑑 2 𝑛
1 𝑟
𝑑𝑛 2𝑛−2𝑟
(𝑡 − 1) = ∑ 𝑛 𝐶𝑟 (−1) 𝑡 (3)
2𝑛 𝑛! 𝑑𝑡 𝑛 2𝑛 𝑛! 𝑑𝑡 𝑛
𝑟=0

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

43

Using formula (2) & (3), we get

𝑛
[ ]
2
1 𝑛
(2𝑛 − 2𝑟)! 2𝑛−2𝑟−𝑛
∑ 𝐶𝑟 (−1)𝑟 𝑡
2𝑛 𝑛! (𝑛 − 2𝑟 − 𝑛)!
𝑟=0
𝑛
[ ]
2
1 𝑛! (−1)𝑟 (2𝑛 − 2𝑟)! 2𝑛−2𝑟−𝑛
=∑ 𝑛 𝑡
2 𝑛! 𝑟! (𝑛 − 𝑟)! (2𝑛 − 2𝑟 − 𝑛)!
𝑟=0
𝑛
[ ]
2
1 𝑛! (−1)𝑟 (2𝑛 − 2𝑟)! 𝑛−2𝑟
=∑ 𝑡 (4)
2𝑛 𝑛! 𝑟! (𝑛 − 𝑟)! (𝑛 − 2𝑟)!
𝑟=0

which proves (i)

(ii) The coefficient of 𝑡 𝑛 will correspond to r=o in the above expansion. So when r=0
we get
1 𝑛! 2𝑛! 1 2𝑛!
= 𝑛
2𝑛 𝑛! 𝑛!
𝑛! 2 (𝑛!)2
which proves (ii)
(iii) To prove (iii), we get from (4)
𝑛
[ ]
2
1 𝑛! (−1)𝑟 (2𝑛 − 2𝑟)! 𝑛−2𝑟
𝑃𝑛 (𝑡) = ∑ 𝑡 (5)
2𝑛 𝑛! 𝑟! (𝑛 − 𝑟)! (𝑛 − 2𝑟)!
𝑟=0

Replacing t by –t in (5) , w get

𝑛
[ ]
2
1 𝑛! (−1)𝑟 (2𝑛 − 2𝑟)!
𝑃𝑛 (−𝑡) = ∑ (−1)𝑛−2𝑟 𝑡 𝑛−2𝑟
2𝑛 𝑛! 𝑟! (𝑛 − 𝑟)! (𝑛 − 2𝑟)!
𝑟=0

Noting that (−1)(2𝑟) is 1 , we get 𝑃𝑛 (– 𝑡) = (−1)𝑛 𝑃𝑛 (𝑡)

As a particular case , let us take t=1 , then we have 𝑃𝑛 (– 1) = (−1)𝑛 𝑃𝑛 (1)
Corollary .
𝑛
(−1) 2 (1
. 3 . 5 . (𝑛 − 1))
𝑃𝑛 (0) =
2 ..4 .𝑛
if n is even and 𝑃𝑛 (0) = 0 if n is odd .
Proof ;- Using (i) , we find the value of 𝑃𝑛 (𝑡) at t =0 . The constant term in 𝑃𝑛 𝑡) gives the value
of 𝑃𝑛 (0) 𝑖𝑛 (𝑖)
𝑛
To obtain this when n is even, let us replace r by
2
𝑖𝑛 (4)

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

44

Now P(t)= -2t and Q(t) = 2p. Since the polynomial -2t and constant 2p are analytic near t =0, the
point t=0 is an ordinary point.
Let us consider the power series and its derived series
𝑥(𝑡) = 𝑎0 + 𝑎1 𝑡 + 𝑎2 𝑡 2 + 𝑎3 𝑡 3 + . . . . . +𝑎𝑘 𝑡 𝑘 + . . . . (2)

𝑥 ′(𝑡) = 𝑎1 + 2𝑎2 𝑡 + 3𝑎3 𝑡 2 + . . . . . +𝑘𝑎𝑘 𝑡 𝑘−1 + . . . .

𝑥 ′′(𝑡) = 2𝑎2 + 3.2𝑎3 𝑡 + 4.3𝑎4 𝑡 2 + . . . . . +(𝑘 + 2)(𝑘 + 1)𝑎𝑘+2 𝑡 𝑘 + . . . . (3)

−2𝑡𝑥 ′ (𝑡) = −2[𝑎1 𝑡 + 2𝑎2 𝑡 2 + 3𝑎3 𝑡 3 + . . . . . +𝑘𝑎𝑘 𝑡 𝑘 + . . . ] (4)

2𝑝𝑥(𝑡) = 2𝑝[𝑎0 + 𝑎1 𝑡 + 𝑎2 𝑡 2 + . . . . +𝑎𝑘 𝑡 𝑘 + . . . . ] (5)

Adding the power series (3),(4) and (5) term by term , we get
(2𝑎2 + 2𝑝𝑎0 ) + (3.2𝑎3 − 2𝑎1 + 2𝑝𝑎1 )𝑡
+ (4.3𝑎4 − 2.2𝑎2 + 2𝑝𝑎2 )𝑡 2 + . . . +[(𝑘 + 1)(𝑘 + 2)𝑎𝑘+2 − 2𝑘𝑎𝑘 + 2𝑝𝑎𝑘 ]𝑡 𝑘 (6)

Thus substituting for x’’, ‘x’ and x in the given differential equation, we get the power series (6)
whose sum is zero near t=0. This is possible if and only if all the coefficients in (6) are zero. Hence
equating the successive coefficients to zero, we obtain

2𝑎2 + 2𝑝𝑎0 = 0,3.2𝑎3 − 2𝑎1 + 2𝑝𝑎1 = 0

4.3 . 𝑎4 − 2.2 𝑎2 + 2𝑝𝑎2 = 0

In general, we have
(𝑘 + 1)(𝑘 + 2)𝑎𝑘+2 − 2𝑘𝑎𝑘 + 2𝑝𝑎𝑘 = 0 (7)

From the above equations and (7) , we get

2𝑝𝑎0 2(𝑝 − 1) 2(𝑝 − 𝑘)
𝑎2 = − , 𝑎3 = − 𝑎1 , 𝑎𝑘+2 = − 𝑎 (8 )
2 3.2 (𝑘 + 1)(𝑘 + 2) 𝑘

Equation (8) is a two term recursion formula from which we shall construct even and odd
coefficients as follow. First, replacing k by 2k-2 in (8), we get
2[𝑝 − (2𝑘 − 2)]
𝑎2𝑘 = − 𝑎2𝑘−2
(2𝑘 − 1)2𝑘

Where
2[𝑝 − (2𝑘 − 4)]
𝑎2𝑘−2 = − 𝑎
(2𝑘 − 3)(2𝑘 − 2) 2𝑘−4

so that we have
(−2)2 [𝑝 − (2𝑘 − 2)][𝑝 − (2𝑘 − 4)]
𝑎2𝑘 = 𝑎2𝑘−4
(2𝑘 − 3)(2𝑘 − 2)(2𝑘 − 1)𝑘

Proceeding in a similar manner we get

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

45
Hermite Polynomials

Is a solution of the Hermite equation,

x’’(t)-2tx’(t)+2nx(t) =0
where n is a non-negative integer n
2
Proof:- Let us take 𝑢(𝑡) = 𝑒 −𝑡
2
Differentiating this, we get 𝑢′ (𝑡) = −2𝑡𝑒 −𝑡
which gives 𝑢′ (𝑡) + 2𝑡𝑢 = 0 (2)
Using Leibnitz formula,let us differentiate (2) n-times . Then
𝑢𝑛+1 (𝑡) + 2𝑡𝑢𝑛 (𝑡) + 2𝑛𝑢𝑛−1 (𝑡) = 0 (3)
𝑡2
Multiplying (3) throughout by (−1)𝑛+1 𝑒 2 , , 𝑤𝑒 𝑔𝑒𝑡
𝑡2 𝑡2 𝑡2
(−1)𝑛+1 𝑒 2 𝑢(𝑛+1) (𝑡) + (−1)𝑛+1 𝑒 2 2𝑡𝑢(𝑛) (𝑡) + (−1)𝑛+1 𝑒 2 2𝑛𝑢(𝑛−1) (𝑡) = 0
which we can rewrite after substituting for u as
𝑡2 𝑑𝑛+1 −
𝑡2 𝑡2 𝑑𝑛 −
𝑡2 𝑡2 𝑑𝑛−1 −
𝑡2
(−1)𝑛+1 𝑒2 (𝑒 2 𝑛
) − (−1) 𝑒 2𝑡 𝑛 ( 𝑒
2 2 ) + (−1)𝑛−1
𝑒 2𝑛 𝑛−1 (𝑒 2 )
2 =0
𝑑𝑡 𝑛+1 𝑑𝑡 𝑑𝑡

Using (1) in the equation, we get

𝐻𝑛+1 − 2𝑡 𝐻𝑛 + 2𝑛 𝐻𝑛−1 = 0 (4)
Let us now differentiate 𝐻𝑛 (𝑥) directly as follows:
2 𝑑𝑛 −𝑡 2 𝑛 𝑡2
𝑑𝑛+1 2
𝐻𝑛′ (𝑡) = (−1)𝑛 2𝑡𝑒 𝑡 𝑛
𝑒 + (−1) 𝑒 𝑛+1 (𝑒 −𝑡 )
𝑑𝑡 𝑑𝑡

which gives
𝐻𝑛′ (𝑡) = 2𝑡 𝐻𝑛 − 𝐻𝑛+1 (5)
that expresses 𝐻𝑛′ in terms of 𝐻𝑛 𝑎𝑛𝑑 𝐻𝑛+1
We shall show that 𝐻𝑛 is the solution of the Hermite equation
x’’(t) - 2tx’(t)+2nx(t) =0
We find 𝐻𝑛′ 𝑎𝑛𝑑 𝐻𝑛 ′′ and simplify them by using (4) and (5)
From (5) we get 𝐻𝑛′ − 2𝑡𝐻𝑛 − 𝐻𝑛+1 (6)
Using (4) in (6) we get 𝐻𝑛′ = 2𝑡𝐻𝑛 − [2𝑡𝐻𝑛 − 2𝑛𝐻𝑛−1 ]
Hence we have 𝐻𝑛′ = 2𝑛𝐻 ′ 𝑛−1 (7)

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

46

′
From (7) we have 𝐻𝑛′′ = 2𝑛𝐻𝑛−1
Using (5) we obtain 𝐻𝑛′′ = 2𝑛[2𝑡𝐻𝑛−1 − 𝐻𝑛 ] (8)
Now

𝑥 ′′ (𝑡) − 2𝑡𝑥 ′ (𝑡) + 2𝑛𝑥(𝑡) = 2𝑛[2𝑡𝐻𝑛−1 − 𝐻𝑛 ] − 2𝑡[2𝑛𝐻𝑛−1 ] + 2𝑛𝐻𝑛

= 4𝑛𝑡 𝐻𝑛−1 − 2𝑛𝐻𝑛 − 4𝑛𝑡 𝐻𝑛−1 + 2𝑛𝐻𝑛 = 0
Hence 𝐻𝑛 defined by (1) is a solution of the Hermite equation

Definition :- 𝐻𝑛 ′𝑠 for n=1,2,3 . . . . . . . are called Hermite polynomials

Let us now find out 𝐻𝑛 , 𝐻1 , 𝐻2 𝑎𝑛𝑑 𝐻4
2 𝑑0 2 2 2
(i) 𝐻0 = (−1)0 𝑒 𝑡 (𝑒 −𝑡 ) = 𝑒 𝑡 . 𝑒 −𝑡 = 1
𝑑𝑡 0

2 𝑑 2 2 2
(ii) 𝐻1 = (−1)1 𝑒 𝑡 𝑑𝑡
(𝑒 −𝑡 ) = −1 𝑒 𝑡 [ −𝑒 −𝑡 2𝑡] = 2𝑡

2 𝑑2 2 2 𝑑 2 2 2 2
(iii) 𝐻2 = (−1)2 𝑒 𝑡 𝑑𝑡 2
(𝑒 −𝑡 ) = 𝑒 𝑡 𝑑𝑡
[−𝑒 −𝑡 2𝑡] = 𝑒 𝑡 [−2𝑒 −𝑡 + 4𝑡 2 𝑒 −𝑡 ] = 4𝑡 2 − 2

2 𝑑3 2 2 𝑑 2 2
(iv) 𝐻3 = (−1)3 𝑒 𝑡 𝑑𝑡 3
(𝑒 −𝑡 ) = −𝑒 𝑡 𝑑𝑡
[−2𝑒 −𝑡 + 4𝑡 2 𝑒 −𝑡 ]

2 2 2 2
= −𝑒 𝑡 [4𝑒 −𝑡 𝑡 + 8𝑡𝑒 −𝑡 − 8𝑡 3 𝑒 −𝑡 ]

= 8𝑡 3 − 8𝑡 − 4𝑡 = 8𝑡 3 − 12𝑡

Generating function of the Hermite function: -

If 𝐻𝑛 (𝑡) are Hermite polynomial, show that
∞
2𝑡𝑥−𝑥 2
𝑥𝑛
𝑒 =∑ 𝐻 (𝑡 )
𝑛! 𝑛
𝑛=0

Now
∞ ∞
2𝑡𝑥−𝑥 2 2𝑡𝑥 −𝑥 2
(2𝑡𝑥)𝑟 (−𝑥 2 )𝑠
𝑒 =𝑒 . 𝑒 =∑ ∑
𝑟! 𝑠!
𝑟=0 𝑠=0

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

47

∞
(−1)𝑠 (2𝑡)𝑟 𝑟+2𝑠
= ∑ 𝑥
𝑟! 𝑠!
𝑟,𝑠=0

If n= r+2s, the coefficient of 𝑥 𝑛 for a fixed value of s is given by

(−1)𝑠 (2𝑡)𝑛−2𝑠
(𝑛 − 2𝑠)! 𝑠!
The coefficient of 𝑥 𝑛 is obtained by summing up for all permissible values of s given by n-2s =r.
𝑛 𝑛
Since r ≥ 0, we have n- 2s ≥0 which gives s ≤ .Thus if n is even, s varies from 0 to 2 and if n is
2
𝑛−1
odd s varies from 0 𝑡𝑜
2
. Hence the coefficient of 𝑥 𝑛 is
𝑛
[ ]
2 ∞
(−1)𝑠 (2𝑡)𝑛−2𝑠 𝐻𝑛 (𝑡) 2 𝑥𝑛
∑ = 𝑠𝑜 𝑡ℎ𝑎𝑡 𝑒 2𝑡𝑥−𝑥 = ∑ 𝐻 (𝑡)
(𝑛 − 2𝑠)! 𝑠! 𝑛! 𝑛! 𝑛
𝑠=0 𝑛=0
2
The function 𝑒 2𝑡𝑥−𝑥 is called the generating function of the Hermite function
● Using generating function of the Hermite polynomial, Prove the following properties
(2𝑛)!
(i) 𝐻2𝑛 (0) =
𝑛!

(ii) 𝐻2𝑛+1 (0) = 0

2 𝑑𝑛 2
(iii) 𝐻𝑛 (𝑡) = (−1)𝑛 𝑒 𝑡 𝑑𝑡 𝑛
(𝑒 −𝑡 )

From the generating function of the Hermite Polynomials

∞
𝑥𝑛 2
∑ 𝐻𝑛 (𝑡) = 𝑒 2𝑡𝑥−𝑥 (1)
𝑛!
𝑛=0

Let us take t=0 in (1) . Then we have

𝑥𝑛 2
∑ 𝐻𝑛 (0) = 𝑒 −𝑥
𝑛!
𝑥 2 (𝑥 2 )2 (−1)𝑛 (𝑥 2 )𝑛
=1− + − . . . . . + . . . (2)
1! 2! 𝑛!
Equating the coefficient of 𝑥 2𝑛 on both sides of (2) , we get
1 .1
𝐻2𝑛 (0) = (−1)𝑛
2𝑛! 𝑛!
Which proves (i)

Downloaded by Akansha Baliyan ([email protected])

 lOMoARcPSD|42174153

48

1
To prove (ii) , let us equate the coefficient of 𝑥 2𝑛+1 . Then we have 𝐻
(2𝑛+1)! 2𝑛+1
(0) = 0 because
the expansion (2) does not have odd powers of x.
iii) Rewriting the generating function by completing the square, we get
2 −(𝑥−𝑡)2 𝐻1 (𝑡) 𝐻2 (𝑥) 2 𝐻𝑛 (𝑡) 𝑛
𝑒𝑡 = 𝐻0 (𝑡) + 𝑥+ 𝑥 + . . . .+ 𝑥 + . . . .
1! 2! 𝑛!
Partially differentiating both sides of the above equation with respect to x for n-times, we get
𝐻𝑛 (𝑡) 𝜕 𝑛 −(𝑥−𝑡) 2 𝑡 2
𝑛! = [ 𝑛 𝑒 ] 𝑒 𝑎𝑡 𝑥 = 0 (3)
𝑛! 𝜕𝑥
Let u= x - t. When x = 0, u = - t
Hence
𝜕 𝑛 −(𝑥−𝑡) 2 𝜕𝑛 2
[ 𝑛
𝑒 ] = 𝑛
( 𝑒 −𝑢 )
𝜕𝑥 𝑎𝑡 𝑥=0 𝜕𝑢
𝜕𝑛 𝑑𝑛
But we note that = (−1)𝑛
𝜕𝑢𝑛 𝑑𝑡 𝑛

Hence we get
𝜕𝑛 −𝑢2 𝑛
𝑑𝑛 2
( 𝑒 ) = (−1) ( 𝑒 −𝑡 ) (4)
𝜕𝑢𝑛 𝑑𝑡 𝑛
Using (4) in (3), we get
𝑑𝑛 −𝑡 2 𝑡2 𝑛 𝑡2
𝑑𝑛 2
𝐻𝑛 (𝑡) = (−1)𝑛 (𝑒 )𝑒 = (−1) 𝑒 (𝑒 −𝑡 )
𝑑𝑡 𝑛 𝑑𝑡 𝑛
Which is the requires result.

Bessel equation and Bessel function

The differential equation
𝑑2𝑦 𝑑𝑦
𝑥2 2
+𝑥 + (𝑥 2 − 𝑝2 )𝑦 = 0 (1)
𝑑𝑥 𝑑𝑥
Where ‘p’ is a parameter and is called ‘Bessel equation of order p. Any solution of Bessel
equation of order p is called Bessel function of order p

Power series solution of Bessel equation of order p

The differential eq. (1) can be written as

Downloaded by Akansha Baliyan ([email protected])

 ACKNOWLEDGEMENT
The successful and final outcome of this project required lots of guidance and
assistance from many people. I am extremely fortunate to have complete my
project work. Whatever I have done is only due to such guidance and
assistance and I should not forget to thank them.

First of all , I would like to convey my heartfelt thanks to DR. SANJAY KUMAR
SIR. Head of department , who always gave valuable suggestions and the
guidance for completion of my project. He helped me to understand and
remember important details of the project.

Secondly , I would like to express my special thanks of gratitude to others

teacher of mathematics department, who gave me the golden opportunity to
do this wonderful project of “ordinary differential equations and
its significant” , who also helped me in the completion my project and
moreover , on time.

Then, I would also like to thank our librarian who had been very helpful and
cooperative in project properly.

Besides, I would like to thanks my parent for giving encouragement,

enthusiasm and valuable assistance to me. Without all this, I might not be able
to complete this project properly.

Last but not least, I would like to thanks other staff member and my friends for
their most valuable efforts to help me.
 MAHARAJ SINGH COLLEGE
(AFFILIATED TO M.S. UNIVERSITY, SAHARANPUR)

CERTIFICATE

THE CANDIDATE HAS EARNED THE ELIGIBILITY FOR THE

SUBMISSION CERTIFIED THAT MRS. HARSH BHARTI
UNDER THE ROLL NO. 210081302125 WORKED ON
THE PROJECT ENTITLED “ORDINARY HAS
DIFFERENTIAL EQUATIONS AND ITS
SIGNIFICANT” OF THE PROJECT REPORT AS HE
HAS PUT IN THE REQUIRED ATTENDANCE AS PER RULE.

DR. ANIL KUMAR SIR DR. SANJAY KUMAR SIR

( PRINCIPAL ) ( HEAD OF DEPARTMENT
PROJECT SUPERVISOR )
 ORDINARY DIFFERENTAIL
EQUATIONS AND ITS SIGNIFICANTS

PROJECTSUBMITTED IN PARTIAL FULLFILMENT OF THE

REQUIREMENTS FOR THE DEGREE OF BACHELOR OF SCIENCE
IN

MATHEMATICS
SUBMITTED BY :-
HARSH BHARTI
ROLL NO. :- 210081302125

UNDER THE GUIDANCE OF :-

DR. SANJAY KUMAR SIR

TO
DEPARTMENT OF MATHEMATICS
OF MAHARAJ SINGH COLLEGE
SAHARANPUR, UP
INDIA