UNIT-1

Matrices
A matrix is a rectangular arrangement of elements in rows and columns.
A system of m n numbers (real or complex) arranged in m rows and n columns between
[ ] or ( ) is called a matrix of order 𝑚 𝑋 𝑛.
An 𝑚 𝑋 𝑛 matrix is written as A=

Types of matrices:
Square matrix: In a matrix 𝐴 = (𝑎𝑖𝑗 )𝑚 𝑋 𝑛 if 𝑚 = 𝑛 then A is called a square matrix.

or A matrix in which no. of rows and no. of columns are equal is called a square matrix.
1 7 0
Eg: A= [5 −1 7] is a square matrix of order 3.
8 9 3
Rectangular matrix: In a matrix 𝐴 = (𝑎𝑖𝑗 )𝑚 𝑋 𝑛 if 𝑚 ≠ 𝑛 then A is called a rectangular
matrix.
A matrix in which no. of rows is not equal to no. of columns is called a rectangular matrix.
2 5
Eg: A= [1 8] is a rectangular matrix of order 3 𝑋 2.
4 7
Null matrix: In a matrix 𝐴 = (𝑎𝑖𝑗 )𝑚 𝑋 𝑛 if 𝑎𝑖𝑗 = 0 ∀ 𝑖, 𝑗 then A is called a zero matrix or
Null matrix. It is denoted by 𝑂𝑚 𝑋 𝑛 .
0 0 0
0 0 0
[ ] , [0 0 0] are null matrices of order 2 𝑋 3 and 3 𝑋 3 respectively.
0 0 0 2𝑋3
0 0 0 3𝑋3
Diagonal elements of a square matrix: In a square matrix 𝐴 = (𝑎𝑖𝑗 )𝑚 𝑋 𝑛 the elements
𝑎11, 𝑎22 , … , 𝑎𝑛𝑛 are called diagonal elements of principal diagonal elements.
Sum of all principal diagonal elements of a square matrix is called its trace.
Tr(𝐴) = 𝑎11 + 𝑎22 + ⋯ + 𝑎𝑛𝑛 .

Diagonal matrix. A square matrix in which the elements other than the principal diagonal
elements are zero is called a diagonal matrix.
A matrix 𝐴 = (𝑎𝑖𝑗 )𝑛 𝑋 𝑛 is a diagonal matrix if 𝑎𝑖𝑗 = 0 for 𝑖 = 𝑗.

If 𝑑1 , 𝑑2 , … , 𝑑𝑛 are the principal diagonal elements of a diagonal matrix, then

𝐴 = (𝑑1 , 𝑑2 , … , 𝑑𝑛 ).

Scalar Matrix: A diagonal matrix in which all the principal diagonal elements are equal
is called a scalar matrix i.e.

in a scalar matrix 𝐴 = (𝑎𝑖𝑗 )𝑛 𝑋 𝑛 , 𝑎𝑖𝑗 = 𝑘 for 𝑖 = 𝑗 and 𝑎𝑖𝑗 = 0 for 𝑖 ≠ 𝑗.

6 0 0
Eg: [0 6 0] is a scalar matrix of order 3.
0 0 6

Unit matrix: If 𝐴 = (𝑎𝑖𝑗 )𝑛 𝑋 𝑛 is a square matrix in which 𝑎𝑖𝑗 = 0 for 𝑖 ≠ 𝑗 and

𝑎𝑖𝑗 = 1 for 𝑖 = 𝑗, then A is called a unit matrix. It is denoted by 𝐼𝑛 .

1 0 0
Eg: [0 1 0] is a unit matrix of order 3 and is denoted by 𝐼3 .
0 0 1
Idempotent matrix: A square matrix 𝐴 is called an idempotent matrix if 𝐴2 = 𝐴.
Involutary matrix: A square matrix 𝐴 is called an involutary matrix if 𝐴2 = 𝐼.
Nilpotent matrix: If 𝐴 is a square matrix and m is the least positive integer such that
𝐴𝑚 = 0 then 𝐴 is called nilpotent of index 𝑚.
Symmetric matrix: A square matrix 𝐴 is said to be symmetric if 𝐴𝑇 = 𝐴
1 2 3
Eg: 𝐴 = [2 4 5] then 𝐴 is symmetric since 𝐴𝑇 = 𝐴.
3 5 6
Skew- Symmetric matrix: A square matrix 𝐴 is said to be Skew-symmetric if
𝐴𝑇 = −𝐴.
Every diagonal element of a skew symmetric matrix is zero.
0 2 −3
Eg: 𝐴 = [−2 0 5 ] then 𝐴 is Skew- symmetric since 𝐴𝑇 = −𝐴.
3 −5 0
Orthogonal matrix: A square matrix A is said to be orthogonal if 𝐴𝑇 𝐴 = 𝐼 = 𝐴𝑇 𝐴.
Therefore A is orthogonal ⇔ 𝐴−1 = 𝐴𝑇 .
Properties of orthogonal matrix:
1. Product of two orthogonal matrices is an orthogonal matrix.
 2. Transpose of an orthogonal matrix is an orthogonal matrix.
3. Inverse of an orthogonal matrix is an orthogonal matrix.
4. Determinant of an orthogonal matrix is ±1.

Theorem: Every square matrix can be expressed as the sum of a symmetric and a
skew-symmetric matrix uniquely.
Proof: Let 𝐴 be a square matrix.
1 1 1
Then we write 𝐴 = 2 (𝐴 + 𝐴𝑇 + 𝐴 − 𝐴𝑇 ) = (𝐴 + 𝐴𝑇 ) + (𝐴 − 𝐴𝑇 )= 𝐵 + 𝐶
2 2
1 1
where 𝐵 = 2 (𝐴 + 𝐴𝑇 ), 𝐶 = 2 (𝐴 − 𝐴𝑇 )
1 1
𝐵 𝑇 = 2 (𝐴 + 𝐴𝑇 )𝑇 = 2 (𝐴𝑇 + 𝐴) = 𝐵

Therefore, B is symmetric.
1 1
𝐶𝑇 = (𝐴 − 𝐴𝑇 )𝑇 = (𝐴𝑇 − 𝐴) = −𝐶
2 2
Therefore, C is skew symmetric.
Uniqueness:
Let 𝐴 = 𝑃 + 𝑄 be another such representation of A where P is symmetric and Q is skew
symmetric.
Therefore 𝑃𝑇 = 𝑝 and 𝑄 𝑇 = −𝑄
And 𝐴𝑇 = (𝑃 + 𝑄)𝑇 = 𝑃𝑇 + 𝑄 𝑇 = 𝑃 − 𝑄
1 1
𝐵 = 2 (𝐴 + 𝐴𝑇 ) = 2 (𝑃 + 𝑄 + 𝑃 − 𝑄) = 𝑃 and
1 1
𝐶 = 2 (𝐴𝑇 − 𝐴) = 2 (𝑃 + 𝑄 − 𝑃 + 𝑄) = 𝑄

Which shows that 𝑃 = 𝐵 𝑎𝑛𝑑 𝑄 = 𝐶.

Hence, the representation is unique.
 −𝟏 𝟕 𝟏
1. Express the matrix [ 𝟐 𝟑 𝟒 ] as the sum of a symmetric matrix and a skew
𝟓 𝟎 𝟓
symmetric matrix.
Solution:
−1 7 1 −1 2 5
𝑇
Let A= [ 2 3 4 ] then 𝐴 = [ 7 3 0]
5 0 5 1 4 5
−1 9/2 3
1 𝑇 )=
Symmetric part of A = (𝐴 + 𝐴 [ 9/2 3 2]
2
3 2 5
0 5/2 −2
1 𝑇)
Skew-symmetric part of A = 2 (𝐴 − 𝐴 = [−5/2 0 2]
2 −2 0
−1 9/2 3 0 5/2 −2
Hence, 𝐴 = [ 9/2 3 2 ] + [−5/2 0 2 ].
3 2 5 2 −2 0

𝟏 𝟎 𝟎
2. Show that the matrix [ 𝟎 𝐬𝐢𝐧 𝜽 𝐜𝐨𝐬 𝜽] is orthogonal.
𝟎 − 𝐜𝐨𝐬 𝜽 𝐬𝐢𝐧 𝜽
1 0 0 1 0 0
𝑇
Solution: Let 𝐴 = [ 0 sin 𝜃 cos 𝜃] then 𝐴 = [ 0 sin 𝜃 −cos 𝜃]
0 − cos 𝜃 sin 𝜃 0 cos 𝜃 sin 𝜃
1 0 0 1 0 0 1 0 0
𝐴𝐴𝑇 = [ 0 sin 𝜃 cos 𝜃 ] [ 0 sin 𝜃 −cos 𝜃] = [0 1 0] = 𝐼
0 − cos 𝜃 sin 𝜃 0 cos 𝜃 sin 𝜃 0 0 1
1 0 0 1 0 0 1 0 0
And 𝐴𝑇 𝐴 = [ 0 sin 𝜃 −cos 𝜃] [ 0 sin 𝜃 cos 𝜃] = [0 1 0] = 𝐼
0 cos 𝜃 sin 𝜃 0 − cos 𝜃 sin 𝜃 0 0 1
Therefore 𝐴 is orthogonal.
𝟎 𝟐𝒃 𝒄
3. Determine a, b, c so that [ 𝒂 𝒃 −𝒄] is orthogonal.
𝒂 −𝒃 𝒄
𝟏 𝟏 𝟏
Ans: 𝒂 = ± ,𝒃 = ± =±
√𝟐 √𝟔 √𝟑

𝟏 −𝟐 𝟐
𝟏
4. Find a positive integer ‘a’ such that [ −𝟐 𝟏 𝟐 ] is orthogonal. (Ans: 3)
𝒂
−𝟐 −𝟐 −𝟏
Complex Matrices:
Conjugate of a Matrix: If all the elements of a Matrix 𝐴 are replaced by their complex
conjugates, then the resulting matrix is called the conjugate of 𝐴 and is denoted by 𝐴̅.
If 𝐴 = (𝑎𝑖𝑗 )𝑚 𝑋 𝑛 is a matrix, then 𝐴̅ = ( 𝑎̅𝑖𝑗 )𝑚 𝑋 𝑛 is called conjugate of 𝐴.

Eg;
𝟏+𝒊 𝟐 𝟑 − 𝟐𝒊 𝟏−𝒊 𝟐 𝟑 + 𝟐𝒊
If A = [ 𝟒 𝟓 − 𝟐𝒊 ̅
𝒊 − 𝟏 ] then 𝐴 = [ 𝟒 𝟓 + 𝟐𝒊 −𝒊 − 𝟏 ]
𝟓 − 𝟒𝒊 −𝟕 𝟎 𝟓 + 𝟒𝒊 −𝟕 𝟎
Transposed conjugate of a Matrix: Let 𝐴 = (𝑎𝑖𝑗 ) be any given matrix. Then the
transpose of the conjugate of A is called the transposed conjugate of A and is denoted
by 𝐴∗ or 𝐴𝜃 .
𝟏+𝒊 𝟐 𝟑 − 𝟐𝒊 𝟏−𝒊 𝟒 𝟓 + 𝟒𝒊
If A = [ 𝟒 𝟓 − 𝟐𝒊 ∗ ̅
𝑇
𝒊 − 𝟏 ] then 𝐴 = 𝐴 = [ 𝟐 𝟓 + 𝟐𝒊 −𝟕 ].
𝟓 − 𝟒𝒊 −𝟕 𝟎 𝟑 + 𝟐𝒊 −𝒊 − 𝟏 𝟎

Hermitian Matrix: If A is a square matrix such that 𝐴∗ = 𝐴, then A is said to be a

Hermitian matrix.
In a Hermitian matrix, all the diagonal elements are real.
1 2 + 5𝑖 3 − 2𝑖
Eg: A = [ 2 − 5𝑖 5 −𝑖 ] is a Hermitian matrix.
3 + 2𝑖 𝑖 0
Skew-Hermitian matrix: If A is a square matrix such that 𝐴∗ = −𝐴, then A is said
to be a Skew-Hermitian matrix.
In a Skew-Hermitian matrix, the diagonal elements are either purely imaginary or equal
to zero.
5𝑖 2 + 5𝑖 3 − 2𝑖
Eg: A = [−2 + 5𝑖 0 −𝑖 ] is a Skew-Hermitian matrix.
−3 − 2𝑖 −𝑖 −𝑖
Unitary Matrix: If A is a square matrix such that 𝐴∗ 𝐴 = 𝐼 = 𝐴𝐴∗ , then A is said to
be unitary.
1 0 0
Eg: A= [0 0 −1] is unitary.
1 0 1
Theorem: Every square matrix can be expressed as the sum of a Hermitian matrix and
a Skew-Hermitian matrix uniquely.
Proof is left to the students.
1 1
Hermitian part of A = 2 (𝐴 + 𝐴∗ ), Skew- Hermitian part of A = = 2𝑖 (𝐴 − 𝐴∗ ),

Problems:
𝑎 + 𝑖𝑐 −𝑏 + 𝑖𝑑
1. Show that the matrix [ ] is unitary if 𝑎2 + 𝑏 2 + 𝑐 2 + 𝑑 2 = 1.
𝑏 + 𝑖𝑑 𝑎 − 𝑖𝑐
Solution:
𝑎 + 𝑖𝑐 −𝑏 + 𝑖𝑑
Let A = [ ]
𝑏 + 𝑖𝑑 𝑎 − 𝑖𝑐

𝑎 − 𝑖𝑐 −𝑏 − 𝑖𝑑 𝑇 𝑎 − 𝑖𝑐 𝑏 − 𝑖𝑑
𝐴∗ = 𝐴̅𝑇 = [ ] =[ ]
𝑏 − 𝑖𝑑 𝑎 + 𝑖𝑐 −𝑏 − 𝑖𝑑 𝑎 − 𝑖𝑐

𝑎 + 𝑖𝑐 −𝑏 + 𝑖𝑑 𝑎 − 𝑖𝑐 𝑏 − 𝑖𝑑
𝐴𝐴∗ = [ ][ ]
𝑏 + 𝑖𝑑 𝑎 − 𝑖𝑐 −𝑏 − 𝑖𝑑 𝑎 − 𝑖𝑐
2 2 2 2
= [𝑎 + 𝑏 + 𝑐 + 𝑑 0 ]
0 𝑎 + 𝑏 + 𝑐 2 + 𝑑2
2 2

A is unitary if 𝐴𝐴∗ = 𝐼
i.e. if 𝑎2 + 𝑏 2 + 𝑐 2 + 𝑑2 = 1.
1+𝑖 2 5 − 5𝑖
2. Express the matrix A= [ 2𝑖 2+𝑖 4 + 2𝑖 ] as the sum of a Hermitian
−1 + 𝑖 −4 7
matrix and a Skew-Hermitian matrix.

0 1 + 2𝑖
3. Given that A = [ ] , show that (𝐼 + 𝐴)(𝐼 − 𝐴)−1 is unitary.
−1 + 2𝑖 0
Determinant of a matrix: A determinant of a square matrix is a real number associated
with every square matrix. Determinant of a matrix is the sum of the products of elements
of any row or column with their corresponding cofactors.

Determinant of a square matrix A is denoted by det (𝐴) or |𝐴|.

Rank of a matrix: if A is a non-zero matrix, its rank is r if

1. Every (𝑟 + 1)𝑡ℎ and above order minor of 𝐴 is zero.

2. There exists at least on 𝑟𝑡h order minor of 𝐴 which is not zero.

Rank of matrix 𝐴 is denoted by 𝑟(𝐴) or 𝜌(𝐴).

Properties:

1. Rank of a matrix is unique.

2. Rank of Null matrix is zero.
3. Rank of a non-zero matrix is always ≥ 1.
4. If 𝐴 is a matrix of order 𝑚 𝑋 𝑛 them 𝑟(𝐴) ≤ min(𝑚, 𝑛).

Elementary transformations on a matrix:

1. Interchanging any two rows or columns. 𝑅𝑖 ↔ 𝑅𝑗 or 𝐶𝑖 ↔ 𝐶𝑗

2. Multiplying ith row(column) by a scalar k. 𝑅𝑖 → 𝑘𝑅𝑖 or 𝐶𝑖 → 𝑘𝐶𝑖
3. Multiplying all the elements of 𝑗 𝑡ℎ row(column) by k and adding to corresponding
elements of 𝑖 𝑡ℎ row(column). . 𝑅𝑖 → 𝑅𝑖 + 𝑘𝑅𝑗 or 𝐶𝑖 → 𝐶𝑖 + 𝑘𝐶𝑗 .

Echelon form of a matrix:

A matrix is said to be in Echelon form if it has the following properties:

1. Zero rows if any, are below all non-zero rows.

2. No. of zeros before the first non-zero element in a row is less than no. of such zeros
in the next row.

No. of non-zero rows in the Echelon form of a matrix is its rank.

Note: To reduce a given matrix to Echelon form, apply only elementary row operations.
 𝟐 −𝟒 𝟑 −𝟏 𝟎
3. Find rank of the matrix [ 𝟏 −𝟐 −𝟏 −𝟒 𝟐 ] by reducing to echelon form.
𝟎 𝟏 −𝟏 𝟑 𝟏
𝟒 −𝟕 𝟒 −𝟒 𝟓
2 −4 3 −1 0
Solution: Let A= [ 1 −2 −1 −4 2]
0 1 −1 3 1
4 −7 4 −4 5
Applying 𝑅1 ↔ 𝑅2
1 −2 −1 −4 2
[ 2 −4 3 −1 0 ]
0 1 −1 3 1
4 −7 4 −4 5
Applying 𝑅2 → 𝑅2 − 2𝑅1 , 𝑅4 → 𝑅4 − 4𝑅1
1 −2 −1 −4 2
[ 0 0 5 7 −4]
0 1 −1 3 1
0 1 8 12 −3
Applying 𝑅2 ↔ 𝑅4
1 −2 −1 −4 2
[ 0 1 8 12 −3 ]
0 1 −1 3 1
0 0 5 7 −4
1 −2 −1 −4 2
Applying 𝑅3 → 𝑅3 − 𝑅2 [ 0 1 8 12 −3 ]
0 0 −9 −9 4
0 0 5 7 −4
1 −2 −1 −4 2
Applying 𝑅3 → 𝑅3 + 2𝑅4 [ 0 1 8 12 −3 ]
0 0 1 5 −4
0 0 5 7 −4
1 −2 −1 −4 2
Applying 𝑅4 → 𝑅4 − 5𝑅3 [ 0 1 8 12 −3 ]
0 0 1 4 −4
0 0 0 −18 16
This is echelon form of the given matrix.
Rank(A)= no. of non-zero rows in echelon form = 4.
 𝟑 𝟒 𝟓 𝟐
4. Find rank of the matrix [ −𝟏 𝟐 𝟎 −𝟑] by reducing into echelon form.
𝟏 −𝟏 𝟏 𝟒
3 4 5 2
Solution: Let A = [ −1 2 0 −3]
1 −1 1 4
1 −1 1 4
Applying 𝑅1 ↔ 𝑅3 [ −1 2 0 −3]
3 4 5 2
Applying 𝑅2 → 𝑅2 + 𝑅1 , 𝑅3 → 𝑅3 − 3𝑅1
1 −1 1 4
[0 1 1 1]
0 7 2 −10
applying 𝑅3 → 𝑅3 − 7𝑅2
1 −1 1 4
[0 1 1 1 ]
0 0 −5 −17
This is in echelon form.
Therefore rank of the given matrix = no. of non-zero rows in echelon form = 3.

Find rank of the following matrices by reducing into echelon form:

𝟏 𝟐 𝟑 𝟎 𝟏 𝟐 𝟑 𝟏
5 A= [𝟐 𝟒 𝟑 𝟐] 6 A = [−𝟐 −𝟒 𝟏 𝟐]
𝟑 𝟐 𝟏 𝟑 −𝟑 −𝟒 𝟓 𝟖
𝟔 𝟖 𝟕 𝟓 𝟏 𝟑 𝟏𝟎 𝟏𝟒
 𝐼𝑟 0
Normal form or canonical form of a Matrix: A matrix of the form [ ] , where 𝐼𝑟
0 0
is a unit matrix of order r and 0 is the null matrix is called the normal form or canonical
form.
Rank of the above canonical form is r.
𝐼
Matrices of the form [ 𝑟 ], [𝐼𝑟 ], [𝐼𝑟 0] are also known as normal forms.
0
Note: Every m X n matrix can be reduced to the normal form by a series of elementary
transformations.

Problems:
𝟑 𝟏 𝟒 𝟔
7. Reduce the matrix A = [𝟐 𝟏 𝟐 𝟒] into normal form and find rank.
𝟒 𝟐 𝟓 𝟖
𝟏 𝟏 𝟐 𝟐

3 1 4 6
Solution: A = [2 1 2 4]
4 2 5 8
1 1 2 2
To get 1 in the first position of first row
1 1 2 2
Applying 𝑅1 ↔ 𝑅4 [2 1 2 4]
4 2 5 8
3 1 4 6
Applying 𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 4𝑅1 , 𝑅4 → 𝑅4 − 3𝑅1

1 1 2 2
[ 0 −1 −2 0]
0 −2 −3 0
0 −2 −2 0
Applying 𝐶2 → 𝐶2 − 𝐶1 , 𝐶3 → 𝐶3 − 2𝐶1, 𝐶4 → 𝐶4 − 2𝐶1
1 0 0 0
[0 −1 −2 0]
0 −2 −3 0
0 −2 −2 0
Applying 𝑅2 → −𝑅2
1 0 0 0
[0 1 2 0]
0 −2 −3 0
0 −2 −2 0
Applying 𝑅3 → 𝑅3 + 2𝑅2 , 𝑅4 → 𝑅4 + 2𝑅2
1 0 0 0
[0 1 2 0]
0 0 1 0
0 0 2 0
Applying 𝐶3 → 𝐶3 + 2𝐶2
1 0 0 0
[0 1 0 0]
0 0 1 0
0 0 2 0
Applying 𝑅4 → 𝑅4 − 2𝑅3
1 0 0 0
[0 1 0 0]
0 0 1 0
0 0 0 0
𝐼3 0
Which is of the normal form [ ]
0 0
Therefore R(A)= 3.
𝟏 −𝟏 𝟐 −𝟑
8. Reduce the matrix A = [𝟒 𝟏 𝟎 𝟐 ] into normal form and find rank.
𝟎 𝟑 𝟎 𝟒
𝟎 𝟏 𝟎 𝟐
1 −1 2 −3
Solution: A = [4 1 0 2]
0 3 0 4
0 1 0 2
Applying 𝑅2 → 𝑅2 − 4𝑅1 ,
1 −1 2 −3
[0 5 −8 14 ]
0 3 0 4
0 1 0 2
Applying 𝐶2 → 𝐶2 + 𝐶1 , 𝐶3 → 𝐶3 − 2𝐶1, 𝐶4 → 𝐶4 + 3𝐶1
1 0 0 0
[0 5 −8 14 ]
0 3 0 4
0 1 0 2
Applying 𝑅2 ↔ 𝑅4

1 0 0 0
[0 1 0 2]
0 3 0 4
0 5 −8 14
Applying 𝑅3 → 𝑅3 − 3𝑅2 , 𝑅4 → 𝑅4 − 5𝑅2
1 0 0 0
[0 1 0 2]
0 0 0 −2
0 0 −8 4
Applying, 𝐶4 → 𝐶4 − 2𝐶1

1 0 0 0
[0 1 0 0]
0 0 0 −2
0 0 −8 4
Applying 𝑅3 ↔ 𝑅4
1 0 0 0
[0 1 0 0]
0 0 −8 4
0 0 0 −2
Applying 𝐶4 → 𝐶4 /(−8)
1 0 0 0
[0 1 0 0]
0 0 1 4
0 0 0 −2
Applying 𝐶4 → 𝐶4 − 4𝐶1
1 0 0 0
[0 1 0 0]
0 0 1 0
0 0 0 −2
Applying 𝑅4 → 𝑅4 /(−2)
1 0 0 0
[0 1 0 0]
0 0 1 0
0 0 0 1
Which is of the normal form 𝐼4 .
Therefore R(A) = 4.
Reduce the following matrices into normal form and find rank.
𝟐 𝟑 −𝟏 −𝟏 𝟏 𝟐 −𝟐 𝟑
9 A= [ 𝟏 −𝟏 −𝟐 −𝟒 ] 10 A= [ 𝟐 𝟓 −𝟒 𝟔]
𝟑 𝟏 𝟑 −𝟐 −𝟏 −𝟑 𝟐 −𝟐
𝟔 𝟑 𝟎 −𝟕 𝟐 𝟒 −𝟏 𝟔

PAQ form :
To find two non-singular matrices P and Q such that PAQ is in normal form where A is
any 𝑚 𝑋 𝑛 matrix.
1. Write A = 𝐼𝑚 𝐴 𝐼𝑛 where 𝐼𝑚 and 𝐼𝑛 are unit matrices of orders m and n
respectively.
2. Apply elementary row and column operations on A of LHS and reduce it to normal
form. Perform the same row operations on 𝐼𝑚 of RHS and the same column
operations on 𝐼𝑛 of RHS.

Problems:
Obtain two non-singular matrices P and Q such that PAQ is in normal form
𝑰 𝟎
[ 𝒓 ] and hence obtain its rank.
𝟎 𝟎
𝟏 𝟑 𝟔 −𝟏
1. A = [𝟏 𝟒 𝟓 𝟏]
𝟏 𝟓 𝟒 𝟑

Solution:

Write A = 𝐼3 𝐴 𝐼4
1 0 0 0
1 3 6 −𝟏 1 0 0
𝟏 ] = [0 1 0 ] A [ 0
[1 4 1 0 0]
5 0 0 1 0
1 5 4 𝟑 0 0 1 0 0 0 1
Apply elementary row and column operations on A of LHS and reduce it to normal
form. Perform the same row operations on 𝐼𝑚 of RHS and the same column
operations on 𝐼𝑛 of RHS
Applying 𝑅2 → 𝑅2 − 𝑅1 , 𝑅3 → 𝑅3 − 𝑅1
 1 0 0 0
1 3 6 −1 1 0 0
2 ] = [−1 1 0] A [0
[0 1 1 0 0]
−1 0 0 1 0
0 2 −2 4 −1 0 1 0 0 0 1

Applying 𝐶2 → 𝐶2 − 3𝐶1, 𝐶3 → 𝐶3 − 6𝐶1, 𝐶4 → 𝐶4 + 𝐶1

1 −3 −6 1
1 0 0 0 1 0 0
0] A [0
[0 1 1 0 0]
−1 2] = [−1 1 0 0 1 0
0 2 −2 4 −1 0 1 0 0 0 1

Applying 𝑅3 → 𝑅3 − 2𝑅2

1 −3 −6 1
1 0 0 0 1 0 0
1 0 ] A [0
[0 1 1 0 0]
−1 2] = [−1 0 0 1 0
0 0 0 0 1 −2 1 0 0 0 1

Applying 𝐶3 → 𝐶3 + 𝐶2 , 𝐶4 → 𝐶4 − 2𝐶2
1 −3 −9 7
1 0 0 0 1 0 0
1 0] A [0
[0 1 1 1 −2]
0 0] = [−1 0 0 1 0
0 0 0 0 1 −2 1 0 0 0 1

𝐼2 0
This is of the form [ ] = PAQ
0 0
1 −3 −9 7
1 0 0
1 0], Q = [0
where P = [−1 1 1 −2]
0 0 1 0
1 −2 1
0 0 0 1
Therefore R(A) = 2.
 𝟏 𝟐 𝟑 −𝟐
2. A= [𝟐 −𝟐 𝟏 𝟑]
𝟑 𝟎 𝟒 𝟏
Solution:

Write A = 𝐼3 𝐴 𝐼4
1 0 0 0
1 2 3 −2 1 0 0
0] A [0
[2 −2 1 1 0 0]
3 ] = [0 1 0 0 1 0
3 0 4 1 0 0 1 0 0 0 1
Apply elementary row and column operations on A of LHS and reduce it to normal
form. Perform the same row operations on 𝐼𝑚 of RHS and the same column
operations on 𝐼𝑛 of RHS

Applying 𝑅2 → 𝑅2 − 2𝑅1 , 𝑅3 → 𝑅3 − 3𝑅1

1 0 0 0
1 2 3 −2 1 0 0
[0 −6 −5 − 7] = [−2 1 0] A [0 1 0 0]
0 0 1 0
0 −6 −5 −7 −3 0 1 0 0 0 1

Applying 𝐶2 → 𝐶2 − 3𝐶1, 𝐶3 → 𝐶3 − 3𝐶1 , 𝐶4 → 𝐶4 + 2𝐶1

1 −2 −3 −2
1 0 0 0 1 0 0
[0 −6 −5 −7 ] = [−2 1 0] A [0 1 0 0]
0 0 1 0
0 −6 −5 − 7 −3 0 1 0 0 0 1

Applying 𝑅3 → 𝑅3 − 𝑅2
1 −2 −3 −2
1 0 0 0 1 0 0
1 0] A [00
[0 −6 −5 −7] = [−2 1 0 0]
0 1 0
0 0 0 0 −1 −1 1 0 0 0 1
Applying 𝐶2 → 𝐶2 − 𝐶3 ,
1 1 −3 −2
1 0 0 0 1 0 0
[0 −1 −5 −7] = [−2 0 1 0 0]
1 0] A [0 −1 1 0
0 0 0 0 −1 −1 1 0 0 0 1
Applying 𝑅3 → −𝑅2
1 1 −3 −2
1 0 0 0 1 0 0
[0 0 1 0 0]
1 5 7 ] = [ 2 −1 0] A [0 −1 1 0
0 0 0 0 1 −1 1 0 0 0 1
 Applying 𝐶3 → 𝐶3 − 5𝐶2 , 𝐶4 → 𝐶4 − 7𝐶2
1 1 −8 −9
1 0 0 0 1 0 0
[0 0 1 −5 −7]
1 0 0 ] = [ 2 −1 0] A [0 −1 6 7
0 0 0 0 1 −1 1 0 0 0 1
𝐼2 0
This is of the form [ ] = PAQ where
0 0
1 1 −8 −9
1 0 0
P = [ 2 −1 0], Q = [ 0 1 −5 −7]
0 −1 6 7
1 −1 1 0 0 0 1
Therefore R(A) = 2.

1 −1 −1 2
3. A = [4 2 2 −1]
2 2 0 −2

1 2 3
4. A = [ 2 −1 0]
3 1 2

Finding inverse of a square matrix using Gauss Jordan method:

Suppose A is a non-singular square matrix of order n. then write 𝐴 = 𝐼𝐴 .
Apply elementary row operations on A of LHS to reduce it to 𝐼. Apply the same operations
on 𝐼 of RHS.
Then 𝐴 = 𝐼𝐴 reduces to 𝐼 = 𝐵𝐴.
Therefore, B is the inverse of A.

Problems:
−𝟏 −𝟑 𝟑 −𝟏
1. Find the inverse of A= [ 𝟏 𝟏 −𝟏 𝟎 ] using Gauss Jordan method.
𝟐 −𝟓 𝟐 −𝟑
−𝟏 𝟏 𝟎 𝟏
Solution:
Write 𝐴 = 𝐼𝐴

−1 −3 3 −1 1 0 0 0
[ 1 1 −1 0]= [0 1 0 0] A
2 −5 2 −3 0 0 1 0
−1 1 0 1 0 0 0 1
Applying 𝑅1 → −𝑅1
 1 3 −3 1 −1 0 0 0
[ 1 1 −1 0] = [ 0 1 0 0] A
2 −5 2 −3 0 0 1 0
−1 1 0 1 0 0 0 1
Applying 𝑅2 → 𝑅2 − 𝑅1 , 𝑅3 → 𝑅3 − 2𝑅1, 𝑅4 → 𝑅4 + 𝑅1
1 3 −3 1 −1 0 0 0
[ 0 −2 2 −1] = [ 1 1 0 0] A
0 −11 8 −5 2 0 1 0
0 4 −3 2 −1 0 0 1

1
Applying 𝑅2 → − 2 𝑅2

−1 −3 3 −1 1 0 0 0
1 1 1
0 1 −1 −2 −2 0 0 A
2 =
0 −11 8 −5 2 1 1 0
[ 0 4 −3 2 ] [ −1 0 0 1]

Applying 𝑅1 → 𝑅1 − 3𝑅2 , 𝑅3 → 𝑅3 + 11𝑅2 , 𝑅4 → 𝑅4 − 4𝑅2

1 0 0 −5/2 1/2 3/2 0 0
−1 1/2 −1/2 −1/2
[0 1 ]= [ 0 0] A
0 0 −3 1/2 −7/2 −11/2 1 0
0 0 1 0 1 2 0 1

Applying 𝑅3 ↔ 𝑅4
1 0 0 −5/2 1/2 3/2 0 0
0 1 −1 1/2 −1/2 −1/2 0 0
[ ]= [ ] A
0 0 1 0 1 2 0 1
0 0 −3 1/2 −7/2 −11/2 1 0
Applying 𝑅2 → 𝑅2 + 𝑅3 , 𝑅4 → 𝑅4 + 3𝑅3

1 0 0 −5/2 1/2 3/2 0 0

0 1 0 1/2 −1/2 3/2 0 1
[ ]= [ ] A
0 0 1 0 1 2 0 1
0 0 0 1/2 −1/2 1/2 1 3
 Applying 𝑅1 → 𝑅1 + 𝑅4 , 𝑅2 → 𝑅2 − 𝑅4 ,

1 0 0 0 0 2 1 3
0 1 0 0 1 1 −1 −2
[ 1 0 ]= [ 1 2 ] A
0 0 0 1
0 0 0 1/2 −1/2 1/2 1 3
Applying 𝑅4 → 2𝑅4 ,
1 0 0 0 0 2 1 3
[0 1 0 0] = [ 1 1 −1 −2 ] A
0 0 1 0 1 2 0 1
0 0 0 1 −1 1 2 6
This is of the form 𝐼 = 𝐵𝐴

0 2 1 3
Therefore 𝐴−1 = [ 1 1 −1 −2 ]
1 2 0 1
−1 1 2 6

𝟎 𝟏 𝟐 𝟐
2. Find the inverse of A= [𝟏 𝟏 𝟐 𝟑] using Gauss Jordan method.
𝟐 𝟐 𝟐 𝟑
𝟐 𝟑 𝟑 𝟑
Solution: Write 𝐴 = 𝐼4 𝐴
𝟎 𝟏 𝟐 𝟐 1 0 0 0
[𝟏 𝟏 𝟐 𝟑] = [0 1 0 0] A
𝟐 𝟐 𝟐 𝟑 0 0 1 0
𝟐 𝟑 𝟑 𝟑 0 0 0 1

Applying 𝑅1 ↔ 𝑅2
1 1 2 3 0 1 0 0
[0 1 2 2 ] = [1 0 0 0] A
2 2 2 3 0 0 1 0
2 3 3 3 0 0 0 1
Applying 𝑅3 → 𝑅3 − 2𝑅1 , 𝑅4 → 𝑅4 − 2𝑅1
1 1 2 3 0 1 0 0
[0 1 2 2] = [ 1 0 0 0] A
0 0 −2 −3 0 −2 1 0
0 1 −1 −3 0 −2 0 1
 Applying 𝑅1 → 𝑅1 − 𝑅2 , 𝑅4 → 𝑅4 − 𝑅2
1 0 0 1 −1 1 0 0
[0 1 2 2] = [ 1 0 0 0] A
0 0 −2 −3 0 −2 1 0
0 0 −3 −5 −1 −2 0 1
Applying 𝑅2 → 𝑅2 + 𝑅3 , 𝑅4 → 2𝑅4 − 3𝑅3
1 0 0 1 −1 1 0 0
[0 1 0 −1 ] = [ 1 −2 1 0] A
0 0 −2 −3 0 −2 1 0
0 0 0 −1 −2 2 −3 2
Applying 𝑅1 → 𝑅1 + 𝑅4 , 𝑅2 → 𝑅2 − 𝑅4 , 𝑅3 → 𝑅3 − 3𝑅4
1 0 0 0 −3 3 −3 2
[0 1 0 0 ] = [ 3 −4 4 −2] A
0 0 −2 0 6 −8 10 −6
0 0 0 −1 −2 2 −3 2
𝑅
Applying 𝑅3 → −23 , 𝑅4 → (−1)𝑅4

1 0 0 0 −3 3 −3 2
[0 1 0 0] = [ 3 −4 4 −2] A
0 0 1 0 −3 4 −5 3
0 0 0 1 2 −2 3 −2
This is of the form 𝐼 = 𝐵𝐴
−3 3 −3 2
Therefore 𝐴−1 = [ 3 −4 4 −2]
−3 4 −5 3
2 −2 3 −2
Practice problems:
Compute the inverse or the matrix by Gauss-Jordan method:
2 −1 3 −1 1 2
1. A= [ 1 1 1] Ans: [ 0 1/2 −1/2]
1 −1 1 1 −1/2 −3/2

1 2 1 −1/2 3/4 −1
2. A = [ 3 2 3] Ans: [ 3/4 −1/4 0]
1 1 2 −1/4 −1/2 1

1 0 1 1 −1 −1
3. A = [ −2 1 0] Ans: (1/3) [ 2 1 −2]
0 −1 1 2 1 1
Consistency of the system of equations 𝑨𝑿 = 𝑩:
The system of equations 𝐴𝑋 = 𝐵 is consistent, if R(A/B)= R(A) and the system will have
i) Unique solution if 𝑅(𝐴/𝐵) = 𝑅(𝐴) = no. of variables.
ii) Infinite no. of solutions if 𝑅(𝐴/𝐵) = 𝑅(𝐴) < no. of variables.
The system of equations 𝐴𝑋 = 𝐵 is inconsistent if 𝑅(𝐴/𝐵) ≠ 𝑅(𝐴)

1. Test for consistency and solve the system if it is consistent:

𝒙 + 𝟐𝒚 + 𝟐𝒛 = 𝟐, 𝟑𝒙 − 𝟐𝒚 − 𝒛 = 𝟓, 𝟐𝒙 − 𝟓𝒚 + 𝟑𝒛 = −𝟒, 𝒙 + 𝒚 + 𝟔𝒛 = 𝟎
Solution: given system can be expressed in matrix form as 𝐴𝑋 = 𝐵

1 2 2 𝑥 2
[ 3 −2 −1] [𝑦 ] = [ 5 ]
2 −5 3 −4
𝑧
1 4 6 0

here 𝑛 = no. of variables = 3

1 2 2 2
[𝐴/𝐵] = (3 −2 −1| 5)
2 −5 3 −4
1 4 6 0
Applying 𝑅2 → 𝑅2 − 3𝑅1, 𝑅3 → 𝑅3 − 2𝑅1 , 𝑅4 → 𝑅4 − 𝑅1
1 2 2 2
( 0 −8 −7 | −1)
0 −9 −1 −8
0 2 4 −2
Applying 𝑅2 → 𝑅2 − 𝑅3
1 2 2 2
(0 1 −6 | 7 )
0 −9 −1 8
0 2 4 −2
Applying 𝑅3 → 𝑅3 + 9𝑅2 , 𝑅4 → 𝑅4 − 2𝑅2
1 2 2 2
(0 1 −6 | 7 )
0 0 −55 55
0 0 16 −16
 3 𝑅
Applying 𝑅3 → (−55) , 𝑅4 → 𝑅4 /16

1 2 2 2
( 0 1 −6 | 7 )
0 0 1 −1
0 0 1 −1
1 2 2 2
Applying 𝑅4 → 𝑅4 − 𝑅3 , (0 1 −6| 7 )
0 0 1 −1
0 0 0 0
R[𝐴/𝐵] = 3 and R(A) =3
𝑛 = no. of variables = 3
so 𝑅[𝐴/𝐵] = 𝑅(𝐴) = 𝑛
therefore, the system is consistent and has unique solution.
Now, the system reduces to
𝑥 + 2𝑦 + 2𝑧 = 2 -------------------(1)
𝑦 − 6𝑧 = 7 -------------------(2)
𝑧 = −1 -------------------------(3)
Using backward substitution,

𝑦 − 6𝑧 = 7 gives 𝑦 − 6(−1) = 7 ⇒ 𝑦 = 1
𝑥 + 2𝑦 + 2𝑧 = 2 gives 𝑥 + 2(1) + 2(−1) = 2 ⇒ 𝑥 = 2
Therefore the solution is 𝑥 = 2, 𝑦 = 1, 𝑧 = −1.

2. Test for consistency and solve the system if it is consistent

𝒙 + 𝒚 + 𝒛 = 𝟑, 𝟑𝒙 − 𝟓𝒚 + 𝟐𝒛 = 𝟖, 𝟓𝒙 − 𝟑𝒚 = 𝟒 = 𝟏𝟒
Solution:
given system can be expressed in matrix form as 𝐴𝑋 = 𝐵
1 1 1 𝑥 3
𝑦
[3 −5 2] [ ] = [ 8 ]
5 −3 4 𝑧 14

here 𝑛 = no. of variables = 3

1 1 1 3
[𝐴/𝐵] = (3 −5 2| 8 )
5 −3 4 14
 Applying 𝑅2 → 𝑅2 − 3𝑅1 , 𝑅3 → 𝑅3 − 5𝑅1 ,
1 1 1 3
(0 −8 −1|−1)
0 −8 −1 −1

Applying 𝑅3 → 𝑅3 − 𝑅2
1 1 1 3
(0 −8 −1|−1)
0 0 0 0
This is in echelon form.
R[𝐴/𝐵] = 2 and R[𝐴] = 2
So R[𝐴/𝐵] =R[𝐴] = 2 < 𝑛
Therefore, the system is consistent and has infinite no. of solutions.
Now the system can be written as
𝑥 + 𝑦 + 𝑧 = 3 --------------(1)
−8𝑦 − 𝑧 = −1 --------------(2)
Let 𝑧 = 𝑘 then
1−𝑧 1−𝑘
(2) gives 𝑦= =
8 8
1−𝑘 23−7𝑘
Substituting in (1), we get 𝑥 =3− −𝑘 =
8 8

The solutions are given by

23−7𝑘
𝑥 8
[𝑦] = [ 1−𝑘 ] where 𝑘 is a parameter.
𝑧 8
𝑘
3. Test for consistency and solve the system if it is consistent
𝑥 + 𝑦 + 𝑧 = 6, 𝑥 − 𝑦 + 2𝑧 = 5, 3𝑥 + 𝑦 + 𝑧 = 8, 2𝑥 − 2𝑦 + 3𝑧 = 7.
(Ans: inconsistent)
4. Solve the system if it is consistent:
3𝑥 + 3𝑦 + 2𝑧 = 1, 𝑥 + 2𝑦 = 4, 10𝑦 + 3𝑧 = −2, 2𝑥 − 3𝑦 − 𝑧 = 5.
(Ans: 𝑥 = 2, 𝑦 = 1 , 𝑧 = −4)
Consistency of the system of equations 𝑨𝑿 = 𝟎:
Since b’s are zeros, 𝑅(𝐴/𝐵) = 𝑅(𝐴)
Therefore, the system is always consistent and zero solution is always a solution of the
system which is called trivial solution.
The system will have
i) Only trivial solution if 𝑅(𝐴) = no. of variables.
ii) Infinite no. of non-trivial solutions if 𝑅(𝐴) < no. of variables, it will have (n-
r) linearly independent solutions.
iii) Non trivial solution if the no. of equations is less than the no. of variables.
Problems:
1. Solve completely the system of equations 𝒙 + 𝟑𝒚 − 𝟐𝒛 = 𝟎, 𝟐𝒙 − 𝒚 + 𝟒𝒛 = 𝟎,
𝒙 − 𝟏𝟏𝒚 + 𝟏𝟒𝒛 = 𝟎.
Solution:
given system can be expressed in matrix form as 𝐴𝑋 = 𝑂
1 3 −2 𝑥 0
[2 −1 4 ] [𝑦] = [0]
1 −11 14 𝑧 0
Here 𝑛 = no. of variables = 3
Now we reduce the coefficient matrix A to Echelon form by applying elementary
row operations on it
Performing 𝑅2 → 𝑅2 − 2𝑅1, 𝑅3 → 𝑅3 − 𝑅1 , we have

1 3 −2
A= [0 −7 8]
0 −14 16

Performing 𝑅3 → 𝑅3 − 2𝑅2 , we have

1 3 −2
[0 −7 8]
0 0 0
This is echelon form of A.
𝑟 = 𝑅(𝐴) = 2 < 𝑛.
Therefore, the system will have infinitely many non trivial solutions.
 The given system 𝐴𝑋 = 𝑂 is equivalent to
𝑥 + 3𝑦 − 2𝑧 = 0------(1)
−7𝑦 + 8𝑧 = 0---------(2)
Assuming z = k,
8𝑘
(2) gives 𝑦 = 7
−10𝑘
Substituting in (1), we get 𝑥 = 7

The solutions are given by

−10𝑘
𝑥 7
[𝑦] = [ 8𝑘 ] where 𝑘 is an arbitrary constant.
𝑧 7
𝑘

2. Solve completely the system of equations:

𝟒𝒙 + 𝟐𝒚 + 𝒛 + 𝟑𝒖 = 𝟎, 𝟔𝒙 + 𝟑𝒚 + 𝟒𝒛 + 𝟕𝒖 = 𝟎, 𝟐𝒙 + 𝒚 + 𝒖 = 𝟎
Solution: given system can be expressed in matrix form as 𝐴𝑋 = 𝑂
𝑥
4 2 1 3 0
𝑦
[6 3 4 7] [ 𝑧 ] = [0]
2 1 0 1 𝑢 0

Now we reduce the coefficient matrix A to Echelon form by applying elementary

row operations on it,

4 2 1 3
A= [6 3 4 7]
2 1 0 1
Interchanging 𝑅1 and 𝑅2 , we get

2 1 0 1
[6 3 4 7]
4 2 1 3

Performing 𝑅2 → 𝑅2 − 3𝑅1, 𝑅3 → 𝑅3 − 2𝑅1 , we have

2 1 0 1
[0 0 4 4]
0 0 1 1
Performing 𝑅3 → 𝑅3 − 4𝑅2 , we have
2 1 0 1
[0 0 4 4]
0 0 0 0
 𝑅2
Performing 𝑅2 → , we have
4
2 1 0 1
[0 0 1 1]
0 0 0 0
This is Echelon form of Matrix A.
𝑅(𝐴) = 2 < 𝑛.
The system will have infinitely many non-trivial solutions of which (4 − 2 = 2)
are linearly independent.
Now the system 𝑨𝑿 = 𝑶 is equivalent to
2𝑥 + 𝑦 + 𝑢 = 0 -------------(1)
𝑧+𝑢 =0 ----------------(2)
From (2), we have 𝑧 = − 𝑢.
Let 𝑧 = 𝑘1 , 𝑥 = 𝑘2 then 𝑢 = −𝑘1 ,
Then from (1), 𝑦 = −2𝑥 − 𝑢 = −2𝑘2 + 𝑘1
The general solution is
𝑥 𝑘2 𝑘2 0 1 0
𝑦 −2𝑘2 + 𝑘1 −2𝑘2 𝑘
[𝑧] = [ ]=[ ]+[ 1
] = [ ] 𝑘2 + [ 1 ] 𝑘1 where 𝑘1 and 𝑘2
−2
𝑘1 0 𝑘 1 0 1
𝑢 −𝑘1 0 −𝑘1 0 −1
are arbitrary constants.
1 0
Two linearly independent solutions are [ ] , [ 1 ].
−2
0 1
0 −1

3. Show that the only real number 𝝀 for which the system
𝒙 + 𝟐𝒚 + 𝟑𝒛 = 𝝀𝒙, 𝟑𝒙 + 𝒚 + 𝟐𝒛 = 𝝀𝒚, 𝟐𝒙 + 𝟑𝒚 + 𝒛 = 𝝀𝒛 has non-zero
solution is 6 and solve them when 𝝀 = 𝟔.

Solution: The given system can be written as

(1 − 𝜆)𝑥 + 2𝑦 + 3𝑧 = 0
3𝑥 + (1 − 𝜆)𝑦 + 2𝑧 = 0
2𝑥 + 3𝑦 + (1 − 𝜆)𝑧 = 0
No. of variables= n = 3
This system can be written in Matrix form as 𝐴𝑋 = 𝑂
1−𝜆 2 3 𝑥 0
[ 3 1−𝜆 2 ] [𝑦]= [0]
2 3 1−𝜆 𝑧 0
 Given, the system has non-trivial solution
Therefore, 𝑅(𝐴) < 𝑛
i.e., 𝑅(𝐴) < 3.
Since A is 3 𝑋 3 matrix, det(A)=0
1−𝜆 2 3
| 3 1−𝜆 2 | = 0 ⇒ 𝜆3 − 3𝜆2 − 15𝜆 − 18 = 0
2 3 1−𝜆
−3±√−3
⇒ 𝜆 = 6, 𝜆 = 2

Here 𝜆 = 6 is the real value of 𝜆.

When 𝜆 = 6 the given system of equations will have a non-trivial solution.
For 𝜆 = 6, the system becomes
−5 2 3 𝑥 0
[ 3 −5 2 ] [𝑦]= [0]
2 3 −5 𝑧 0
Reducing the coefficient matrix A to Echelon form,
−5 2 3
A = [ 3 −5 2 ]
2 3 −5

1 −8 7
We get [0 −1 1]
0 0 0

Now the system AX = O becomes

𝑥 − 8𝑦 + 7𝑧 = 0
−𝑦 + 𝑧 = 0
Solving these equations, we get 𝑦 = 𝑧 and 𝑥 = 𝑧.

Let 𝑧 = 𝑘
𝑥 𝑘 1
Then [𝑦] = [𝑘] = 𝑘 [1]
𝑧 𝑘 1
1
when 𝜆 = 6, non-trivial solution is [1]
1
4. Solve the system of equations 𝟐𝒙 − 𝒚 + 𝟑𝒛 = 𝟎, 𝟑𝒙 + 𝟐𝒚 + 𝒛 = 𝟎, 𝒙 − 𝟒𝒚 +
𝟓𝒛 = 𝟎.
Ans; (-1, 1, 1)
5. Determine whether the system of equations 𝒙 + 𝒚 + 𝒘 = 𝟎, 𝒚 + 𝒛 = 𝟎. 𝒙 +
𝒚 + 𝒛 + 𝒘 = 𝟎, 𝒙 + 𝒚 + 𝟐𝒛 = 𝟎 have non-trivial solution, if so find the
solution.
Ans; (-10/7, 8/7, 1)
Gauss elimination method:
In this method we reduce the system of equations to an equivalent upper triangular system
which can be solved using back substitution.
Consider the system of equations AX = B
𝑎11 𝑥 + 𝑎12 𝑦 + 𝑎13 𝑧 = 𝑏1
𝑎21 𝑥 + 𝑎22 𝑦 + 𝑎23 𝑧 = 𝑏2
𝑎31 𝑥 + 𝑎32 𝑦 + 𝑎33 𝑧 = 𝑏3
This system can be written in Matrix form as 𝐴𝑋 = 𝐵

𝑎11 𝑎12 𝑎13 𝑏1

The augmented matrix (𝐴|𝐵) 𝑎
= ( 21 𝑎22 𝑎23|𝑏2)
𝑎31 𝑎32 𝑎33 𝑏3

Reduce the matrix (𝐴|𝐵) to upper triangular form by applying elementary row operations.
If 𝑎11 = 0, two rows can be interchanged so that 𝑎11 ≠ 0.
In the first stage making 𝑎21 , 𝑎31 to zeroes.
𝑎 𝑎
Applying 𝑅2 → 𝑅2 − (𝑎21 ) 𝑅1 , and 𝑅3 → 𝑅3 − 𝑎31 𝑅1
11 11

the augmented matrix reduces to

𝑎11 𝑎12 𝑎13 𝑏1
( 0 𝑎′22 𝑎′23|𝑏′2 )
0 𝑎′32 𝑎′33 𝑏′3

In the second stage making 𝑎′32 to zero,

𝑎′32
Applying 𝑅3 → 𝑅3 − 𝑅2 ,
𝑎′22

The augmented matrix becomes

𝑎11 𝑎12 𝑎13 𝑏1
( 0 𝑎′22 𝑎′23 | 𝑏′2 )
0 0 𝑎′′33 𝑏′′3

which is an upper triangular matrix.

Now the system becomes
𝑎11 𝑥 + 𝑎12 𝑦 + 𝑎13 𝑧 = 𝑏1
𝑎′22 𝑦 + 𝑎′23 𝑧 = 𝑏′2
𝑎′′33 𝑧 = 𝑏′′3
Then 𝑧 = 𝑏′′3 /𝑎′′33

Obtain the values of 𝑦, 𝑧 using back substitution.

Problems:
1. Solve the system of equations using Gauss elimination method.
𝒙 + 𝒚 + 𝒛 = 𝟔, 𝟑𝒙 + 𝟑𝒚 + 𝟒𝒛 = 𝟐𝟎, 𝟐𝒙 + 𝒚 + 𝟑𝒛 = 𝟏𝟑

Solution:

Given system can be expressed in matrix form as AX=B

1 1 1 𝑥 6
𝑦
[3 3 4] [ ] = [20]
2 1 3 𝑧 13
The augmented matrix is
1 1 1 6
(𝐴|𝐵) = (3 3 4 |20)
2 1 3 13
Applying 𝑅2 → 𝑅2 − 3𝑅1 , 𝑅3 → 𝑅3 − 2𝑅1 , we have

1 1 1 6
(0 0 1 |2)
0 −1 1 1
Applying 𝑅2 ↔ 𝑅3 , we have
1 1 1 6
(0 −1 1 |1)
0 0 1 2

Now the system becomes,

𝑥+𝑦+𝑧 =6
−𝑦 + 𝑧 = 1
𝑧=2
Using back substitution, we get
𝑧 = 2.
−𝑦 + 𝑧 = 1 ⇒ 𝑦 = 𝑧 − 1 = 1
𝑥 + 𝑦 + 𝑧 = 6 ⇒ x = 6 − y − z = 3.
The solution is 𝑥 = 3, 𝑦 = 1, 𝑧 = 2.
2. Solve the system of equations using Gauss elimination method:

𝑥1 − 𝑥2 + 3𝑥3 = 5, 2𝑥1 − 4𝑥2 + 7𝑥3 = 7, 2𝑥1 − 9𝑥2 + 2𝑥3 = −15.

3. Solve the system of equations using Gauss elimination method:

𝟐𝒙 + 𝒚 + 𝒛 = 𝟏𝟎, 𝟑𝒙 + 𝟐𝒚 + 𝟑𝒛 = 𝟏𝟖, 𝒙 + 𝟒𝒚 + 𝟗𝒛 = 𝟏𝟔

(Ans: 𝑥 = 7, 𝑦 = −9, 𝑧 = 5)
 UNIT-2

Eigenvalues and Eigenvectors

Let 𝐴 be a square matrix. A non-zero vector 𝑋 is said to be an Eigenvector of a matrix 𝐴 if
there exists a scalar λ such that 𝐴𝑋 = 𝜆𝑋. The scalar λ is called Eigenvalue of A
corresponding to the Eigenvector 𝑋.
The equation |(𝐴 − 𝜆)𝑋| = 0 is called characteristic equation of A. Roots of characteristic
equation are called characteristic roots or Eigenvalues or Latent roots of Matrix A.
Note: If A = (𝑎𝑖𝑗 )3 𝑥 3 then

|(𝐴 − 𝜆𝐼)| = −λ3 + (𝑎11 + 𝑎22 + 𝑎33 )λ2 − (𝑎11 𝑎22 + 𝑎22 𝑎33 + 𝑎33 𝑎11 )λ + |𝐴|.
Problems:
1. Determine the characteristic roots and the corresponding characteristic vectors
𝟖 −𝟔 𝟐
of the matrix A=[−𝟔 𝟕 −𝟒].
𝟐 −𝟒 𝟑
Sol: Characteristic equation of A is |A − λI| = 0
8−λ −6 2
⇒ | −6 7 − λ −4 | = 0
2 −4 3 − λ
⇒−λ3 + 18λ2 − 45λ = 0
⇒ λ = 0, 15, 3
Therefore, the characteristic roots or eigenvalues are 0, 15, 3
𝑥
Let 𝑋 = [𝑦] be the characteristic vector corresponding to the characteristic root λ
𝑧
Consider the equation (𝐴 − λI)X = 0
8−λ −6 2 𝑥 0
⇒[ −6 7−λ −4 ] [𝑦 ] = [ 0]
2 −4 3−λ 𝑧 0
Case(i):
When λ = 0
8−0 −6 2 𝑥 0
(𝐴 − λI)X = 0 ⇒ [ −6 7 − 0 −4 ] [𝑦] = [0]
2 −4 3−0 𝑧 0
8 −6 2 𝑥 0
⇒[−6 𝑦
7 −4 ] [ ] = [0]
2 −4 3 𝑧 0
Applying 𝑅1 ↔ 𝑅3
 2 −4 3 𝑥 0
⇒[ −6 7 −4] [𝑦] = [0]
8 −6 2 𝑧 0
Applying 𝑅2 → 𝑅2 + 3𝑅1 ; 𝑅3 → 𝑅3 − 4𝑅1
2 −4 3 𝑥 0
⇒[0 −5 5 ] [𝑦] = [0]
0 10 −10 𝑧 0
Applying 𝑅3 → 𝑅3 + 2𝑅2
2 −4 3 𝑥 0
⇒[0 𝑦
−5 5] [ ] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
2𝑥 − 4𝑦 + 3𝑧 = 0
−5𝑦 + 5𝑧 = 0
Let 𝑧 = 𝑘 then −5𝑦 + 5𝑧 = 0 ⇒−5𝑦 + 5𝑘 = 0 ⇒𝑦 = 𝑘
Substituting the values of 𝑦, 𝑧 in 2𝑥 − 4𝑦 + 3𝑧 = 0, We get
𝑘
⇒2𝑥 − 4𝑘 + 3𝑘 = 0 ⇒𝑥 = 2

𝑥 𝑘/2 1
𝑘
Therefore, 𝑋 = [𝑦] = [ 𝑘 ] = 2 [2] is the eigenvector corresponding to the
𝑧 𝑘 2
eigenvalue λ = 0.
Case(ii):
When λ = 15
8 − 15 −6 2 𝑥 0
(𝐴 − λI)X = 0 ⇒ [ −6 7 − 15 −4 ] [𝑦] = [0]
2 −4 3 − 15 𝑧 0
−7 −6 2 𝑥 0
⇒[−6 −8 −4 ] [𝑦] = [0]
2 −4 −12 𝑧 0
Applying 𝑅1 ↔ 𝑅3
2 −4 −12 𝑥 0
⇒[−6 −8 −4 ] [𝑦 ] = [ 0]
−7 −6 2 𝑧 0
1 1
Applying 𝑅1 → 𝑅1 ; 𝑅2 → − 𝑅2
2 2

1 −2 −6 𝑥 0
⇒[ 3 4 𝑦
2 ] [ ] = [0]
−7 −6 2 𝑧 0
Applying 𝑅2 → 𝑅2 − 3𝑅1 ; 𝑅3 → 𝑅3 + 7𝑅1
1 −2 −6 𝑥 0
⇒[0 10 20 ] [𝑦] = [0]
0 −20 −40 𝑧 0
1 1
Applying 𝑅1 → 10 𝑅1 ; 𝑅2 → − 20 𝑅2

1 −2 −6 𝑥 0
⇒[0 1 2 ] [𝑦] = [0]
0 1 2 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
1 −2 −6 𝑥 0
⇒[0 1 2 ] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥 − 2𝑦 − 6𝑧 = 0
𝑦 + 2𝑧 = 0
Let 𝑧 = 𝑘 then 𝑦 + 2𝑧 = 0 ⇒𝑦 + 2𝑘 = 0 ⇒𝑦 = −2𝑘
Substituting 𝑦, 𝑧 values in 𝑥 − 2𝑦 − 6𝑧 = 0, We get
⇒𝑥 − 2(−2𝑘) − 6𝑘 = 0 ⇒𝑥 = 2𝑘
𝑥 2𝑘 2
Therefore, 𝑋 = [𝑦] = [−2𝑘] = 𝑘 [−2] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 15.
Case(iii):
Consider the equation (𝐴 − λI)X = 0
8−λ −6 2 𝑥 0
⇒[ −6 7−λ 𝑦
−4 ] [ ] = [0]
2 −4 3−λ 𝑧 0
When λ = 3
8−3 −6 2 𝑥 0
[ −6 7−3−4 ] [𝑦] = [0]
2 3−3 𝑧
−4 0
5 −6 2 𝑥 0
⇒[−6 4 −4] [𝑦] = [0]
2 −4 0 𝑧 0
Applying 𝑅1 ↔ 𝑅3
2 −4 0 𝑥 0
⇒[−6 4 −4] [𝑦] = [0]
5 −6 2 𝑧 0
 1 1
Applying 𝑅1 → 2 𝑅1 ; 𝑅2 → 2 𝑅2

1 −2 0 𝑥 0
⇒[−3 𝑦
2 −2] [ ] = [0]
5 −6 2 𝑧 0
Applying 𝑅2 → 𝑅2 + 3𝑅1 ; 𝑅3 → 𝑅3 − 5𝑅1
1 −2 0 𝑥 0
⇒[0 −4 −2] [𝑦] = [0]
0 4 2 𝑧 0
1 1
Applying 𝑅1 → − 2 𝑅1 ; 𝑅2 → 2 𝑅2

1 −2 0 𝑥 0
⇒[0 2 1 ] [𝑦 ] = [0]
0 2 1 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
1 −2 0 𝑥 0
⇒[0 2 1] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥 − 2𝑦 = 0
2𝑦 + 𝑧 = 0
−1
Let 𝑧 = 𝑘 then 2𝑦 + 𝑧 = 0 ⇒2𝑦 + 𝑘 = 0 ⇒𝑦 = 𝑘
2

Substituting 𝑦, 𝑧 values in 𝑥 − 2𝑦 = 0, We get

−1
⇒𝑥 − 2 ( 2 𝑘) = 0 ⇒𝑥 = −𝑘

𝑥 −𝑘 −2
−1 1
Therefore, 𝑋 = [𝑦] = [ 2 𝑘] = 2 𝑘 [−1] is the eigenvector corresponding to the
𝑧 𝑘 2
eigenvalue λ = 3.
1 2 −2
Hence, [2] , [−2] , [−1] are eigenvectors corresponding to the eigenvalues
2 1 2
λ = 0,15, 3 respectively.
 −𝟐 𝟐 −𝟑
2. Find the eigenvalues and the corresponding eigenvectors of A=[ 𝟐 𝟏 −𝟔].
−𝟏 −𝟐 𝟎
Sol: Characteristic equation of A is |A − λI| = 0
−2 − λ 2 −3
⇒| 2 1−λ −6 | = 0
−1 −2 0−λ
⇒−λ3 + (−2 + 1 + 0)λ2 − [−2 + 0 + 0 − (4 + 3 + 12)λ + |𝐴| = 0
⇒−λ3 − λ2 + 21λ + 45 = 0
⇒λ3 + λ2 − 21λ − 45 = 0
⇒ λ = 5, −3, −3
Therefore the eigenvalues are 5, −3, −3
𝑥
Let 𝑋 = [𝑦] be the eigenvector corresponding to the eigenvalue λ
𝑧
Case(i):
Consider the equation (𝐴 − λI)X = 0
−2 − λ 2 −3 𝑥 0
⇒[ 2 1−λ −6 ] [𝑦 ] = [ 0]
−1 −2 0−λ 𝑧 0
When λ = 5
−2 − 5 2 −3 𝑥 0
[ 2 1 − 5 −6 ] [𝑦] = [0]
−1 −2 0 − 5 𝑧 0
−7 2 −3 𝑥 0
[ 2 −4 −6] [𝑦] = [0]
−1 −2 −5 𝑧 0
Applying 𝑅1 ↔ 𝑅3
−1 −2 −5 𝑥 0
⇒[ 2 −4 −6] [ 𝑦 ] = [ 0]
−7 2 −3 𝑧 0
Applying 𝑅1 → −𝑅1
1 2 5 𝑥 0
⇒[ 2 −4 −6] [𝑦] = [0]
−7 2 −3 𝑧 0
Applying 𝑅2 → 𝑅2 − 2𝑅1 ; 𝑅3 → 𝑅3 + 7𝑅1
1 2 5 𝑥 0
⇒[0 −8 −16] [𝑦] = [0]
0 16 32 𝑧 0
 1
Applying 𝑅2 → − 8 𝑅2 ; 𝑅3 → 𝑅3 /16

1 2 5 𝑥 0
⇒[0 1 2] [𝑦] = [0]
0 1 2 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
1 2 5 𝑥 0
⇒[0 1 2 ] [ 𝑦 ] = [ 0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥 + 2𝑦 + 5𝑧 = 0
𝑦 + 2𝑧 = 0
Let 𝑧 = 𝑘 then 𝑦 + 2𝑧 = 0 ⇒𝑦 + 2𝑘 = 0 ⇒𝑦 = −2𝑘
Substituting 𝑦, 𝑧 values in 𝑥 + 2𝑦 + 5𝑧 = 0, We get
⇒𝑥 + 2(−2𝑘) + 5𝑘 = 0 ⇒𝑥 = −𝑘
𝑥 −𝑘 −1
Therefore, 𝑋 = [𝑦] = [−2𝑘] = 𝑘 [−2] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 5.
Case(ii):
Consider the equation (𝐴 − λI)X = 0
−2 − λ 2 −3 𝑥 0
⇒[ 2 1−λ −6 ] [𝑦] = [0]
−1 −2 0−λ 𝑧 0
When λ = −3
−2 + 3 2 −3 𝑥 0
[ 2 1 + 3 −6 ] [𝑦] = [0]
−1 −2 0 + 3 𝑧 0
1 2 −3 𝑥 0
[2 4 𝑦
−6] [ ] = [0]
−1 −2 3 𝑧 0
Applying 𝑅2 → 𝑅2 − 2𝑅1 ; 𝑅3 → 𝑅3 + 𝑅1
1 2 −3 𝑥 0
⇒[0 0 0 ] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥 + 2𝑦 − 3𝑧 = 0
Let 𝑧 = 𝑘1 and 𝑦 = 𝑘2
Substituting 𝑦, 𝑧 values in 𝑥 + 2𝑦 − 3𝑧 = 0, we get
⇒𝑥 + 2𝑘2 − 3𝑘1 = 0 ⇒𝑥 = 3𝑘1 − 2𝑘2
𝑥 3𝑘1 − 2𝑘2 3 −2
Therefore, 𝑋 = [𝑦] = [ 𝑘2 ] = 𝑘1 [0] + 𝑘2 [ 1 ] is the eigenvector
𝑧 𝑘1 1 0
corresponding to the eigenvalue λ = −3.
−1 3 −2
Hence, [−2] , [0] , [ 1 ] are eigenvectors corresponding to the eigenvalues
1 1 2
λ = 5, −3, −3 respectively.
Properties of eigenvalues and eigenvectors of a matrix:
Theorem- 2
Theorem-3
Theorem-4
Theorem- 5
Theorem-6
Theorem- 7
Theorem-8
Theorem-9
Theorem-10
Theorem-11
Theorem-12
13. Eigenvalues of unitary matrix are ±𝟏.
14. Eigenvalues of orthogonal matrix are of unit modulus.
15. Eigenvectors corresponding to two distinct eigenvalues are linearly independent.
16. if λ is an eigenvalue of a matrix 𝑨 and 𝒇(𝑨) is a polynomial in 𝑨, then 𝒇(𝝀) is an
eigenvalue of 𝒇(𝑨).
17. In a real symmetric matrix the eigenvectors corresponding to two distinct
eigenvalues are orthogonal.
𝟏
18. If λ is an eigenvalue of an orthogonal matrix, then is also an eigenvalue.
𝛌

1 1 1
Problem: If A = ( 0 2 1), find eigenvalues of
−4 4 3
1) 𝐴𝑇 2) 𝐴 + 3𝐼 3) 𝐴−1 4) adj (A) 5) 5A 6) 𝐴3 + 5𝐴2 − 2𝐴+3I
Diagonalization of a square Matrix:

An 𝑛 × 𝑛 matrix A is diagonalizable if it is similar to a diagonal matrix: that is, if there exists

an invertible 𝑛 × 𝑛 matrix P such that 𝑃−1 𝐴𝑃 is a diagonal matrix.

Suppose A is a square matrix of order n and λ be an eigenvalue of A.

1) If λ is and eigenvalue of order ‘p’ then ‘p’ is called algebraic multiplicity of λ.
2) No. of linearly independent eigenvectors corresponding to the eigenvalue λ is called
its geometric multiplicity.

Conditions for diagonalizability of a square matrix:

1. A square matrix is diagonalizable if all is eigenvalues are distinct.
or
2. If eigenvalues are not distinct, then the matrix is diagonalizable if for every
eigenvalue, its algebraic multiplicity is equal to geometric multiplicity.

Steps to diagonalize a square matrix.

Let A be a square matrix of order 3.
1. Find eigenvalues and corresponding eigenvectors of A.
2. Check the condition of diagonalizability.
3. If a is diagonalizable, write the Modal matrix P = [𝑥1 𝑥2 𝑥3 ] where 𝑥1, 𝑥2 , 𝑥3 are
eigenvectors of A.
4. Find 𝑃−1.
5. Find 𝐷 = 𝑃−1 𝐴𝑃. This is the diagonal form of A.

To find higher powers of A: find 𝐴𝑚 , using the relation

𝐴𝑚 = 𝑃𝐷𝑚 𝑃 −1

𝟐 𝟑 𝟒
1. Show that the matrix 𝑨 = [𝟎 𝟐 −𝟏] cannot be diagonalized.
𝟎 𝟎 𝟏
Sol: Characteristic equation of A is |A − λI| = 0
2−λ 3 4
⇒| 0 2 − λ −1 | = 0
0 0 1−λ
⇒(2 − λ)[(2 − λ)(1 − λ)] = 0
⇒ λ = 2, 2, 1
Therefore the eigenvalues are 2, 2, 1.
𝑥
Let 𝑋 = [𝑦] be the eigenvector corresponding to the eigenvalue λ
𝑧
Case(i):
Consider the equation (𝐴 − λI)X = 0
2−λ 3 4 𝑥 0
⇒[ 0 2−λ −1 ] [𝑦] = [0]
0 0 1−λ 𝑧 0
When λ = 2
2−2 3 4 𝑥 0
[ 0 𝑦
2 − 2 −1 ] [ ] = [0]
0 0 1−2 𝑧 0
0 3 4 𝑥 0
[0 0 −1] [𝑦] = [0]
0 0 −1 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
0 3 4 𝑥 0
⇒[0 0 −1] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
3𝑦 + 4𝑧 = 0
−𝑧 = 0 ⇒𝑧 = 0
Substituting 𝑧 = 0 in 3𝑦 + 4𝑧 = 0, we get
3𝑦 = 0 ⇒𝑦 = 0
Let 𝑥 = 𝑘
𝑥 𝑘 1
Therefore, 𝑋 = [𝑦] = [0] = 𝑘 [0] is the eigenvector corresponding to the eigenvalue
𝑧 0 0
λ = 2.
For eigenvalue λ = 2, the geometric multiplicity is 1

and the algebraic multiplicity is 2.

Since the algebraic multiplicity is not equal to geometric multiplicity, matrix A is not
diagonalizable.
 𝟏 𝟎 −𝟏
2. Find a matrix P which transform the matrix 𝑨 = [𝟏 𝟐 𝟏 ] to diagonal form.
𝟐 𝟐 𝟑
Hence calculate 𝑨𝟒 .

Sol: Characteristic equation of A is |A − λI| = 0

1−λ 0 −1
⇒| 1 2−λ 1 |=0
2 2 3−λ
⇒−λ3 + (1 + 2 + 3)λ2 − [2 + 6 + 3 − (0 − 2 + 2)]λ + |𝐴| = 0
⇒−λ3 + 6λ2 − 11λ + 6 = 0
⇒λ3 − 6λ2 + 11λ − 6 = 0
⇒ λ = 1, 2, 3
Therefore the eigenvalues are 1, 2, 3.
𝑥
Let 𝑋 = [𝑦] be the eigenvector corresponding to the eigenvalue λ
𝑧

Case(i):
Consider the equation (𝐴 − λI)X = 0
1−λ 0 −1 𝑥 0
⇒[ 1 2−λ 1 ] [𝑦 ] = [ 0]
2 2 3−λ 𝑧 0
When λ = 1
1−1 0 −1 𝑥 0
[ 1 2−1 1 ] [𝑦] = [0]
2 2 3−1 𝑧 0
0 0 −1 𝑥 0
⇒[1 1 1 ] [𝑦] = [0]
2 2 2 𝑧 0
Applying, 𝑅3 → 𝑅3 − 2𝑅2
0 0 −1 𝑥 0
⇒[1 1 1 ] [𝑦 ] = [0]
0 0 0 𝑧 0
Applying 𝑅1 ↔ 𝑅2
1 1 1 𝑥 0
⇒[0 0 −1] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥+𝑦+𝑧 =0
−𝑧 = 0 ⇒ 𝑧 = 0
Substituting 𝑧 = 0 in 𝑥 + 𝑦 + 𝑧 = 0, we get
𝑥+𝑦 =0
Let 𝑦 = 𝑘 then 𝑥 + 𝑦 = 0 ⇒𝑥 + 𝑘 = 0 ⇒𝑥 = −𝑘
𝑥 −𝑘 −1
𝑦
Therefore, 𝑋 = [ ] = [ 𝑘 ] = 𝑘 [ 1 ] is the eigenvector corresponding to the
𝑧 0 0
eigenvalue λ = 1.
Case(ii):
Consider the equation (𝐴 − λI)X = 0
1−λ 0 −1 𝑥 0
⇒[ 1 2−λ 1 ] [𝑦] = [0]
2 2 3−λ 𝑧 0
When λ = 2
1−2 0 −1 𝑥 0
[ 1 2−2 1 ] [ 𝑦 ] = [ 0]
2 2 3−2 𝑧 0
−1 0 −1 𝑥 0
⇒[ 1 0 1 ] [𝑦] = [0]
2 2 1 𝑧 0
Applying, 𝑅2 → 𝑅2 + 𝑅1
−1 0 −1 𝑥 0
⇒[ 0 0 0 ] [𝑦] = [0]
2 2 1 𝑧 0
Applying 𝑅2 ↔ 𝑅3 ; 𝑅1 → 𝑅1 /(−1)
1 0 1 𝑥 0
⇒[2 2 1 ] [ 𝑦 ] = [ 0]
0 0 0 𝑧 0
Applying 𝑅2 → 𝑅2 − 2𝑅1
1 0 1 𝑥 0
⇒[0 2 −1] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥+𝑧 =0
2𝑦 − 𝑧 = 0
Let 𝑧 = 𝑘 then 𝑥 + 𝑧 = 0 ⇒𝑥 + 𝑘 = 0 ⇒𝑥 = −𝑘
and 2𝑦 − 𝑧 = 0 ⇒2𝑦 − 𝑘 = 0 ⇒𝑦 = 𝑘/2
 𝑥 −𝑘 −2
𝑘
Therefore, 𝑋 = [𝑦] = [𝑘/2] = 2 [ 1 ] is the eigenvector corresponding to the
𝑧 𝑘 2
eigenvalue λ = 2.
Case(iii):
Consider the equation (𝐴 − λI)X = 0
1−λ 0 −1 𝑥 0
⇒[ 1 2−λ 1 ] [𝑦] = [0]
2 2 3−λ 𝑧 0
When λ = 3
1−3 0 −1 𝑥 0
[ 1 2−3 1 ] [ 𝑦 ] = [ 0]
2 2 3−3 𝑧 0
−2 0 −1 𝑥 0
⇒[ 1 −1 1 ] [𝑦] = [0]
2 2 0 𝑧 0
Applying 𝑅1 ↔ 𝑅2
1 −1 1 𝑥 0
⇒[−2 0 −1] [𝑦] = [0]
2 2 0 𝑧 0
Applying 𝑅2 → 𝑅2 + 2𝑅1 ; 𝑅3 → 𝑅3 − 2𝑅1
1 −1 1 𝑥 0
⇒[0 −2 1 ] [𝑦] = [0]
0 4 −2 𝑧 0
Applying, 𝑅3 → 𝑅3 + 2𝑅2
1 −1 1 𝑥 0
⇒[0 −2 1] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥−𝑦+𝑧 =0
−2𝑦 + 𝑧 = 0
1
Let 𝑧 = 𝑘 then −2𝑦 + 𝑧 = 0 ⇒−2𝑦 + 𝑘 = 0 ⇒𝑦 = 2 𝑘

Substituting 𝑦, 𝑧 values in 𝑥 − 𝑦 + 𝑧 = 0, we get

1
⇒𝑥 − 𝑘 + 𝑘 = 0 ⇒𝑥 = −𝑘/2
2

𝑥 −𝑘/2 −1
𝑘
Therefore, 𝑋 = [𝑦] = [ 𝑘/2 ] = 2 [ 1 ] is the eigenvector corresponding to the
𝑧 𝑘 2
eigenvalue λ = 3.
 −1 −2 −1
Hence 𝑋1 = [ 1 ] , 𝑋2 = [ 1 ] , 𝑋3 = [ 1 ] are eigenvectors corresponding to the
0 2 2
eigenvalues λ = 1,2, 3 respectively.
Therefore A is diagonalizable
−1 −2 −1
Now the modal matrix 𝑃 = [𝑋1 𝑋2 𝑋3 ] = [ 1 1 1]
0 2 2
0 2 −1 0 1 −1/2
−1 1
and 𝑃 = [−2 −2 0 ] = [−1 −1 0 ]
2
2 2 1 1 1 1/2
1
0 1 − 2 1 0 −1 −1 −2 −1
−1
Consider 𝑃 𝐴𝑃 = [−1 −1 0 ] [1 2 1 ] [ 1 1 1]
1
1 1 2 2 3 0 2 2
2

1 0 0
= [0 2 0] = 𝐷
0 0 3
Now 𝐴 = 𝑃𝐷𝑃 −1 then 𝐴4 = 𝑃𝐷4 𝑃−1 =
1
−1 −2 −1 1 0 0 0 1 −2
[1 1 1 ] [0 16 0 ] [−1 −1 0 ]
1
0 2 2 0 0 81 1 1 2

−49 −50 −40

= [ 65 66 40 ]
130 130 81

−𝟗 𝟒 𝟒
3. Show that the matrix 𝑨 = [ −𝟖 𝟑 𝟒] is diagonalizable. Also find the diagonal
−𝟏𝟔 𝟖 𝟕
form and a modal matrix P.

Sol: Characteristic equation of A is |A − λI| = 0

−9 − λ 4 4
⇒ | −8 3−λ 4 |=0
−16 8 7−λ
⇒−λ3 + (−9 + 3 + 7)λ2 − [−27 + 21 − 63 − (−32 − 64 + 32)]λ + |𝐴| = 0
⇒−λ3 + λ2 + 5λ + 3 = 0
⇒λ3 − λ2 − 5λ − 3 = 0
⇒ λ = −1, −1, 3
Therefore the eigenvalues are −1, −1, 3.
 𝑥
Let 𝑋 = [𝑦] be the eigenvector corresponding to the eigenvalue λ
𝑧
Case(i):

Consider the equation (𝐴 − λI)X = 0

−9 − λ 4 4 𝑥 0
⇒[ −8 3−λ 4 ] [𝑦] = [0]
−16 8 7−λ 𝑧 0
When λ = −1
−8 4 4 𝑥 0
[ −8 4 4] [𝑦] = [0]
−16 8 8 𝑧 0
Applying 𝑅2 → 𝑅2 − 𝑅1 ; 𝑅3 → 𝑅3 − 2𝑅1
−8 4 4 𝑥 0
𝑦
⇒[ 0 0 0] [ ] = [0]
0 0 0 𝑧 0
Applying 𝑅1 → 𝑅1 /4
−2 1 1 𝑥 0
⇒[ 0 0 0] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are
−2𝑥 + 𝑦 + 𝑧 = 0
Let 𝑧 = 𝑘1 and 𝑦 = 𝑘2
Substituting 𝑦, 𝑧 values in −2𝑥 + 𝑦 + 𝑧 = 0, we get
𝑘1 +𝑘2
⇒−2𝑥 + 𝑘2 + 𝑘1 = 0 ⇒𝑥 = 2
𝑘 +𝑘
1 2
𝑥 2
1/2 1/2 1 1
𝑘 𝑘
Therefore, 𝑋 = [𝑦] = [ 𝑘2 ] = 𝑘1 [ 0 ] + 𝑘2 [ 1 ] = 21 [0] + 22 [2] is the
𝑧 𝑘1 1 0 2 0
eigenvector corresponding to the eigenvalue λ = −1.
Case(ii):
Consider the equation (𝐴 − λI)X = 0
−9 − λ 4 4 𝑥 0
⇒[ −8 3−λ 𝑦
4 ] [ ] = [0 ]
−16 8 7−λ 𝑧 0
When λ = 3
 −12 4 4 𝑥 0
[ −8 0 4] [𝑦] = [0]
−16 8 4 𝑧 0
Applying, 𝑅2 → 𝑅2 − 𝑅1 ; 𝑅3 → 𝑅3 − 𝑅1

−12 4 4 𝑥 0
⇒[ 4 𝑦
−4 0] [ ] = [0]
−4 4 0 𝑧 0
Applying, 𝑅3 → 𝑅3 + 𝑅2
−12 4 4 𝑥 0
⇒[ 4 −4 0] [𝑦] = [0]
0 0 0 𝑧 0
𝑅1
Applying 𝑅1 → ; 𝑅2 → 𝑅2 /4
4

−3 1 1 𝑥 0
⇒[ 1 −1 0] [𝑦 ] = [ 0]
0 0 0 𝑧 0
Applying 𝑅1 ↔ 𝑅2
1 −1 0 𝑥 0
⇒[−3 1 1] [𝑦] = [0]
0 0 0 𝑧 0
Applying, 𝑅2 → 𝑅2 + 3𝑅1
1 −1 0 𝑥 0
⇒[0 𝑦
−2 1] [ ] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥−𝑦 =0
−2𝑦 + 𝑧 = 0
Let 𝑧 = 𝑘 then −2𝑦 + 𝑧 = 0 ⇒−2𝑦 + 𝑘 = 0 ⇒𝑦 = 𝑘/2
Substituting 𝑦 value in 𝑥 − 𝑦 = 0, we get
⇒ 𝑥 − 𝑘/2 = 0 ⇒ 𝑥 = 𝑘/2
𝑥 𝑘/2 1
𝑘
Therefore, 𝑋 = [𝑦] = [𝑘/2] = 2 [1] is the eigenvector corresponding to the
𝑧 𝑘 2
eigenvalue λ = 3.
1 1 1
Hence 𝑋1 = [0] , 𝑋2 = [2] , 𝑋3 = [1] are eigenvectors corresponding to the
2 0 2
eigenvalue λ = −1, −1, 3 respectively.
 For eigenvalue λ = −1, the geometric multiplicity =algebraic multiplicity=2
and for eigenvalue λ = 3, the geometric multiciplicity =algebraic multiplicity =1.
Thus each eigenvalue of A has its geometric multiplicity equal to its algebraic
multiplicity.
Therefore, A is diagonalizable.
1 1 1
Now modal matrix 𝑃 = [𝑋1 𝑋2 𝑋3 ] = [0 2 1]
2 0 2
−1
2 −1 2
then 𝑃 −1 = [ 1 0
−1]
2
−2 1 1
−1
2 −1 −9 4 4 1 1 1
2
Consider 𝑃−1 𝐴𝑃 = [ 1 0
−1] [ −8 3 4] [0 2 1]
2 −16 8 7 2 0 2
−2 1 1
−1 0 0
= [ 0 −1 0] = 𝐷
0 0 3

𝟏 𝟏 𝟑
4. Determine the modal matrix 𝑷 for 𝑨 = [𝟏 𝟓 𝟏] and hence diagonalize 𝑨.
𝟑 𝟏 𝟏
Sol: Characteristic equation of A is |A − λI| = 0
1−λ 1 3
⇒| 1 5−λ 1 |=0
3 1 1−λ
⇒−λ3 + (1 + 5 + 1)λ2 − [5 + 5 + 1 − (1 + 9 + 1)]λ + |𝐴| = 0
⇒−λ3 + 7λ2 − 36 = 0
⇒λ3 − 7λ2 + 36 = 0
⇒ λ = −2,3,6
Therefore the eigenvalues are −2,3,6.
𝑥
Let 𝑋 = [𝑦] be the eigenvector corresponding to the eigenvalue λ
𝑧
Case(i):

Consider the equation (𝐴 − λI)X = 0

1−λ 1 3 𝑥 0
⇒[ 1 5−λ 1 ] [𝑦 ] = [ 0]
3 1 1−λ 𝑧 0
When λ = −2
3 1 3 𝑥 0
[1 7 1] [𝑦] = [0]
3 1 3 𝑧 0
Applying 𝑅1 ↔ 𝑅2
1 7 1 𝑥 0
⇒[3 1 3] [𝑦] = [0]
3 1 3 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
1 7 1 𝑥 0
⇒[3 1 3] [𝑦] = [0]
0 0 0 𝑧 0
Applying, 𝑅2 → 𝑅2 − 3𝑅1
1 7 1 𝑥 0
⇒[0 −20 0] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥 + 7𝑦 + 𝑧 = 0
−20𝑦 = 0 ⇒𝑦 = 0
Let 𝑧 = 𝑘
Substituting 𝑦, 𝑧 values in 𝑥 + 7𝑦 + 𝑧 = 0, we get
⇒ 𝑥 + 𝑘 = 0 ⇒ 𝑥 = −𝑘
𝑥 −𝑘 −1
Therefore, 𝑋 = [𝑦] = [ 0 ] = 𝑘 [ 0 ] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = −2.
Case(ii):

Consider the equation (𝐴 − λI)X = 0

1−λ 1 3 𝑥 0
⇒[ 1 5−λ 1 ] [𝑦 ] = [ 0]
3 1 1−λ 𝑧 0
When λ = 3
−2 1 3 𝑥 0
[ 1 2 1 ] [𝑦] = [0]
3 1 −2 𝑧 0
Applying 𝑅1 ↔ 𝑅2
 1 2 1 𝑥 0
⇒[−2 1 3 ] [𝑦] = [0]
3 1 −2 𝑧 0
Applying 𝑅2 → 𝑅2 + 2𝑅1 ; 𝑅3 → 𝑅3 − 3𝑅1
1 2 1 𝑥 0
⇒[0 5 5 ] [𝑦] = [0]
0 −5 −5 𝑧 0
Applying, 𝑅3 → 𝑅3 + 𝑅2
1 2 1 𝑥 0
⇒[0 𝑦
5 5] [ ] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥 + 2𝑦 + 𝑧 = 0
5𝑦 + 5𝑧 = 0 ⇒𝑦 + 𝑧 = 0
Let 𝑧 = 𝑘 then 𝑦 + 𝑧 = 0⇒ 𝑦 = −𝑘
Substituting 𝑦, 𝑧 values in 𝑥 + 2𝑦 + 𝑧 = 0, we get
⇒ 𝑥 − 2𝑘 + 𝑘 = 0 ⇒ 𝑥 = 𝑘
𝑥 𝑘 1
Therefore, 𝑋 = [𝑦] = [−𝑘] = 𝑘 [−1] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 3.
Case(iii):

Consider the equation (𝐴 − λI)X = 0

1−λ 1 3 𝑥 0
⇒[ 1 5−λ 1 ] [𝑦] = [0]
3 1 1−λ 𝑧 0
When λ = 6
−5 1 3 𝑥 0
[ 1 −1 1 ] [𝑦] = [0]
3 1 −5 𝑧 0
Applying 𝑅1 ↔ 𝑅2
1 −1 1 𝑥 0
⇒[−5 1 3 ] [𝑦] = [0]
3 1 −5 𝑧 0
Applying 𝑅2 → 𝑅2 + 5𝑅1 ; 𝑅3 → 𝑅3 − 3𝑅1
1 −1 1 𝑥 0
⇒[0 −4 8 ] [𝑦] = [0]
0 4 −8 𝑧 0
Applying, 𝑅3 → 𝑅3 + 𝑅2
1 −1 1 𝑥 0
⇒[0 −4 8] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
𝑥−𝑦+𝑧 =0
−4𝑦 + 8𝑧 = 0 ⇒−𝑦 + 2𝑧 = 0
Let 𝑧 = 𝑘 then −𝑦 + 2𝑧 = 0⇒ 𝑦 = 2𝑘
Substituting 𝑦, 𝑧 values in 𝑥 − 𝑦 + 𝑧 = 0, we get
⇒ 𝑥 − 2𝑘 + 𝑘 = 0 ⇒ 𝑥 = 𝑘
𝑥 𝑘 1
Therefore, 𝑋 = [𝑦] = [2𝑘] = 𝑘 [2] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 6.
−1 1 1
Hence 𝑋1 = [ 0 ] , 𝑋2 = [−1] , 𝑋3 = [2] are eigenvectors corresponding to the
1 1 1
eigenvalue λ = −2,3, 6 respectively.
Therefore A is diagonalizable.
−1 1 1
Now the modal matrix 𝑃 = [𝑋1 𝑋2 𝑋3 ] = [ 0 −1 2]
1 1 1
−1 0 1
𝑃 −1 = [ 1 −1 1]
1 1 1
−1 0 1 1 1 3 −1 1 1
Therefore, 𝑃−1 𝐴𝑃 == [ 1 −1 1] [1 5 1] [ 0 −1 2]
1 1 1 3 1 1 1 1 1
−2 0 0
=[ 0 3 0]=D
0 0 6
Cayley-Hamilton theorem: Every square matrix satisfies its own characteristic equation.
Applications:
1. To find higher powers of a square matrix.
2. To find inverse of a square matrix.
Let 𝑨𝟑𝒙𝟑 be a square matrix, and its characteristic equation be
𝑎3 λ3 + 𝑎2 λ2 + 𝑎1 λ + 𝑎0 = 0--------------(1)
By Cayley-Hamilton theorem, A satisfies (1)
⸫ 𝑎3 𝐴3 + 𝑎2 A2 + 𝑎1 A + 𝑎0 𝐼 = 0------------------(2)
i) From (2), we have
A(𝑎3 𝐴3 + 𝑎2 A2 + 𝑎1 A + 𝑎0 𝐼) = 0
⇒ 𝑎3 𝐴4 + 𝑎2 A3 + 𝑎1 A2 + 𝑎0 𝐴 = 0
−1
⇒ 𝐴4 = 𝑎 (𝑎2 A3 + 𝑎1 A2 + 𝑎0 𝐴 )
3
ii) From (2), we have
𝐴−1 (𝑎3 𝐴3 + 𝑎2 A2 + 𝑎1 A + 𝑎0 𝐼) = 0
⇒ 𝑎3 𝐴2 + 𝑎2 𝐴 + 𝑎1 𝐼 + 𝑎0 𝐴−1 = 0
−1
⇒ 𝐴−1 = = ( 𝑎3 𝐴2 + 𝑎2 𝐴 + 𝑎1 𝐼).
𝑎0

Problems:
𝟏 𝟐 −𝟏
1. If 𝑨 = [𝟐 𝟏 −𝟐] verify Cayley-Hamilton theorem. Find 𝑨−𝟏 and 𝑨𝟒 using
𝟐 −𝟐 𝟏
Cayley-Hamilton theorem.

Sol: Characteristic equation of A is |A − λI| = 0

1−λ 2 −1
⇒| 2 1 − λ −2 | = 0
2 −2 1 − λ
⇒(1 − λ)[(1 − λ)(1 − λ) − 4] − 2[2(1 − λ) + 4] − 1[−4 − 2(1 − λ)] = 0
⇒(1 − λ)[λ2 − 2λ − 3] − 2[−2λ + 6] − 1[2λ − 6] = 0
⇒−λ3 + 3λ2 + 3λ − 9 = 0
⇒λ3 − 3λ2 − 3λ + 9 = 0----------------------------------(1)
By Cayley-Hamilton theorem, matrix A should satisfy its characteristic equation.
i.e., A3 − 3A2 − 3A + 9I =O-----------------------------(2)
1 2 −1 1 2 −1 3 6 −6
Now, 𝐴2 = [2 1 −2] [2 1 −2] = [0 9 −6]
2 −2 1 2 −2 1 0 0 3
 3 6 −6 1 2 −1 3 24 −21
and 𝐴3 = 𝐴2 𝐴 = [0 9 −6] [2 1 −2] = [6 21 −24]
0 0 3 2 −2 1 6 −6 3
Consider A3 − 3A2 − 3A + 9I
3 24 −21 3 6 −6 1 2 −1 1 0 0
= [6 21 −24] − 3 [0 9 −6] − 3 [2 1 −2] + 9 [0 1 0]
6 −6 3 0 0 3 2 −2 1 0 0 1
0 0 0
= [0 0 0]=O
0 0 0
Therefore, A3 − 3A2 − 3A + 9I =O
Hence Cayley-Hamilton theorem is verified.
To find 𝐴−1
Multiplying equation (2) with 𝐴−1 on both sides, we get
𝐴−1 [A3 − 3A2 − 3A + 9I] = 𝐴−1(O)
⇒A2 − 3𝐴 − 3I + 9𝐴−1 =O
⇒9𝐴−1 = −A2 + 3𝐴 + 3I
1
⇒𝐴−1 = 9 (−A2 + 3𝐴 + 3I)

3 6 −6 1 2 −1 1 0 0 3 0 3
1 1
= 9 {− [0 9 −6] + 3 [2 1 −2] + 3 [0 1 0]} = 9 [6 −3 0]
0 0 3 2 −2 1 0 0 1 6 −6 3
1 1
0
3 3
−1 2 1
Therefore, 𝐴 = −3 0
3
2 2 1
[3 − 3 3]

To find 𝐴4
Multiplying equation (2) with A on both sides, we get
𝐴[A3 − 3A2 − 3A + 9I] = A(O)
⇒A4 − 3A3 − 3A2 + 9A =O
⇒𝐴4 = 3A3 + 3A2 − 9A
3 24 −21 3 6 −6 1 2 −1 9 72 −72
= 3 [6 21 −24] + 3 [0 9 −6] − 9 [2 1 −2]=[0 81 −72]
6 −6 3 0 0 3 2 −2 1 0 0 9
9 72 −72
4
Therefore, 𝐴 = [0 81 −72]
0 0 9
 𝟕 𝟐 −𝟐
2. Using Cayley-Hamilton theorem find 𝑨−𝟏 and 𝑨𝟒 where 𝑨 = [−𝟔 −𝟏 𝟐 ].
𝟔 𝟐 −𝟏
Sol: Characteristic equation of A is |A − λI| = 0
7−λ 2 −2
⇒ | −6 −1 − λ 2 |=0
6 2 −1 − λ
⇒−λ3 + (7 − 1 − 1)λ2 − [−7 + 1 − 7 − (−12 − 12 + 4)λ + |𝐴| = 0
⇒−λ3 + 5λ2 − 7λ + 3 = 0
⇒λ3 − 5λ2 + 7λ − 3 = 0----------------------------(1)

By Cayley-Hamilton theorem, we have

A3 − 5A2 + 7A − 3I =O-----------------------------(2)
7 2 −2 7 2 −2 25 8 −8
2
We have, 𝐴 = [−6 −1 2 ] [−6 −1 2 ] = [−24 −7 8 ]
6 2 −1 6 2 −1 24 8 −7
25 8 −8 7 2 −2 79 26 −26
and 𝐴3 = 𝐴2 𝐴 = [−24 ] [
−7 8 −6 −1 2 ] = [−78 −25 26 ]
24 8 −7 6 2 −1 78 26 −25
To find 𝐴−1
Multiplying equation (2) with 𝐴−1 on both sides, we get
𝐴−1 [A3 − 5A2 + 7A − 3I] = 𝐴−1(O)
⇒A2 − 5𝐴 + 7I − 3𝐴−1 =O
⇒3𝐴−1 = A2 − 5𝐴 + 7I
1
⇒𝐴−1 = 3 (A2 − 5𝐴 + 7I)

1 25 8 −8 7 2 −2 1 0 0
= {[−24 −7 8 ] − 5 [−6 −1 2 ] + 7 [0 1 0]}
3
24 8 −7 6 2 −1 0 0 1
−3 −2 2
−1 1
Therefore, 𝐴 = 3[ 6 5 −2]
−6 −2 5
To find 𝐴4
Multiplying equation (2) with A on both sides, we get
𝐴[A3 − 5A2 + 7A − 3I] = A(O)
⇒A4 − 5A3 + 7A2 − 3A =O
⇒𝐴4 = 5A3 − 7A2 + 3A
 79 26 −26 25 8 −8 7 2 −2
= 5 [−78 −25 26 ] − 7 [ −24 −7 8 ] + 3 [ −6 −1 2 ]
78 26 −25 24 8 −7 6 2 −1
241 80 −80
Therefore, 𝐴4 = [−240 −79 80 ]
240 80 −79

𝟐 𝟏 𝟏
3. If 𝑨 = [𝟎 𝟏 𝟎], find the value of the matrix 𝑨𝟖 − 𝟓𝑨𝟕 + 𝟕𝑨𝟔 − 𝟑𝑨𝟓 + 𝑨𝟒 − 𝟓𝑨𝟑 +
𝟏 𝟏 𝟐
𝟐
𝟖𝑨 − 𝟐𝑨 + 𝑰.
Sol: Characteristic equation of A is |A − λI| = 0
2−λ 1 1
⇒| 0 1−λ 0 |=0
1 1 2−λ
⇒−λ3 + (2 + 1 + 2)λ2 − [2 + 2 + 4 − (0 + 1 + 0)λ + |𝐴| = 0
⇒−λ3 + 5λ2 − 7λ + 3 = 0
⇒λ3 − 5λ2 + 7λ − 3 = 0

By Cayley-Hamilton theorem, we have

A3 − 5A2 + 7A − 3I =O-----------------------------(1)
Rewriting the given expression as
𝐴8 − 5𝐴7 + 7𝐴6 − 3𝐴5 + 𝐴4 − 5𝐴3 + 8𝐴2 − 2𝐴 + 𝐼
= 𝐴5 (A3 − 5A2 + 7A − 3I) + A(A3 − 5A2 + 8A − 2I) + I
= 𝐴5 (O)+𝐴[(A3 − 5A2 + 7A − 3I) + A + I] + I , using(1)
= O+𝐴(A3 − 5A2 + 7A − 3I) + 𝐴2 + 𝐴 + 𝐼, using(1)
= 𝐴2 + 𝐴 + 𝐼
2 1 1 2 1 1 5 4 4
We have, 𝐴2 = [0 ] [
1 0 0 1 0 ] = [ 0 1 0]
1 1 2 1 1 2 4 4 5
Therefore, value of the matrix 𝐴8 − 5𝐴7 + 7𝐴6 − 3𝐴5 + 𝐴4 − 5𝐴3 + 8𝐴2 − 2𝐴 + 𝐼
= 𝐴2 + 𝐴 + 𝐼
5 4 4 2 1 1 1 0 0
= [0 1 0] + [0 1 0] + [0 1 0]
4 4 5 1 1 2 0 0 1
8 5 5
= [0 3 0]
5 5 8
 𝟑 𝟏
4. If 𝑨 = [ ] write 𝟐𝑨𝟓 − 𝟑𝑨𝟒 + 𝑨𝟐 − 𝟒𝑰 as a linear polynomial in A.
−𝟏 𝟐
Sol: Characteristic equation of A is |A − λI| = 0
3−λ 1
⇒| |=0
−1 2 − λ

⇒λ2 − (𝑡𝑟(𝐴))λ + 𝑑𝑒𝑡𝐴 = 0

⇒λ2 − 5λ + 7 = 0

By Cayley-Hamilton theorem, we have

A2 − 5A + 7𝐼 = 0
⇒A2 = 5A − 7𝐼------------------------(1)
Multiplying on both sides with A, we get
A3 = 5A2 − 7𝐴-------------------------(2)
A4 = 5A3 − 7A2 ------------------------(3)
A5 = 5A4 − 7A3 ------------------------(4)
Given expression is 2𝐴5 − 3𝐴4 + 𝐴2 − 4𝐼
= 2(5A4 − 7A3 ) − 3𝐴4 + 𝐴2 − 4𝐼, using(4)
= 10A4 − 14A3 − 3𝐴4 + 𝐴2 − 4𝐼
= 7A4 − 14A3 + 𝐴2 − 4𝐼
= 7(5A3 − 7A2 ) − 14A3 + 𝐴2 − 4𝐼, using (3)
= 21A3 − 48A2 − 4𝐼
= 21(5A2 − 7𝐴) − 48A2 − 4𝐼, using (2)
= 57A2 − 147𝐴 − 4𝐼
= 57(5A − 7𝐼) − 147𝐴 − 4𝐼, using (1)
= 138𝐴 − 403𝐼 which is a linear polynomial in A

1 0 0
5. If A = [1 0 1], find A50 .
0 1 0
Quadratic forms
An expression of the form 𝑄 = X 𝑇 𝐴𝑋 = ∑𝑛𝑖=1 ∑𝑛𝑗=1 𝑎𝑖𝑗 𝑥𝑖 𝑥𝑗 where 𝑎𝑖𝑗 ′𝑠 are real numbers, is
called a quadratic form in n variables 𝑥1 , 𝑥2 , … , 𝑥𝑛 .
Example:
1. 2𝑥 2 + 7𝑥𝑦 + 5𝑦 2 is a symmetric form in two variables x and y.
2. 6𝑥 2 + 7𝑥𝑦 + 3𝑦 2 + 14𝑧 2 + 4𝑦𝑧 + 18𝑥𝑧 + 4𝑥𝑦 is a quadratic form in three variables
𝑥, 𝑦 and 𝑧.

To write the symmetric matrix corresponding to the QF:

Writing the coefficients of square terms along the diagonal and the coefficients of
product terms divided by 2 at the appropriate places as:
𝑥 𝑦 𝑧

𝑥 Coeff of 𝑥 2 1
Coeff of 𝑥𝑦
1
Coeff of 𝑥𝑧
2 2

𝑦 1
Coeff of 𝑥𝑦 Coeff of 𝑦 2 1
Coeff of 𝑦𝑧
2 2

1 1
𝑧 Coeff of 𝑥𝑧 Coeff of 𝑦𝑧 Coeff of 𝑧 2
2 2

To write the QF when corresponding symmetric matrix is given:

𝑎 ℎ 𝑔 𝑎 ℎ 𝑔 𝑥
If A = [ ℎ 𝑏 𝑓 ] then the QF = 𝑋 𝑇 𝐴𝑋 = [𝑥 𝑦 𝑧] [ℎ 𝑏 𝑓 ] [𝑦]
𝑔 𝑓 𝑐 𝑔 𝑓 𝑐 𝑧

= 𝑎𝑥 2 + 𝑏𝑦 2 + 𝑐𝑧 2 + 2ℎ𝑥𝑦 + 2𝑓𝑦𝑧 + 2𝑔𝑧𝑥.

Problems:
𝟑 𝟐 −𝟏
1. Write QF corresponding to the Symmetric matrix [ 𝟐 𝟐 𝟑]
−𝟏 𝟑 𝟏
Solution:
𝑥 3 2 −1
𝑦
Let X=[ ], A=[ 2 2 3]
𝑧 −1 3 1
3 2 −1 𝑥
QF= X 𝑇 𝐴𝑋 = [𝑥 𝑦 𝑧] [ 2 2 3 ] [𝑦 ]
−1 3 1 𝑧
= 3𝑥 2 + 2𝑦 2 + 𝑧 2 + 4𝑥𝑦 + 6𝑦𝑧 − 2𝑥𝑧
2. Write the symmetric matrix corresponding to the QF 𝟑𝒙𝟐 − 𝟑𝒚𝟐 − 𝟓𝒛𝟐 − 𝟐𝒙𝒚 +
𝟔𝒛𝒙 − 𝟔𝒚𝒛.

Solution:

3 −1 3
Symmetric matrix corresponding to the given QF is A =[ −1 −3 −3]
3 −3 −5

Canonical form or Normal form of a QF:

A real Quadratic form in which the product terms are missing and which contains
only square terms of variables is called canonical form.
i.e. 𝜆1 𝑦1 2 + 𝜆2 𝑦2 2 +. . . +𝜆𝑛 𝑦𝑛 2 is a canonical form.
Let 𝑄 = X 𝑇 𝐴𝑋 be a QF in n variables. A non singular transformation 𝑋 = 𝑃𝑌 which
transforms the QF to another form 𝜆1 𝑦1 2 + 𝜆2 𝑦2 2 +. . . +𝜆𝑛 𝑦𝑛 2 is called a linear
transformation.
If P is an orthogonal matrix then the transformation 𝑋 = 𝑃𝑌 is called an orthogonal
transformation.
Rank, index and signature of a QF:
Rank of QF is rank of the corresponding symmetric matrix and is denoted by r.
No. of positive terms in canonical form is called as index of QF and is denoted by s.
The difference of no. of positive terms and the no. of non-positive terms in canonical
form is called signature of QF.
Signature = 𝑠 − (𝑟 − 𝑠) = 2𝑠 − 𝑟.
 Nature of QF:
Let 𝑄 = X 𝑇 𝐴𝑋 be a QF in n variables and rank of A be 𝑟.
The QF is
1. Positive definite if all the eigenvalues of A are positive i.e if 𝑟 = 𝑛, 𝑠 = 𝑛.
2. Negative definite if all the eigenvalues of A are negative i.e 𝑟 = 𝑛, 𝑠 = 0.
3. Positive semi-definite if all the eigenvalues of A are ≥ 0, and atleast one
eigenvalue is zero i.e if 𝑟 < 𝑛, 𝑠 = 𝑛.
4. Negative semi-definite if all the eigenvalues of A are ≤ 0, and atleast one
eigenvalue is zero i.e if 𝑟 < 𝑛, 𝑠 = 0
5. Indefinite if A has positive as well as negative eigenvalues.

Problems:
3. Identify the nature of the quadratic form:
𝒙𝟏 𝟐 + 𝟒𝒙𝟐 𝟐 + 𝒙𝟑 𝟐 + 𝟒𝒙𝟏 𝒙𝟐 + 𝟐𝒙𝟏 𝒙𝟑 − 𝟒𝒙𝟐 𝒙𝟑 .

Solution:
Given quadratic form is 𝑥1 2 + 4𝑥2 2 + 𝑥3 2 + 4𝑥1 𝑥2 + 2𝑥1 𝑥3 − 4𝑥2 𝑥3 and it is in
three variables 𝑥1 , 𝑥2 and 𝑥3 .

1 −2 1
Symmetric matrix of the given quadratic form is 𝐴 = [−2 4 −2]
1 −2 1
Characteristic equation of 𝐴 is |𝐴 − λI| = 0
1 − λ −2 1
⇒ | −2 4 − λ −2 | = 0
1 −2 1 − λ
⇒−λ + (1 + 4 + 1)λ2 − [4 + 4 + 1 − (4 + 1 + 4)]λ + |𝐴| = 0
3

⇒−λ3 + 6λ2 + 0 = 0
⇒λ3 − 6λ2 = 0 ⇒ λ2 (λ − 6) = 0
⇒ λ2 = 0 or λ − 6 = 0
⇒ λ = 0, 0, 6
Therefore the eigenvalues are λ = 0, 0, 6, which are positive and two values are
zero.
Therefore, the nature of the given quadratic form is positive semi-definite.

4. Discuss the nature of the quadratic form 𝒙𝟐 + 𝟒𝒙𝒚 + 𝟔𝒙𝒛 − 𝒚𝟐 + 𝟐𝒚𝒛 + 𝟒𝒛𝟐 .

Solution:
Given quadratic form is 𝑥1 2 + 4𝑥2 2 + 𝑥3 2 + 4𝑥1 𝑥2 + 2𝑥1 𝑥3 − 4𝑥2 𝑥3 and it is in
three variables 𝑥1 , 𝑥2 and 𝑥3 .

1 2 3
Therefore, the symmetric matrix of the given quadratic form is 𝐴 = [2 −1 1]
3 1 4
Characteristic equation of 𝐴 is |𝐴 − λI| = 0
1−λ 2 3
⇒| 2 −1 − λ 1 |=0
3 1 4−λ
⇒−λ3 + (1 − 1 + 4)λ2 − [−1 − 4 + 4 − (4 + 9 + 1)]λ + |𝐴| = 0
⇒−λ3 + 4λ2 + 13λ + 0 = 0
⇒λ3 − 4λ2 − 13λ = 0 ⇒ λ(𝜆2 − 4λ − 13) = 0
⇒ λ = 0 or 𝜆2 − 4λ − 13 = 0
⇒ λ = 0, 2 + √17, 2 − √17.
Therefore the eigenvalues are λ = 0, 2 + √17, 2 − √17.
Therefore, the nature of the given quadratic form is indefinite.

Reduction of a QF to canonical form(Sum of squares form):

1. Using Diagonalization by linear transformation

2. Orthogonalization or orthogonal transformation.
Using Diagonalization by linear transformation

Problems:

1. Reduce the quadratic form to the canonical form 𝒙𝟐 + 𝒚𝟐 + 𝟐𝒛𝟐 − 𝟐𝒙𝒚 +

𝟒𝒛𝒙 + 𝟒𝒚𝒛 and also find the rank, nature, index and signature of the
quadratic form.

Solution:
Given quadratic form is 𝑥 2 + 𝑦 2 + 2𝑧 2 − 2𝑥𝑦 + 4𝑧𝑥 + 4𝑦𝑧 and it is in three
variables 𝑥, 𝑦 and 𝑧.

1 −1 2
Symmetric matrix of the quadratic form is 𝐴 = [−1 1 2]
2 2 2
We write 𝐴 = 𝐼3 𝐴𝐼3
1 −1 2 1 0 0 1 0 0
i.e. [−1 1 2] = [0 1 0] 𝐴 [0 1 0]
2 2 2 0 0 1 0 0 1
By applying row and column operations we reduce 𝐴 on the L.H.S to the diagonal
form. Same row and column operations will be applied on the pre-factor and post-
factor of 𝐴 on R.H.S respectively to reduce it to the form P 𝑇 𝐴P.
Applying 𝑅2 → 𝑅2 + 𝑅1 ; 𝑅3 → 𝑅3 − 2𝑅1
1 −1 2 1 0 0 1 0 0
⇒[0 0 4 ] = [ 1 1 0 ] 𝐴 [0 1 0]
0 4 −2 −2 0 1 0 0 1
Applying 𝐶2 → 𝐶2 + 𝐶1 ; 𝐶3 → 𝐶3 − 2𝐶1
1 0 0 1 0 0 1 1 −2
⇒[0 0 4 ] = [ 1 1 0] 𝐴 [0 1 0 ]
0 4 −2 −2 0 1 0 0 1
Applying 𝑅2 ↔ 𝑅3
1 0 0 1 0 0 1 1 −2
⇒[0 4 −2] = [−2 0 1] 𝐴 [0 1 0 ]
0 0 4 1 1 0 0 0 1
Applying 𝐶2 ↔ 𝐶3
1 0 0 1 0 0 1 −2 1
⇒[0 −2 4] = [−2 0 1] 𝐴 [0 0 1]
0 4 0 1 1 0 0 1 0
Applying 𝑅3 → 𝑅3 + 2𝑅2
 1 0 0 1 0 0 1 −2 1
⇒[0 −2 4] = [−2 0 1] 𝐴 [0 0 1]
0 0 8 −3 1 2 0 1 0
Applying 𝐶3 → 𝐶3 + 2𝐶2
1 0 0 1 0 0 1 −2 −3
⇒[0 −2 0] = [−2 0 1] 𝐴 [ 0 0 1]
0 0 8 −3 1 2 0 1 2
Thus this is of the form 𝐷 = P 𝑇 𝐴P
1 0 0 1 −2 −3
where 𝐷 = [0 −2 0]is the diagonal matrix and P = [0 0 1]
0 0 8 0 1 2
1 0 0 𝑦1
𝑇 [
The canonical form is 𝑌 𝐷𝑌 = 1𝑦 𝑦2 𝑦3 [0 −2 0] [𝑦2 ]
]
0 0 8 𝑦3
= 𝑦1 2 − 2𝑦2 2 + 8𝑦3 2
This is done by using the linear transformation X = PY
𝑥 1 −2 −3 𝑦1
where X = [𝑦], P = [0 0 1 ] and Y = [𝑦2 ].
𝑧 0 1 2 𝑦3
Number of variables in the quadratic form 𝑛 = 3.
Rank of the given quadratic form
𝑟 = number of non-zero terms in its canonical form = 3.
The index of the given quadratic form
𝑠 =number of positive terms in its canonical form=2.
The nature of the given quadratic form is indefinite (since 𝑟 = 𝑛, 𝑠 < 𝑛).
The signature of the given quadratic form = difference between the number of positive
terms and number of negative terms in its canonical form = 2-1=1
(or) 2𝑠 − 𝑟 = 4 − 3 = 1.
2. Reduce the quadratic form to canonical form
𝟐𝒙𝟏 𝟐 + 𝒙𝟐 𝟐 − 𝟑𝒙𝟑 𝟐 + 𝟏𝟐𝒙𝟏 𝒙𝟐 − 𝟒𝒙𝟏 𝒙𝟑 − 𝟖𝒙𝟐 𝒙𝟑 into sum of squares form.

Solution:
Given quadratic form is 2𝑥1 2 + 𝑥2 2 − 3𝑥3 2 + 12𝑥1 𝑥2 − 4𝑥1 𝑥3 − 8𝑥2 𝑥3 and it is in
three variables 𝑥1 , 𝑥2 and 𝑥3 .
2 6 −2
Symmetric matrix of the given quadratic form is 𝐴 = [ 6 1 −4]
−2 −4 −3
We write 𝐴 = 𝐼3 𝐴𝐼3
 2 6 −2 1 0 0 1 0 0
i.e. [ 6 1 −4] = [0 1 0] 𝐴 [0 1 0]
−2 −4 −3 0 0 1 0 0 1
By applying row and column operations we reduce 𝐴 on the L.H.S to the diagonal
form. Same row and column operations will be applied on the pre-factor and post-
factor of 𝐴 on R.H.S respectively to reduce it to the form P 𝑇 𝐴P.
Applying 𝑅2 → 𝑅2 − 3𝑅1 ; 𝑅3 → 𝑅3 + 𝑅1
2 6 −2 1 0 0 1 0 0
⇒[0 −17 2 ] = [−3 1 0] 𝐴 [0 1 0]
0 2 −5 1 0 1 0 0 1
Applying 𝐶2 → 𝐶2 − 3𝐶1 ; 𝐶3 → 𝐶3 + 𝐶1
2 0 0 1 0 0 1 −3 1
⇒[0 −17 2 ] = [−3 1 0] 𝐴 [0 1 0]
0 2 −5 1 0 1 0 0 1
2
Applying 𝑅3 → 𝑅3 + 17 𝑅2
2 0 0 1 0 0 1 −3 1
⇒[0 −17 2 ] = [−3 1 0] 𝐴 [0 1 0]
81 11 2
0 0 − 17 1 0 0 1
17 17
2
Applying 𝑅3 → 𝑅3 + 17 𝑅2
11
2 0 0 1 0 0 1 −3 17
⇒[0 −17 0 ] = [−3
81 11
1
2
0] 𝐴 [
0 1
2]
0 0 − 17 1 17
17 17 0 0 1
This is of the form 𝐷 = P 𝑇 𝐴P
11
2 0 0 1 −3
17
where 𝐷 = [0 −17 0 ]is the diagonal matrix and P = [
81 0 1
2]
0 0 − 17
17 0 0 1
2 0 0 𝑦1
The canonical form is 𝑌 𝐷𝑌 = [𝑦1
𝑇 𝑦2 𝑦3 ] [ 0 −17 0 ] [𝑦2 ]
81
0 0 − 17 𝑦3

81 2
= 2𝑦1 2 − 17𝑦2 2 − 𝑦
17 3
This is done by using the linear transformation X = PY
11
𝑥 1 −3 𝑦1
17
where X = [𝑦], P = [0 1
2] and Y = [𝑦2 ].
𝑧 17 𝑦3
0 0 1
3. Reduce the QF 𝒙𝟐 + 𝟐𝒚𝟐 + 𝟐𝒛𝟐 − 𝟐𝒙𝒚 − 𝟐𝒚𝒛 + 𝒛𝒙 to canonical form and
find its rank, index and signature.
Solution:
Symmetric matrix of the given quadratic form is
1 −1 1/2
A =[ −1 2 −1 ]
1/2 −1 2
We write 𝐴 = 𝐼 𝐴 𝐼
1 −1 1/2 1 0 0 1 0 0
[ −1 2 −1 ] = [0 1 0] 𝐴 [0 1 0]
1/2 −1 2 0 0 1 0 0 1
1
Applying 𝑅2 → 𝑅2 + 𝑅1 ; 𝑅3 → 𝑅3 − 2 𝑅1

1 −1 1/2 1 0 0 1 0 0
[0 1 −1/2] = [ 1 1 0] 𝐴 [0 1 0]
0 −1/2 7/4 −1/2 0 1 0 0 1
1
Applying 𝐶2 → 𝐶2 + 𝐶1 ; 𝐶3 → 𝐶3 − 2 𝐶1

1 0 0 1 0 0 1 1 −1/2
[0 1 −1/2] = [ 1 1 0] 𝐴 [0 1 0 ]
0 −1/2 7/4 −1/2 0 1 0 0 1
1
Applying 𝑅3 → 𝑅3 + 2 𝑅2

1 0 0 1 0 0 1 1 −1/2
[0 1 −1/2] = [1 1 0] 𝐴 [0 1 0 ]
0 0 3/2 0 1/2 1 0 0 1
1
Applying 𝐶3 → 𝐶3 + 2 𝐶2

1 0 0 1 0 0 1 1 0
[ 0 1 0 ] = [1 1 0] 𝐴 [0 1 1/2]
0 0 3/2 0 1/2 1 0 0 1
This is of the form 𝐷 = P 𝑇 𝐴P
1 0 0 1 1 0
where 𝐷 = [0 1 0 ]is the diagonal matrix and P = [0 1 1/2]
0 0 3/2 0 0 1
1 0 0 𝑦1
The canonical form is 𝑌 𝐷𝑌 = [𝑦1
𝑇 𝑦2 𝑦3 ] [0 1 0 ] [𝑦2 ]
0 0 3/2 𝑦3
3
= 𝑦1 2 + 𝑦2 2 + 𝑦3 2
2
This is done by using the linear transformation X = PY
𝑥 1 1 0 𝑦1
where X = [𝑦], P = [0 1 1/2] and Y = [𝑦2 ].
𝑧 0 0 1 𝑦3
Rank of the QF = r= 3
Index = s = 3
Signature = 2s – r = 3
Given QF is positive definite.

4. Reduce the QF 𝒙𝟐 − 𝟐𝒚𝟐 + 𝟑𝒛𝟐 − 𝟒𝒚𝒛 + 𝟔𝒛𝒙 to canonical form and find its
rank, index and signature.

Solution:
Symmetric matrix of the given quadratic form is
1 0 3
A =[ 0 −2 −2]
3 −2 3
We write 𝐴 = 𝐼 𝐴 𝐼
1 0 3 1 0 0 1 0 0
[ 0 −2 −2] = [0 1 0] 𝐴 [0 1 0]
3 −2 3 0 0 1 0 0 1
Applying 𝑅3 → 𝑅3 − 3𝑅1
1 0 3 1 0 0 1 0 0
[ 0 −2 −2 ] = [ 0 1 0] 𝐴 [ 0 1 0]
0 −2 −6 −3 0 1 0 0 1
Applying 𝐶3 → 𝐶3 − 3𝐶1
1 0 0 1 0 0 1 0 −3
[ 0 −2 −2 ] = [ 0 1 0] 𝐴 [0 1 0]
0 −2 −6 −3 0 1 0 0 1

Applying 𝑅3 → 𝑅3 − 𝑅2
1 0 0 1 0 0 1 0 −3
[ 0 −2 −2 ] = [ 0 1 0] 𝐴 [0 1 0]
0 0 −4 −3 −1 1 0 0 1
Applying 𝐶3 → 𝐶3 − 𝐶2

1 0 0 1 0 0 1 0 −3
[ 0 −2 0 ]=[ 0 1 0] 𝐴 [0 1 −1]
0 0 −4 −3 −1 1 0 0 1
This is of the form 𝐷 = P 𝑇 𝐴P
1 0 0 1 0 −3
where 𝐷 = [ 0 −2 0 ]is the diagonal matrix and P = [0 1 −1]
0 0 −4 0 0 1
1 0 0 𝑦1
𝑇
The canonical form is 𝑌 𝐷𝑌 = [ 1𝑦 𝑦2 𝑦 3] [ 0 −2 0 ] [𝑦2 ]
0 0 −4 𝑦3
= 𝑦1 2 − 2𝑦2 2 − 4𝑦3 2
This is done by using the linear transformation X = PY
𝑥 1 0 −3 𝑦1
where X = [𝑦], P = [0 1 −1] and Y = [𝑦2 ].
𝑧 0 0 1 𝑦3
Rank of the QF = r= 3
Index = s = 1
Signature = 2s – r = -1
Given QF is indefinite.
 Orthogonal transformation:

1. Reduce the quadratic form 𝟑𝒙𝟐 + 𝟐𝒚𝟐 + 𝟑𝒛𝟐 − 𝟐𝒙𝒚 − 𝟐𝒚𝒛 to canonical form by
orthogonal transformation and also find the nature, index and signature of the
quadratic form.

Sol: Given quadratic form is 3𝑥 2 + 2𝑦 2 + 3𝑧 2 − 2𝑥𝑦 − 2𝑦𝑧 and it is in three

variables 𝑥, 𝑦 and 𝑧.

3 −1 0
the matrix of the quadratic form is 𝐴 = [−1 2 −1]
0 −1 3
Characteristic equation of 𝐴 is |𝐴 − λI| = 0
3 − λ −1 0
⇒ | −1 2 − λ −1 | = 0
0 −1 3 − λ
⇒−λ + (3 + 2 + 3)λ2 − [6 + 6 + 9 − (1 + 0 + 1)]λ + |𝐴| = 0
3

⇒−λ3 + 8λ2 − 19λ + 12 = 0

⇒λ3 − 8λ2 + 19λ − 12 = 0
⇒ λ = 3, 1, 4
Therefore the eigenvalues are 3, 1, 4.
𝑥
𝑦
Let 𝑋 = [ ] be the eigenvector corresponding to the eigenvalue λ
𝑧
Consider the equation (𝐴 − λI)X = 0
3−λ −1 0 𝑥 0
⇒[ −1 2−λ −1 ] [𝑦] = [0]
0 −1 3−λ 𝑧 0
Case(i): When λ = 3

0 −1 0 𝑥 0
(𝐴 − λI)X = 0 ⇒ [−1 −1 −1] [𝑦] = [0]
0 −1 0 𝑧 0
Applying 𝑅1 ↔ 𝑅2
−1 −1 −1 𝑥 0
⇒[ 0 −1 0 ] [𝑦] = [0]
0 −1 0 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
−1 −1 −1 𝑥 0
⇒[ 0 −1 0 ] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
−𝑥 − 𝑦 − 𝑧 = 0
−𝑦 = 0
Let 𝑧 = 𝑘
Substituting 𝑦, 𝑧 values in −𝑥 − 𝑦 − 𝑧 = 0, we get
⇒−𝑥 − 𝑘 = 0 ⇒ 𝑥 = −𝑘
𝑥 −𝑘 −1
Therefore, 𝑋 = [𝑦] = [ 0 ] = 𝑘 [ 0 ] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 3.
Case(ii): When λ = 1

2 −1 0 𝑥 0
(𝐴 − λI)X = 0 ⇒ [−1 1 −1] [𝑦] = [0]
0 −1 2 𝑧 0
Applying 𝑅1 ↔ 𝑅2
−1 1 −1 𝑥 0
⇒[ 2 −1 0 ] [𝑦] = [0]
0 −1 2 𝑧 0
Applying 𝑅2 → 𝑅2 + 2𝑅1
−1 1 −1 𝑥 0
⇒[ 0 1 −2] [ 𝑦 ] = [ 0]
0 −1 2 𝑧 0
Applying 𝑅3 → 𝑅3 + 𝑅2
−1 1 −1 𝑥 0
⇒[ 0 1 −2] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
−𝑥 + 𝑦 − 𝑧 = 0
𝑦 − 2𝑧 = 0
Let 𝑧 = 𝑘 then 𝑦 − 2𝑧 = 0 ⇒ 𝑦 = 2𝑘
Substituting 𝑦, 𝑧 values in −𝑥 + 𝑦 − 𝑧 = 0, we get
⇒−𝑥 + 2𝑘 − 𝑘 = 0 ⇒ 𝑥 = 𝑘
𝑥 𝑘 1
𝑦
Therefore, 𝑋 = [ ] = [2𝑘] = 𝑘 [2] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 1.
Case(iii): When λ = 4

3 − λ −1 0 𝑥 0
(𝐴 − λI)X = 0 ⇒[ −1 2 − λ −1 ] [𝑦] = [0]
0 −1 3−λ 𝑧 0
−1 −1 0 𝑥 0
⇒ [−1 −2 −1] [𝑦] = [0]
0 −1 −1 𝑧 0
Applying 𝑅2 → 𝑅2 − 𝑅1
−1 −1 0 𝑥 0
⇒[ 0 −1 −1] [ 𝑦 ] = [ 0]
0 −1 −1 𝑧 0
Applying 𝑅3 → 𝑅3 − 𝑅2
−1 −1 0 𝑥 0
⇒[ 0 −1 −1] [𝑦] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
−𝑥 − 𝑦 = 0
−𝑦 − 𝑧 = 0
Let 𝑧 = 𝑘 then −𝑦 − 𝑧 = 0 ⇒ 𝑦 = −𝑘
Substituting 𝑦 value in −𝑥 − 𝑦 = 0, we get
⇒−𝑥 + 𝑘 = 0 ⇒ 𝑥 = 𝑘
𝑥 𝑘 1
𝑦
Therefore, 𝑋 = [ ] = [−𝑘] = 𝑘 [−1] is the eigenvector corresponding to the
𝑧 𝑘 1
eigenvalue λ = 4.
−1 1 1
Hence 𝑋1 = [ 0 ] , 𝑋2 = [2] , 𝑋3 = [−1] are eigenvectors corresponding to the
1 1 1
eigenvalue λ = 3, 1, 4 respectively.
Since 𝑋1 , 𝑋2 , 𝑋3 are pairwise orthogonal,

‖𝑋1 ‖ = √(−1)2 + 02 + 12 = √2

‖𝑋2 ‖ = √12 + 22 + 12 = √6

‖𝑋3 ‖ = √12 + (−1)2 + 12 = √3

 Normalizing 𝑃, we get
−1 1 1 −1 1
0
√2 √6 √3 √2 √2
𝑋1 𝑋2 𝑋3 2 −1 1 2 1
𝑃 = [‖𝑋 ] = 0 and 𝑃𝑇 =
1‖ ‖𝑋2 ‖ ‖𝑋3 ‖ √6 √3 √6 √6 √6
1 1 1 1 −1 1
[ √2 √6 √3] [√3 √3 √3]

Since P is orthogonal, 𝑃−1 = 𝑃𝑇

−1 1 −1 1 1
0
√2 √2 3 −1 0 √2 √6 √3
−1 𝑇 1 2 1 2 −1
Therefore, 𝑃 𝐴𝑃 = 𝑃 𝐴𝑃 = √6 √6 √6
[−1 2 −1] 0 √6 √3
1 −1 1 0 −1 3 1 1 1
[√3 √3 √3] [√2 √6 √3 ]

3 0 0
= [0 1 0]=D
0 0 4
3 0 0 𝑦1 𝑦1
The canonical form is 𝑌 𝐷𝑌 = [𝑦1 𝑇 𝑦2 𝑦3 ] [0 1 0] [𝑦2 ] where 𝑌 = [𝑦2 ]
0 0 4 𝑦3 𝑦3

= 3𝑦1 2 + 𝑦2 2 + 4𝑦3 2
Rank of the given quadratic form 𝑟 = number of non-zero terms in its canonical form

So r = 3.

The index of the given quadratic form 𝑠 =number of positive terms in its canonical
form=3.

The nature of the given quadratic form is positive definite (Since all the eigenvalues are
positive).

The signature of the given quadratic form= difference between the number of positive
terms and number of negative terms in its canonical form= 3-0=3 (or) 2𝑠 − 𝑟 = 6 −
3 = 3.
2. Reduce the quadratic form 𝟐𝒙𝟏 𝟐 + 𝟐𝒙𝟐 𝟐 + 𝟐𝒙𝟑 𝟐 − 𝟐𝒙𝟏 𝒙𝟐 − 𝟐𝒙𝟐 𝒙𝟑 − 𝟐𝒙𝟑 𝒙𝟏 into sum
of squares form by orthogonal transformation.

Sol: Given quadratic form is 2𝑥1 2 + 2𝑥2 2 + 2𝑥3 2 − 2𝑥1 𝑥2 − 2𝑥2 𝑥3 − 2𝑥3 𝑥1 and it is in
three variables 𝑥1 , 𝑥2 and 𝑥3 .
2 −1 −1
Therefore, the matrix of the given quadratic form is 𝐴 = [−1 2 −1]
−1 −1 2
Characteristic equation of 𝐴 is |𝐴 − λI| = 0
2−λ −1 −1
⇒ | −1 2 − λ −1 | = 0
−1 −1 2 − λ
⇒−λ3 + (2 + 2 + 2)λ2 − [4 + 4 + 4 − (1 + 1 + 1)]λ + |𝐴| = 0
⇒−λ3 + 6λ2 − 9λ + 0 = 0
⇒λ3 − 6λ2 + 9λ = 0 ⇒ λ(𝜆2 − 6λ + 9) = 0
⇒ λ = 0 or 𝜆2 − 6λ + 9 = 0
⇒ λ = 0, 3,3
Therefore the eigenvalues are λ = 0, 3, 3.
𝑥
Let 𝑋 = [𝑦] be the eigenvector corresponding to the eigenvalue λ
𝑧
Case(i):

2−λ −1 −1 𝑥 0
Consider the equation (𝐴 − λI)X = 0 ⇒[ −1 2−λ 𝑦
−1 ] [ ] = [0]
−1 −1 2−λ 𝑧 0
When λ = 0
2 −1 −1 𝑥 0
[−1 2 −1] [𝑦] = [0]
−1 −1 2 𝑧 0

Applying 𝑅1 ↔ 𝑅2
−1 2 −1 𝑥 0
⇒[ 2 𝑦
−1 −1] [ ] = [0]
−1 −1 2 𝑧 0
Applying 𝑅2 → 𝑅2 + 2𝑅1 ; 𝑅3 → 𝑅3 − 𝑅1
−1 2 −1 𝑥 0
⇒[ 0 3 −3] [𝑦] = [0]
0 −3 3 𝑧 0
Applying 𝑅3 → 𝑅3 + 𝑅2
−1 2 −1 𝑥 0
⇒[ 0 3 −3] [𝑦] = [0]
0 0 0 𝑧 0

By Backward substitution, the equations for the above are:

−𝑥 + 2𝑦 − 𝑧 = 0
3𝑦 − 3𝑧 = 0
Let 𝑧 = 𝑘 then 3𝑦 − 3𝑧 = 0 ⇒𝑦 = 𝑘
Substituting 𝑦, 𝑧 values in −𝑥 + 2𝑦 − 𝑧 = 0, we get
⇒−𝑥 + 2𝑘 − 𝑘 = 0 ⇒ 𝑥 = 𝑘
𝑥 𝑘 1
Therefore, 𝑋 = [𝑦] = [𝑘] = 𝑘 [1] is the eigenvector corresponding to the eigenvalue
𝑧 𝑘 1
λ = 0.
Case(ii):

Consider the equation (𝐴 − λI)X = 0

2−λ −1 −1 𝑥 0
[ −1 2 − λ −1 ] [𝑦] = [0]
−1 −1 2 − λ 𝑧 0
When λ = 3
−1 −1 −1 𝑥 0
[−1 −1 −1] [𝑦] = [0]
−1 −1 −1 𝑧 0
Applying 𝑅2 → 𝑅2 − 2𝑅1 ; 𝑅3 → 𝑅3 − 3𝑅1
−1 −1 −1 𝑥 0
⇒[ 0 0 𝑦
0 ] [ ] = [0]
0 0 0 𝑧 0
By Backward substitution, the equations for the above are:
−𝑥 − 𝑦 − 𝑧 = 0
Let 𝑧 = 𝑘1 and 𝑦 = 𝑘2
Substituting 𝑦, 𝑧 values in −𝑥 − 𝑦 − 𝑧 = 0, we get
⇒−𝑥 − 𝑘2 − 𝑘1 = 0 ⇒𝑥 = −𝑘2 − 𝑘1
 𝑥 −𝑘1 − 𝑘2 −1 −1
Therefore, 𝑋 = [𝑦] = [ 𝑘2 ] = 𝑘1 [ 0 ] + 𝑘2 [ 1 ] is the eigenvector
𝑧 𝑘1 1 0
corresponding to the eigenvalue λ = 3.
1 −1 −1
Hence 𝑋1 = [1] , 𝑋2 = [ 0 ] , 𝑋3 = [ 1 ] are eigenvectors corresponding to the
1 1 0
eigenvalue λ = 0, 3, 3 respectively.
Since 𝑋2 , 𝑋3 are not pairwise orthogonal
𝑥
Let 𝑋3 = [𝑦] be the new eigenvector which is orthogonal to 𝑋1 and 𝑋2
𝑧
then 𝑋1 𝑋3 𝑇 = 𝑥 + 𝑦 + 𝑧 = 0------------------(1)

and 𝑋2 𝑋3 𝑇 = −𝑥 + 𝑧 = 0-------------(2)
Let 𝑧 = 𝑘 then −𝑥 + 𝑧 = 0⇒𝑥 = 𝑘
Substituting 𝑥, 𝑧 values in 𝑥 + 𝑦 + 𝑧 = 0, we get
𝑘 + 𝑦 + 𝑘 = 0 ⇒ 𝑦 = −2𝑘
𝑘 1
Therefore the new eigenvector 𝑋3 = [−2𝑘] = 𝑘 [−2]
𝑘 1
1 −1 1
Thus 𝑋1 = [1] , 𝑋2 = [ 0 ] , 𝑋3 = [−2] are mutually orthogonal.
1 1 1
then

‖𝑋1 ‖ = √12 + 12 + 12 = √3

‖𝑋2 ‖ = √(−1)2 + 02 + 12 = √2

‖𝑋3 ‖ = √12 + (−2)2 + 12 = √6

Normalizing 𝑃, we get
1 −1 1 1 1 1
√3 √2 √6 √3 √3 √3
𝑋1 𝑋2 𝑋3 1 −2 −1 1
𝑃 = [‖𝑋 ] = 0 and 𝑃𝑇 = 0
1‖ ‖𝑋2 ‖ ‖𝑋3 ‖ √3 √6 √2 √2
1 1 1 1 −2 1
[√3 √2 √6] [√6 √6 √6]

then 𝑃 −1 = 𝑃𝑇
 1 1 1 1 −1 1
√3 √3 √3 2 −1 −1 √3 √2 √6
−1 𝑇 −1 1 1 −2
Therefore, 𝑃 𝐴𝑃 = 𝑃 𝐴𝑃 = 0 [−1 2 −1] 0
√2 √2 √3 √6
1 −2 1 −1 −1 2 1 1 1
[√6 √6 √6] [√3 √2 √6 ]

0 0 0
= [0 3 0]=D
0 0 3
0 0 0 𝑦1 𝑦1
The canonical form is 𝑌 𝐷𝑌 = [𝑦1
𝑇 𝑦2 𝑦3 ] [0 3 0] [𝑦2 ] where 𝑌 = [𝑦2 ]
0 0 3 𝑦3 𝑦3
= 3𝑦2 2 + 3𝑦3 2
This is the sum of squares form of the given quadratic form.
Mathematics-I Mathematics Faculty 1

Subject: Mathematics-I
Unit-III
Ordinary Differential Equations

1 Introduction
Ordinary Differential Equation: An equation involving dependent vari-
able(y) and its derivative with respect to one independent variable(x) is called
Ordinary Differential Equation(ODE).
Examples:

dy
1. dx = −ky
2
d y
2. m dx 2 = −ky

Partial Differential Equation: An equation involving dependent variable

and its derivative with respect to two or more independent variables is called
Partial differential equation(PDE).
Examples:

∂2w ∂2w ∂2w

1. ∂x2 + ∂y 2 + ∂z 2 = 0 (Laplace’ s equation)
 
∂2w ∂2w 2
2. a2 ∂x2 + ∂y 2 + ∂∂zw2 = ∂w
∂t (Heat equation)
 
∂2w ∂2w ∂2w ∂2w
3. a2 ∂x2 + ∂y 2 + ∂z 2 = ∂t2 (Wave equation)

Order: The order of a differential equation is the order of the highest derivative
present in the equation.
Examples:

dy
1. dx = −ky order=1
d2 y dy
2. dx2 − 5 dx + 6y = 0 order=2
3
3. x2 y 00 − (xy 0 ) + 6y = log x order=2
h i1
2 2
4. 1 + (y 0 ) = 5y order=1

Degree: The degree of a differential equation is the degree(or power) of the

highest order derivative occurring in the equation, after the equation has been
made free of radicals and factions in its derivatives.
Examples:

1
Mathematics-I Mathematics Faculty 2

3
1. x2 y 00 − (xy 0 ) + 6y = log x order=2 & degree=1
h i1
2 2
2. 1 + (y 0 ) = 5y order=1 & degree=2

3. (y 00 )2 = 1 + (y 0 )3 order=2 & degree=2

dy
4. y + sin( dx )=0 order=1 & degree=Not defined

Theorem: The necessary and sufficient condition for the differential equation
∂M ∂N
M (x, y)dx + N (x, y)dy = 0 to be exact is =
∂y ∂x
Example 1.
(2x − y + 1)dx + (2y − x − 1)dy = 0 (1)
is Exact or not?
Solution:

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = 2x − y + 1
and N (x, y) = 2y − x − 1

∂M ∂N
= −1 and = −1
∂y ∂x

∂M ∂N
∴ =
∂y ∂x

∴ (1) is an exact Differential equation

Example 2.
(ey + 1) cos xdx + ey sin xdy = 0 (1)
is Exact or not?
Solution:

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = (ey + 1) cos x
and N (x, y) = ey sin x

∂M ∂N
= ey cos x and = ey cos x
∂y ∂x

2
Mathematics-I Mathematics Faculty 3

∂M ∂N
∴ =
∂y ∂x

∴ (1) is an exact Differential equation

Example 3. Solve

(y 2 − 2xy)dx − (x2 − 2xy)dy = 0 (1)

Solution:

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = y 2 − 2xy
and N (x, y) = −x2 + 2xy

∂M ∂N
= 2y − 2x and = −2x + 2y
∂y ∂x

∂M ∂N
∴ =
∂y ∂x

∴ (1) is an exact Differential equation

2 Solution of Exact differential equation

Working rule to solve an Exact Differential Equation:
Let the given Differential equation be of the form M (x, y)dx + N (x, y)dy = 0
Step-I: check the condition for exactness
∂M ∂N
= if it is exact then proceed to step II
∂y ∂x
Step-II:
The general solution is
Z Z
M (x, y)dx + N (x, y)dy = C
(y constant) (Only those terms which do not contain x)

Example 1.

(2x − y + 1) dx + (2y − x − 1) dy = 0 (1)

is Exact or not? If it Exact then find the solution.

Solution:

3
Mathematics-I Mathematics Faculty 4

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = 2x − y + 1
and N (x, y) = 2y − x − 1

∂M ∂N
= −1 and = −1
∂y ∂x

∂M ∂N
∴ =
∂y ∂x
∴ (1) is an exact Differential equation
The general solution is
Z Z
M (x, y) dx + N (x, y) dy = C
(y constant) (Only those terms which do not contain x)

Z Z
(2x − y + 1) dx + (2y − 1) dy = C

Z Z Z Z Z
2 x dx − y dx + dx + 2 y dy − dy = C

∴ The general solution is

x2 − yx + x + y 2 − y = c (3)

Example 2.
y sin 2x dx − (y 2 + cos2 x) dy = 0 (1)
is Exact or not? If it Exact then find the solution.

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = y sin 2x
and N (x, y) = −y 2 − cos2 x

∂M ∂N
= sin 2x and = −2 cos x.(− sin x) = 2 sin x cos x = sin 2x
∂y ∂x

∂M ∂N
∴ =
∂y ∂x

4
Mathematics-I Mathematics Faculty 5

∴ (1) is an exact Differential equation

The general solution is
Z Z
M (x, y) dx + N (x, y) dy = C
(y constant) (Only those terms which do not contain x)

Z Z
y sin 2x dx + −y 2 dy = C

y cos 2x y 3
− − =C
2 3
∴ The general solution is
y cos 2x y 3
+ =C (3)
2 3

Example 3.
(ey + 1) cos x dx + ey sin x dy = 0 (1)
is Exact or not? If it Exact then find the solution.

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = (ey + 1) cos x
and N (x, y) = ey sin x

∂M ∂N
= ey cos x and = ey cos x
∂y ∂x

∂M ∂N
∴ =
∂y ∂x
∴ (1) is an exact Differential equation
The general solution is
Z Z
M (x, y) dx + N (x, y) dy = C
(y constant) (Only those terms which do not contain x)

Z Z
(ey + 1) cos x dx + 0 dy = C

Z
y
(e + 1) cos x dx + 0 = C

5
Mathematics-I Mathematics Faculty 6

∴ The general solution is

(ey + 1) sin x = c (3)

Example 4.
(y 2 − 2xy) dx − (x2 − 2xy) dy = 0 (1)
is Exact or not? If it Exact then find the solution.

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = y 2 − 2xy
and N (x, y) = −x2 + 2xy

∂M ∂N
= 2y − 2x and = −2x + 2y
∂y ∂x

∂M ∂N
∴ =
∂y ∂x
∴ (1) is an exact Differential equation
The general solution is
Z Z
M (x, y) dx + N (x, y) dy = C
(y constant) (Only those terms which do not contain x)

Z Z
(y 2 − 2xy) dx + 0 dy = C

Z Z
2
y dx − 2y x dx + 0 = C

∴ The general solution is

y 2 x − yx2 = c (3)

Example 5.
2 2
(y 2 exy + 4x3 ) dx + (2xy.exy − 3y 2 ) dy = 0 (1)

is Exact or not? If it Exact then find the solution.

M (x, y)dx + N (x, y)dy = 0 (2)

6
Mathematics-I Mathematics Faculty 7

Compare equation (1) with (2)

2
where M (x, y) = y 2 exy + 4x3
2
and N (x, y) = 2xy.exy − 3y 2

∂M 2 2 ∂N 2 2
= 2yexy + y 2 exy (2xy) and = 2yexy + 2xy.exy (y 2 )
∂y ∂x

∂M ∂N
∴ =
∂y ∂x
∴ (1) is an exact Differential equation
The general solution is
Z Z
M (x, y) dx + N (x, y) dy = C
(y constant) (Only those terms which do not contain x)

Z Z
2 xy 2 3
(y e + 4x ) dx + (−3y 2 ) dy = C

Z Z
2 xy 2
y e dx + 4x3 dx − y 3 = C

∴ The general solution is

2
exy
y2 + x4 − y 3 = c (3)
y2

Example 6.
1
y(1 + ) + cos y dx + (x + log x − x sin y) dy = 0 (1)
x
is Exact or not? If it Exact then find the solution.

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = y(1 + x1 ) + cos y
and N (x, y) = (x + log x − x sin y)

∂M 1 ∂N 1
= (1 + ) − sin y and = 1 + − sin y
∂y x ∂x x

∂M ∂N
∴ =
∂y ∂x

7
Mathematics-I Mathematics Faculty 8

∴ (1) is an exact Differential equation

The general solution is
Z Z
M (x, y) dx + N (x, y) dy = C
(y constant) (Only those terms which do not contain x)

Z   Z
1
(1 + )y + cos y dx + 0 dy = C
x

Z
(x + log x)y + cos y dx = C

∴ The general solution is

(x + log x)y + x cos y = c (3)

Example 7.
x2 y dx − x3 + y 3


dy = 0 (1)
is Exact or not? If it Exact then find the solution.

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = x2 y

and N (x, y) = − x3 + y 3

∂M ∂N
= x2 and = −3x2
∂y ∂x

∂M ∂N
∴ 6=
∂y ∂x

∴ (1) is not an exact Differential equation

3 Equations Reducible to Exact

3.1 Method-I(Homogeneous)
The differential equation M dx + N dy = 0 is not exact differential equation. If
M dx + N dy = 0 is a homogeneous Differential equation and M x + N y 6= 0 then
1
is an Integrating Factor(I.F).
Mx + Ny

8
Mathematics-I Mathematics Faculty 9

Homogeneous function: A function f (x, y) is said to be homogeneous func-

tion of degree n, if f (kx, ky) = k n f (x, y).
Example 1. Solve
x2 y dx − x3 + y 3


dy = 0 (1)

Solution:

M (x, y)dx + N (x, y)dy = 0 (2)

Compare equation (1) with (2)
where M (x, y) = x2 y

and N (x, y) = − x3 + y 3

∂M ∂N
= x2 and = −3x2
∂y ∂x

∂M ∂N
∴ 6=
∂y ∂x
∴ (1) is not an exact Differential equation
But (1) is homogeneous Differential equation
M x + N y = x3 y − x3 y − y 4 = −y 4 6= 0
1 1
Integrating Factor (I.F)= =− 4
Mx + Ny y
Multiplying equation (1) with − y14
We get
x2 x3 + y 3
− 3 dx + dy = 0 (3)
y y4
The equation (3)of the form M1 dx + N1 dy = 0
x2 x3 + y 3
Where M1 = − 3 and N1 =
y y4

∂M1 3x2 ∂N1 3x2

= 4 and = 4
∂y y ∂x y

∂M 1 ∂N 1
∴ =
∂y ∂x
∴ (3) is an exact Differential equation
The general solution of (1) is same as the general solution of differential equation
(1) The general solution is
Z Z
M1 (x, y) dx + N1 (x, y) dy = C
(y constant) (Only those terms which do not contain x)

9
Mathematics-I Mathematics Faculty 10

x2
Z   Z
1
− dx + dy = C
y3 y
Z
1
x2


− dx + log |y| = C
y3
∴ The general solution of (1)is
x3
− + log |y| = C (4)
3y 2

4 Linear Equation
Linear equation: A differential equation is said to be linear if the dependent
variable and its differential coefficients occur only in the first degree and not
multiplied together.
Thus, the standard form of a linear equation of the first order, commonly known
as Leibnitz’s Linear equation, is
dy
+ P (x)y = Q(x) (1)
dx
where P (x) and Q(x) are the function of x. R
To solve the equation (1) , multiply both sides by e P (x)dx so that we get,
dy R P (x)dx  R  R
e + y e P (x)dx P (x) = Q(x) e P (x)dx (2)
dx
i.e
d  R P (x)dx  R
ye = Q(x) e P (x)dx (3)
dx
Integrating both sides with respect to x on equation (3)
R
Z R
P (x)dx
ye = Q(x) e P (x)dx dx + c (4)

is the required solution

R
Note: The factor e P (x)dx on multiplying by which the left-hand side of equa-
tion (1) becomes the differential coefficient of a single function is called the
Integrating Factor (I.F) of linear equation (1).

It is important to remember that

i) If the given equation is of the form
dy
+ P (x)y = Q(x) (1)
dx
R
P (x)dx
R
then I.F= e and the general solution is y(I.F ) = Q(x)(I.F ) dx+c

10
Mathematics-I Mathematics Faculty 11

ii) If the given equation is of the form

dx
+ P (y)x = Q(y) (2)
dy
R
P (y)dy
R
then I.F= e and the general solution is x(I.F ) = Q(y)(I.F ) dy+c

dy 2
Example 1. Solve the differential equation (x + 1) dx − y = e3x (x + 1)
Solution:
Dividing by (x + 1) on both sides of given equation

dy y
− = e3x (x + 1) (1)
dx (x + 1)

Which is linear in y.
dy R
Now the equation (1) is in the form of + P (x)y = Q(x) then I.F= e P (x)dx
dx
and
R
the general solution is y(I.F ) = Q(x)(I.F ) dx + c

1
Where P (x) = − & Q(x) = e3x (x + 1)
x+1

R 1
− dx 1
x + 1 = e− log(x+1) = elog(1+x)−1 =
R
P (x)dx
I.F= e =e
1+x
R
The general solution is y(I.F ) = Q(x)(I.F ) dx + c

y x+1 e3x
= e3x dx + c = e3x dx + c =
R R
i.e +c
x+1 x+1 3
y e3x
The general solution of given differential equation is = +c
x+1 3

Example 2. Solve the differential equation y (log y) dx + (x − log y) dy = 0

Solution:
Re writing the given differential equation
dx x 1
+ = (1)
dy y (log y) y

Which is a Linear equation in x.

dx R
Now the equation (1) is in the form of + P (y)x = Q(y) then I.F= e P (y)dy
dy

11
Mathematics-I Mathematics Faculty 12

R
and the general solution is x(I.F ) = Q(y)(I.F ) dy + c
1
Where P (y) = y (log y) & Q(y) =
y
R 1
dy
y (log y) = elog(log y) = log y 1
I.F = e (∵ Let log y = t, ⇒ dy = dt)
y
R
The general solution is x(I.F ) = Q(y)(I.F ) dy + c
Z
1
x log y = log y dy + c
y
Z
1
= tdt + c (∵ Let log y = t, ⇒ dy = dt)
y
t2
= +c
2
2
(log y)
= +c
2
2
(log y)
The general solution of given differential equation is x log y = +c
2
Example 3. Solve the differential equation 1 + y 2 dx = tan−1 y − x dy

Solution:
Rewriting the given differential equation

dx dx x tan−1 y
1 + y2 = tan−1 y − x or + = (1)
dy dy (1 + y 2 ) (1 + y 2 )
Which is a Linear equation in x.
dx R
Now the equation (1) is in the form of + P (y)x = Q(y) then I.F= e P (y)dy
R dy
and the general solution is x(I.F ) = Q(y)(I.F ) dy + c
1 tan−1 y
Where P (y) = 2
& Q(y) =
(1 + y ) (1 + y 2 )
1
dy
I.F = e (1 + y 2 ) = etan−1 y
R
The general solution is x(I.F ) = Q(y)(I.F ) dy + c

tan−1 y tan−1 y
Z
−1
xetan y = e dy + c
(1 + y 2 )
Z
dy
= tet dt + c (∵ Let tan−1 y = t, ⇒ = dt)
1 + y2
= et (t − 1) + c
−1
= etan y
tan−1 y − 1 + c

12
 Mathematics-I Mathematics Faculty 13

x = tan−1 y − 1 +


The general solution of given differential equation is
−1
ce− tan y

5 Bernoulli’s Equation
dy
The equation dx + P (x)y = Q(x)y n (1), where P (x), Q(x) are functions
of x, is reducible to the Leibnitz’s linear equation and is usually called the
Bernoulli’s equation.

Converting Bernoulli’s equation into Linear Equation:

dy
1. To solve the above equation, divide both sides by y n , so that y −n dx +
P (x)y 1−n = Q(x) (2)
dy dz
2. Let y 1−n = z so that (1 − n)y 1−n−1 dx = dx
1 dz
3. The equation (2) becomes 1−n dx + P (x)z = Q(x) (3)
dz
4. Equation (3) implies that dx + (1 − n)P (x)z = (1 − n)Q(x) Which is
Leibnitz’s Linear equation in z.

dy
Example 1. Solve the differential equation x + y = x3 y 6
dx
Solution:
Step-1 Rewriting the given equation
dy y
+ = x2 y 6 (1)
dx x
Equation (1)is not Linear equation
Step-2 Dividing above equation (1) on both sides by y −6

dy y −5
y −6 + = x2 (2)
dx x

dy dz
Step-3 Let y −5 = z so that −5y −5−1 =
dx dx
then equation (2) implies
1 dz z
− + = x2
5 dx x
multiplying with −5 on both side
dz 5
− z = −5x2 (3)
dx x

13
 Mathematics-I Mathematics Faculty 14

Which is Leibnitz’s Linear equation in z Now the equation (3) is in the

dz R
form of + P (x)z = Q(x) then I.F = e P (x)dx
dx R
and the general solution is z(I.F ) = Q(x)(I.F ) dx + c
−5
Where P (x) = & Q(x) = −5x2
x
R −5 R 1
dx −5 dx
I.F = e x =e x = e−5 log x = x−5
R
The general solution is z(I.F ) = Q(x)(I.F ) dx + c

−5x2
Z
z
= dx + c
x5 x5
Z
1
= −5 dx + c
x3
Z
= −5 x−3 dx + c

x−2
= −5 +c
−2
5
= +c
2x2

Step-4 Substituting z = y −5 in the general solution

The general solution of given differential equation is

y −5 5
5
= 2 +c
x 2x

dy
Example 2. Solve the differential equation + x sin(2y) = x3 cos2 y
dx
Solution:
Step-1 Dividing the given equation on both sides by cos2 y
dy
sec2 y + 2x tan y = x3 (∵ sin(2y) = 2 sin y cos y) (1)
dx
Equation (1)is not Linear equation
dy dz
Step-2 Let tan y = z so that sec2 y = dx
dx
then equation (1) implies
dz
+ 2xz = x3 (2)
dx
Which is Leibnitz’s Linear equation in z

14
 Mathematics-I Mathematics Faculty 15

dz
Step-3 Now the equation (2) is in the form of + P (x)z = Q(x) then I.F =
R dx
P (x)dx
e
R
and the general solution is z(I.F ) = Q(x)(I.F ) dx + c
Where RP (x) = 2x & Q(x) = x3
2
I.F = e 2xdx = ex
R
The general solution is z(I.F ) = Q(x)(I.F ) dx + c
Z
2 2
zex = x3 ex dx + c
Z
1
= tet dt ( Let x2 = t ⇒ 2x dx = dt)
2
1
= et (t − 1) + c
2
1 2
= ex (x2 − 1) + c
2

Step-4 Substituting z = tan y in the general solution The general solution of given
differential equation is
2 1 x2 2
tan yex = e (x − 1) + c
2

6 General solution of Differential equation:

To find the General Solution (Complementary Function)of f (D)y = 0
Consider the nth order homogeneous linear differential equation with constant
coefficients as
dn y dn−1 y dn−2 y dy
n
+ c1 n−1 + c2 n−2 + . . . cn−1 + cn y = 0 (1)
dx dx dx dx
Here c1 , c2 , . . . , cn are all constants. Introducing the notation of differential
d d2 dn
operation D ≡ and higher order operators as D2 = , . . . , D n
= ,
dx dx2 dxn
etc, the differential equation (1) can be rewritten as

Dn y + c1 Dn−1 y + c2 Dn−2 y + . . . + cn−1 Dy + cn y = 0 (2)

or
f (D)y = 0 (3)
where
f (D) = Dn + c1 Dn−1 + c2 Dn−2 + . . . + cn−1 D + cn (4)
is a polynomial in D.

15
Mathematics-I Mathematics Faculty 16

6.1 Auxiliary Equation (A.E)

The equation obtained by replacing D by m in equation (4) is called the auxiliary
equation.
i.e
f (m) = mn + c1 mn−1 + c2 mn−2 + . . . + cn−1 m + cn = 0 (5)
The General solution of equation (1) is obtained according as the nature of the
roots(n) of A.E (5) as follows
Case Nature of the n roots General solution
1 n distinct and real roots (i.e y(x) = c1 em1 x + c2 em2 x + c3 em3 x + . . . + cn emn x
m1 , m2 , m3 , . . . , mn )
2 Two equal roots and (n-2) distinct real roots (i.e, y(x) = (c1 + c2 x)em1 x + c3 em3 x + . . . + cn emn x
m1 , m1 , m3 , . . . , mn )
3 Three equal roots and (n-3) distinct real y(x) = (c1 +c2 x+c3 x2 )em1 x +c4 em4 x +. . . +cn emn x
roots (i.e, m1 , m1 , m1 , m4 , . . . , mn )
4 complex conjugate roots and (n-2) distinct y(x) = epx (c1 cos qx + c2 sin qx) + c3 em3 x + . . . +
real roots (i.e, m1 = p + iq, m2 = p − cn emn x
iq, m3 , m4 , . . . , mn )
5 Two equal complex conjugate roots and (n-4) y(x) = epx [(c1 + c2 x) cos qx + (c3 + c4 x) sin qx] +
distinct real roots (i.e, m1 = m2 = p + iq, m3 = c5 em5 x + . . . + cn emn x
m4 = p − iq, m5 , . . . , mn )

6.2 Second and higher order linear differential equations

of with constant coefficients
d2 y dy
Example 1. Solve 2
+5 + 6y = 0
dx dx
Solution: The given equation in Operator form is (D2 + 5D + 6)y = 0
i.e f (D)y = 0
The A.E is f (m) = 0 i.e., m2 + 5m + 6 = 0
hence m = −2, −3 are roots of A.E.
Therefore roots are real and distinct
General Solution is : y(x) = c1 e−2x + c2 e−3x

d2 y dy
Example 2. Solve − 12 + 36y = 0
dt2 dt
Solution: The given equation in Operator form is (D2 − 12D + 36)y = 0
i.e f (D)y = 0
The A.E is f (m) = 0 i.e., m2 − 12m + 36 = 0
hence m = 6, 6 are roots of A.E.
Therefore roots are real and double (repeated)
General Solution is : y(t) = (c1 + c2 t) e6t

d4 y d3 y d2 y
Example 3. Solve 4
+6 3 +9 2 =0
dt dt dt

16
Mathematics-I Mathematics Faculty 17

Solution: The given equation in Operator form is (D4 + 6D3 + 9D2 )y = 0

i.e f (D)y = 0

The A.E is f (m) = 0 i.e., m4 + 6m3 + 9m2 = 0 i.e., m2 m2 + 6m + 9 = 0
hence m = 0, 0, −3, −3 are roots of A.E.
Therefore m = 0 and m = −3 are real and double (repeated)
General Solution is : y(t) = (c1 + c2 t) e0.t + (c3 + c4 t) e−3t

d5 y d3 y
Example 4. Solve − 3 =0
dt5 dt
Solution: The given equation in Operator form is (D5 − D3 )y = 0
i.e f (D)y = 0

The A.E is f (m) = 0 i.e., m5 − m3 = 0 i.e., m3 m2 − 1 = 0
hence m = 0, 0, 0, 1, −1 are roots of A.E.
Therefore m = 0 is triple (repeated 3 times) and m = −1, 1 are real and distinct
roots
General Solution is : y(t) = c1 + c2 t + c3 t2 e0.t + c4 e−x + c5 et

Example 5. Solve D2 − 4D + 1 y = 0
Solution: The given equation in Operator form is (D2 − 4D + 1)y = 0
i.e f (D)y = 0
2
The A.E is f (m) √ m − 4m + 1 = 0
√ = 0 i.e.,
hence m = 2 + 3, 2 − 3 are roots of A.E.
Therefore roots are real and distinct
√ √
General Solution is : y(x) = c1 e(2+ 3)x + c2 e(2− 3)x

d3 y d2 y dy
Example 6. Solve 3
+ 2 +4 + 4y = 0
dx dx dx
Solution: The given equation in Operator form is (D3 + D2 + 4D + 4)y = 0
i.e f (D)y = 0
The A.E is f (m) = 0 i.e., m3 + m2 + 4m + 4 = 0 i.e., (m2 + 4)(m + 1) = 0
hence m = −1, ±2i are roots of A.E.
Therefore m = −1 is real and m = ±2i is complex conjugate root
General Solution is : y(x) = c1 e−x + c2 cos 2x + c3 sin 2x

d4 y d2 y
Example 7. Solve 4
+ 8 2 + 16y = 0
dx dx
Solution: The given equation in Operator form is (D4 + 8D2 + 16)y = 0
i.e f (D)y = 0
The A.E is f (m) = 0 i.e., m4 + 8m2 + 16 = 0
hence m = ±2i, ± 2i are roots of A.E.
General Solution is : y(x) = (c1 + c2 x) cos 2x + (c3 + c4 x) sin 2x

d4 y
Example 8. Solve + 4y = 0
dx4

17
Mathematics-I Mathematics Faculty 18

Solution: The given equation in Operator form is (D4 + 4)y = 0

i.e f (D)y = 0
The A.E is f (m) = 0 i.e., m4 + 4 = 0
hence m = 1 ± i and m = −1 ± i are roots of A.E.
General Solution is : y(x) = e−x (c1 cos x + c2 sin x) + ex (c3 cos x + c4 sin x)

d2 y dy dy
Example 9. Solve + 5 + 6y = 0, given y(0) = 0, and at x = 0 =
dx2 dx dx
15
Solution: The given equation in Operator form is (D2 + 5D + 6)y = 0
i.e f (D)y = 0
The A.E is f (m) = 0 i.e., m2 + 5m + 6 = 0
hence m = −2 and m = −3 are roots of A.E.
General Solution is :
y(x) = c1 e−2x + c2 e−3x (1)
dy
= −2c1 e−2x − 3c2 e−3x (2)
dx
given y(0) = 0 equation (1) =⇒ 0 = c1 + c2 =⇒ c1 = −c2
dy
Also given at x = 0, = 15 equation(2) =⇒ , 15 = −2c1 − 3c2
dx
substituting c1 = −c2 , =⇒ 15 = 2c2 − 3c2 =⇒ c2 = −15 and c1 = 15
Hence the required solution is y(x) = 15e−2x − 15e−3x = 15(e−2x − e−3x )

6.3 Inverse Operator

1
x is that function of x, not containing arbitrary constants which when
f (D)
n 1 o
operated upon by f (D) gives X. i.e., f (D) X =X
f (D)
1
Thus X satisfies the equation f (D)y = X and is, therefore, its particular
f (D)
integral(P.I).

7 Rules for finding the particular Integral

Consider the nth order non-homogeneous linear differential equation with con-
stant coefficients as
dn y dn−1 y dn−2 y dy
n
+ c1 n−1 + c2 n−2 + . . . + cn−1 + cn y = X (1)
dx dx dx dx


which is in symbolic form Dn + c1 Dn−1 + c2 Dn−2 + . . . + cn−1 D + cn y = X
or f (D)y = X

18
 Mathematics-I Mathematics Faculty 19

1
Therefore P.I = X
f (D)
The complete solution of the equation (1) is y = C.F + P.I, where (C.F) is the
Complementary Function and (P.I) is the Particular Integral of the equation
(1).

7.1 Non-homogeneous of the type X = eax

Case (i): If f (a) 6= 0
1 ax 1 ax
P.I= e = e , Provided f (a) 6= 0.
f (D) f (a)
Case (ii): If f (a) = 0 but f 0 (a) 6= 0
1 ax 1
P.I= e = x. 0 eax , Provided f 0 (a) 6= 0.
f (D) f (a)
Case (iii): If f (a) = 0, f 0 (a) = 0 but f 00 (a) 6= 0
1 ax 1
P.I= e = x2 . 00 eax , Provided f 00 (a) 6= 0.
f (D) f (a)
Case (iv): (D −a) is a factor of f (D). Suppose f (D) = (D −a)φ(D), where φ(a) 6= 0.
Then
1 ax 1 1 ax 1 n 1 o 1
e = . e = . eax = x.eax
f (D) D − a φ(D) φ(a) D − a φ(a)
1 ax xn ax
Case (v): e = e
(D − a)n n!
1
.X = eax X.e−ax dx, no arbitrary constant being added.
R
Case (vi):
D−a


Example 1. Find the P.I of 4D2 + 4D − 3 y = e2x


Solution: The given equation in Operator form is 4D2 + 4D − 3 y = e2x
i.e f (D)y = e2x
1 2x 1 1
P.I= e = e2x = e2x , since f (2) 6= 0
f (D) 4D2 + 4D − 3 4(2)2 + 4(2) − 3
1 2x
= e
21


Example 2. Find the P.I of 4D2 + 4D − 3 y = 5


Solution: The given equation in Operator form is 4D2 + 4D − 3 y = 5
i.e f (D)y = 5.e0x
1 0x 1 1
P.I= e = e2x = e0x , since f (0) 6= 0
f (D) 4D2 + 4D − 3 4(0)2 + 4(0) − 3
1
=−
3

19
Mathematics-I Mathematics Faculty 20

Example 3. Find the P.I of D3 − 3D2 + 4 y = e2x


Solution: The given equation in Operator form is D3 − 3D2 + 4 y = e2x
i.e f (D) = D3 − 3D2 + 4, f (2) = 0, f 0 (D) = 3D2 − 6D, f 0 (2) = 0
But f 00 (D) = 6D − 6, f 00 (2) = 6 6= 0
1 2x 1 2x
P.I= e = x2 e
f (D) 6


Example 4. Find the P.I of D2 − 1 y = 2 sinh x
Solution: The given equation in Operator form is [(D + 1)(D − 1)] y = ex −
e−x , since 2 sinh x = ex − e−x
1
P.I= [ex − e−x ]
(D + 1)(D − 1)
Let us evaluate each of these terms separately.  
1 x 1 x 1 1 x 1
(i) e = e = e = .xex
(D + 1)(D − 1) (1 + 1)(D − 1) 2 D−1 2
 
1 −x 1 1 1 −x 1
(ii) e = x
e =− e = − .xe−x
(D + 1)(D − 1) (D + 1)(−1 − 1) 2 D+1 2
1 1
Therefore P.I= .xex − .xe−x
2 2
x x
Therefore P.I= [e − e−x ]
2
d2 y dy x
Example 5. Solve: 2
+3 + 2y = ee
dx dx
x
Solution: The given equation in Operator form is (D2 + 3D + 2)y = ee
The A.E is f (m) = 0 =⇒ m2 + 3m + 2 = 0 m = −1, − 2
Therefore C.F=c1 e−x + c2 e−2x
 
1 x 1 1 x
P.I= ee = − ee
(D + 1)(D + 2) D+1 D+2
1 x 1 x
P.I= ee − ee
D+1 D+2
Let us evaluate each of these terms separately.  
1 x 1 x R x 1
ee = ee = e−x ee .ex dx, Since .X = e−ax X.eax dx
R
(i)
D+1 D − (−1) D − (−a)
= e−x et dt where t = ex , =⇒ dt = ex dx
R
x
= e−x .et = e−x .ee
1 x 1 x R x R x
(ii) ee = ee = e−2x ee .e2x dxe−2x ee .ex .ex dx, Since
 D+2 D − (−2) 
1 −ax
R ax
.X = e X.e dx
D − (−a)
= e−2x et .t dt where t = ex , =⇒ dt = ex dx
R

20
 Mathematics-I Mathematics Faculty 21

x
= e−x .et (t − 1) = e−2x .ee (ex − 1)
x x x x
= ee (e−x − e−2x ) Therefore P.I=e−x .ee − ee (e−x − e−2x ) = e−2x .ee
The complete solution is y = C.F + P.I
x
i.e., y(x) = c1 e−x + c2 e−2x + e−2x .ee

7.2 Non-homogeneous of the type X = sin(ax+b) or cos(ax+

b)
Case (i): If f (−a2 ) 6= 0
1 1
P.I= 2
sin(ax + b) = sin(ax + b), Provided f (−a2 ) 6= 0.
f (D ) f (−a2 )
Case (ii): If f (−a2 ) = 0 but f 0 (−a2 ) 6= 0
1 1
P.I= sin(ax + b) = x 0 sin(ax + b), Provided f 0 (−a2 ) 6= 0.
f (D2 ) f (−a2 )
Case (iii): If f (−a2 ) = 0, f 0 (−a2 ) = 0 but f 00 (−a2 ) 6= 0
1 1
P.I= sin(ax+b) = x2 00 sin(ax+b), Provided f 00 (−a2 ) 6= 0.
f (D2 ) f (−a2 )

Example 1. Solve: (D2 + 4)y = cos 2x

Solution: The given equation in Operator form is (D2 + 4)y = cos 2x
Auxiliary equation is m2 + 4 = 0
m = ±2i, Therefore C.F=c1 cos 2x + c2 sin 2x
   
1 1 x 1 x 1 x
P.I= 2 cos 2x = x. cos 2x = cos 2x = sin 2x = sin 2x
D +4 2D 2 D 2 2 4
x
Complete solution is y = C.F + P.I = c1 cos 2x + c2 sin 2x + sin 2x
4

Example 2. Solve: (D3 + 1)y = cos(2x − 1)

Solution: The given equation in Operator form is (D3 + 1)y √ = cos(2x − 1)
√
−1 − i 3 −1 + i 3
Auxiliary equation is m3 + 1 = 0 =⇒ , m = 1, m = ,
2 2
x −x/2
 √ √ 
Therefore C.F=c1 e + e c2 cos( 3/2)x + c3 sin( 3/2)x
1
P.I= cos(2x − 1) Put D2 = −22 = −4
D3
+1
1
= cos(2x − 1) Multiply and divide by 1 + 4D
D(−4) + 1
(1 + 4D) 1
= cos(2x − 1) = (1 + 4D). cos(2x − 1)
(1 − 4D)(1 + 4D) 1 − 16D2
1 1
=(1 + 4D). cos(2x − 1) = [cos(2x − 1) + 4D cos(2x − 1)]
1 − 16(−4) 65

21
Mathematics-I Mathematics Faculty 22

1
= [cos(2x − 1) − 8 sin(2x − 1)]
65
Hence, the complete solution is y(x) = C.F + P.I
√ √  1
y(x) = c1 ex +e−x/2 c2 cos( 3/2)x + c3 sin( 3/2)x + [cos(2x − 1) − 8 sin(2x − 1)]

65

d3 y dy
Example 3. Solve: +4 = sin 2x.
dx3 dx
Solution: The given equation in Operator form is (D3 + 4D)y = sin 2x
The A.E is f (m) = 0 =⇒ m3 + 4m = 0 =⇒ m = 0, ± 2i
Therefore C.F=c1 + (c2 cos 2x + c3 sin 2x)
1  2 
P.I= sin 2x D + 4 = 0, for D2 = −22
D (D2 + 4)
 
1 d
P.I=x. 2 sin 2x, Since (D3 + 4D) = 3D2 + 4
3D + 4 dD
1 x
=x. sin 2x = − sin 2x Put D2 = −22 = −4
3(−4) + 4 8
Hence, the complete solution is y(x) = C.F + P.I = c1 + (c2 cos 2x + c3 sin 2x) −
x
sin 2x
8

d3 y d2 y dy
Example 4. Solve: 3
− 3 3
+4 − 2y = ex + cos x
dx dx dx
Solution: The given equation in Operator form is (D3 − 3D2 + 4D − 2)y =
ex + cos x
The A.E is f (m) = 0 =⇒ m3 − 3m2 + 4m − 2 = 0 =⇒ m = 1, 1 ± i
Therefore C.F=c1 ex + ex (c2 cos x + c3 sin x)
1 1
P.I= ex + 3 cos x
(D − 1)(D2 − 2D + 2) D − 3D2 + 4D − 2
1 1
= ex + cos x
(D − 1)(1 − 2 + 2) (−1)D − 3(−1) + 4D − 2
1 1
= ex + cos x
D−1 3D + 1
3D − 1 (3D − 1)
=x.ex + cos x = x.ex + cos x
9D2 − 1 9(−12 ) − 1
−3 sin x − cos x 1
=x.ex + = x.ex + (3 sin x + cos x)
−10 10
Hence, the complete solution is y(x) = C.F +P.I = c1 ex +ex (c2 cos x + c3 sin x)+
1
x.ex + (3 sin x + cos x)
10

22
Mathematics-I Mathematics Faculty 23

7.3 Non-homogeneous of the type X = xn

1 −1
Here P.I= xn = [f (D)] xn
f (D)
−1
Expand [f (D)] in ascending powers of D as far as the terms in Dn term by
term. Since the (n + 1)th and higher derivatives of xn are zero, we need not
consider terms beyond Dn .


Example 1. Solve: D2 + D − 6 y = x
Solution: The given equation in Operator form is (D2 + D − 6)y = x
The A.E is f (m) = 0 =⇒ m2 + m − 6 = 0 =⇒ m = −3, 2
Therefore C.F=c1 e−3x + c2 e2x
1 1
P.I= 2 x= x
D D2

D +D−6
−6 1 − −
6 6
1 h  i−1
D2
=− 1 − D 6 + 6 x
6
1h 
D2
 i
=− 1 + D 6 + 6 + . . . x
6
   
1 1 1 1
=− x + D(x) = − x +
6 6 6 6
 
−3x 2x 1 1
Hence, the complete solution is y(x) = C.F + P.I = c1 e + c2 e − x+
6 6

Example 2. Find the P.I of (D2 + D)y = x2 + 2x + 4

Solution: The given equation in Operator form is (D2 + D)y = x2 + 2x + 4
1 1 −1

P.I= (x2 + 2x + 4) = [1 + D] x2 + 2x + 4
D(D + 1) D
1  

= 1 − D + D2 − D3 + . . . x2 + 2x + 4
D
1  2 
= x + 2x + 4 − D(x2 + 2x + 4) + D2 (x2 + 2x + 4) + D3 (x2 + 2x + 4)
D
1  2 
= x + 2x + 4 − (2x + 2) + (2) + D3 (0)
D
x3
= (x2 + 4) dx =
R
+ 4x
3

7.4 Non-homogeneous of the type X = eax V (x)

 
1 1
Here P.I= [eax V (x)] = eax .V (x)
f (D) f (D + a)

23
Mathematics-I Mathematics Faculty 24

1
Thus eax which is on the right of may be taken out to the left provided
f (D)
D is replaced by D + a.


Example 1. Find the P.I of D2 − 4D + 3 y = ex cos 2x
1 1
Solution: P.I= ex cos 2x = ex cos 2x
D2 − 4D + 3 (D + 1)2 − 4(D + 1) + 3
1 1  
=ex 2 cos 2x = ex 2 cos 2x Putting D2 = −22
D − 2D −2 − 2D
1 1
=− ex cos 2x
2 2+D
1 2−D 1 2−D
=− ex cos 2x = − ex cos 2x
2 4 − D2 2 4 − (−22 )
1 1
=− ex (2 cos 2x − D(cos 2x)) = − ex (2 cos 2x + 2 sin 2x)
16 16
1
=− ex (cos 2x + sin 2x)
8


Example 2. Solve: D2 − 4D + 4 y = e2x x3
Solution: A.E is f (m) = 0, i.e., m2 − 4m + 4 = 0, =⇒ m = 2, 2
C.F=(c1 + c2 x) e2x
1 1
P.I= 2 e2x x3 = e2x 2
x3
D − 4D + 4 (D + 2) − 4(D + 2) + 4
 4 5
2x 1 3 2x 1 x 2x x
=e x = e . = e .
D2 D 4 20
x5
The complete solution is y(x) = C.F + P.I = (c1 + c2 x) e2x + e2x .
20


Example 3. Solve: D3 − 7D2 + 10D y = e2x sin x
Solution:
A.E is f (m) = 0, i.e., m3 − 7m2 + 10m = 0, =⇒ m = 0, 2, 5
C.F=c1 + c2 e2x + c3 e5x
1 1
P.I= 3 2
e2x sin x = e2x . sin x
D − 7D + 10D (D + 2) − 7(D + 2)2 + 10(D + 2)
3

1
=e2x . sin x
D3 + 6D2 + 12D + 8 − 7D2 − 28D − 28 + 10D + 20
1 1
=e2x 3 sin x = e2x sin x
D − D2 − 6D (−12 )D − (−12 ) − 6D
1 1
=e2x sin x = e2x sin x
−D + 1 − 6D 1 − 7D
1 + 7D 1 + 7D
=e2x sin x = e2x sin x
1 − 49D2 1 − 49(−12 )

24
Mathematics-I Mathematics Faculty 25

1 + 7D e2x
=e2x sin x = [sin x + 7 cos x]
50 50
e2x
The complete solution is y(x) = C.F +P.I = c1 +c2 e2x +c3 e5x + [sin x + 7 cos x]
50

7.5 Non-homogeneous of the type X = x.V (x)

Example 1. Solve: D2 − 2D + 1 y = xex sin x
Solution: A.E is f (m) = 0, i.e., m2 − 2m + 1 = 0, =⇒ m = 1, 1
C.F=(c1 + c2 x)ex
1 1
P.I= 2
ex .x sin x = ex x sin x
(D − 1) (D + 1 − 1)2
1 1 R
=ex 2 x sin x = ex x sin x
D D
Integrating by parts,
1   1
P.I=ex x(− cos x) − 1(− cos x)dx = ex (−x cos x + sin x)
R
R D  D
=ex (−x cos x + sin x) dx
h n o i
=ex − x sin x − 1. sin xdx − cos x
R

=ex [−x sin x − cos x − cos x] = −ex [x sin x + 2 cos x]

The complete solution is y(x) = C.F + P.I = (c1 + c2 x)ex − ex [x sin x + 2 cos x]

Example 2. Solve: D4 + 2D2 + 1 y = x2 cos x
Solution: A.E is f (m) = 0, i.e., m4 + 2m2 + 1 = 0, =⇒ (m2 + 1)2 =
0, =⇒ m = ±i, ± i
C.F=(c1 + c2 x) cos x + (c3 + c4 x) sin x
( )
1 1
1
P.I= 2 x2 cos x = x2 Re.P.of e ix
=Re.P.of eix .x2
(D + 1)2 (D2 + 1)2 (D2 + 1)2
n 1 o
=Re.P.of eix 2 x2
[(D + i)2 + 1)]
( ) ( )
n 1 o 1 1
=Re.P.of eix 2 x2
= Re.P.of
e ix
2 x
2
= Re.P.of e ix
 x
2
[D2 + 2iD] Di 2
D 
−4D2 2i + 1 −4D2 1 + 2i i
" (  −2 )#
1 i
=Re.P.of eix − 1− D x2
4D2 2
" (  2 ) #
1 ix 1 iD iD
=Re.P.of − e . 2 1+2 +3 + . . . x2
4 D 2 2

25
Mathematics-I Mathematics Faculty 26

" ( )# " ( )#
1 ix 1 2 3 1 ix 1 x3 2 3
=Re.P.of − e . 2 x + 2ix − = − e . + ix − x
4 D 2 4 D 3 2
" ( )#
x4 x3
 4
x3
 
1 3 1 x 3
=Re.P.of − eix + i − x2 =− Re.P.of (cos x + i sin x) + i − x2
4 12 3 4 4 12 3 4
( ) (
−1
1

= Re.P.of (cos x + i sin x) x4 + 4ix3 − 9x2 =− x4 − 9x2 cos x −
48 48
)
4x3 sin x

Hence the complete solution is y(x) = C.F + P.I ( )

1 4 2

3
y(x) = (c1 + c2 x) cos x + (c3 + c4 x) sin x − x − 9x cos x − 4x sin x
48


Example 3. Solve: D2 − 1 y = x sin 3x + cos x


Example 4. Solve: D2 − 4D + 4 y = 8x2 e2x sin 2x


Example 5. Solve: D2 + a2 y = sec ax

26
Method of Variation of Parameters
Example 3. Consider
Practice Questions:

1.

2.

3.
Laplace transforms

Unit-4
Laplace transforms
Course Objective:
Develop ability to Analyze properties of Laplace Transform, Inverse Laplace Transform and
convolution theorem.
Course Outcome:
At the end of course, the student would be able to evaluate various problems using Laplace Transform
and Inverse Laplace Transform
Main Contents of the unit:
 Laplace transforms in different cases
 Inverse Laplace transforms in different cases
 Convolution theorem
 Solution of Ordinary Differential Equations by using Laplace transforms
Introduction:
 Laplace transforms is an application of integral transform which produces a new function 𝑓(𝑠)
by integrating the product of kernel and the given function𝑓(𝑡) between the suitable limits. i.e.,
𝑡
the integral transform is 𝑓 (𝑠) = ∫𝑡 2 𝑘 (𝑠, 𝑡) ∗ 𝑓(𝑡)𝑑𝑡
1
 The Laplace transforms is an essential mathematical part for engineers.
 The Laplace transform is a very much useful mathematical tool for solving linear differential
equations with given initial conditions (initial value) without finding the general solution.
 The Laplace transforms is a powerful tool in diverse fields of engineering.

Definition of Laplace transforms:

The Laplace transform of a function 𝑓(𝑡) can be denoted by 𝐿{𝑓(𝑡)} or 𝑓 ̅(𝑠) and is defined as
∞
𝑳{𝒇(𝒕)} = 𝒇̅(𝒔) = ∫𝟎 𝒆−𝒔𝒕 𝒇(𝒕) 𝒅𝒕 for all 𝑡 > 0

where the parameter 𝒔 is a real or complex number and the symbol 𝑳 is known as Laplace transform
operator.
Sufficient conditions for existence of Laplace transform of a function:
A function 𝑓(𝑡) is said to have Laplace transform if it has the following conditions.

1. The function 𝑓(𝑡) must be piece-wise (sectionally) continuous in any limited interval.
2. The function 𝑓(𝑡) is of exponential order.
Note:
1. Piece-wise (sectionally) continuous: A function 𝑓(𝑡) is said to be piece-wise continuous in
[a, b] if [a, b] can be divided into finite number of sub intervals and the following conditions
holds.
i. 𝑓 (𝑡) is continuous in each sub interval
ii. 𝑓 (𝑡) has both right- and left-hand limits at every end points of each sub interval.
𝟐
Ex: 𝒇(𝒕) = {𝒕 , 𝟎 < 𝒕 < 𝟓 is sectionally continuous for 𝒕 > 𝟎
𝟐𝒕 + 𝟑, 𝒕 > 𝟓

1
Laplace transforms

2. Exponential order: A function 𝑓(𝑡) is said to be of exponential order 𝒂 if

−𝑎𝑡 ( )
lim 𝑒 𝑓 𝑡 𝑖𝑠 𝑓𝑖𝑛𝑖𝑡𝑒
𝑡→∞

Ex: 𝒇(𝒕) = 𝒆𝒂𝒕 { 𝒔𝒊𝒏𝒄𝒆 lim 𝑒 −𝑎𝑡 𝑓(𝑡) = lim 𝑒 −𝑎𝑡 𝒆𝒂𝒕 = 𝟏 < ∞ }
𝑡→∞ 𝑡→∞

3. Gamma function: The gamma function of 𝑛 > 0 can be denoted by Γ(n) and is defined as
∞ −𝒙 𝒏−𝟏
𝛤(𝑛) = ∫𝟎 𝒆 𝒙 𝒅𝒙

4. Properties of Gamma function:

i. 𝛤(𝑛 + 1) = 𝑛 𝛤(𝑛), if 𝑛 > 0
ii. 𝛤(𝑛 + 1) = 𝑛! , if 𝑛 ∈ 𝑍 +
iii. 𝛤(1) = 1
1
iv. 𝛤 (2) = √𝜋
1
v. 𝛤 (− ) = −2 √𝜋
2
vi. 𝛤(0), 𝛤(−1), 𝛤(−2), …are all not defined.

General Properties of Laplace transform:

Property 4.1: If 𝑳{𝒇(𝒕)} = 𝒇̅(𝒔) then for any constant k, 𝑳{𝒌 𝒇(𝒕)} = 𝒌 𝑳{𝒇(𝒕)} = 𝒌 𝒇̅(𝒔)
Proof:
∞
Given 𝐿{𝑓(𝑡)} = 𝑓 ̅(𝑠) , by definition, 𝐿{𝑓 (𝑡)} = 𝑓 ̅(𝑠) = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡
∞ ∞
then 𝐿{𝑘 𝑓 (𝑡)} = ∫0 𝑒 −𝑠𝑡 𝑘 𝑓(𝑡) 𝑑𝑡 = 𝑘 ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 = 𝑘 𝐿{𝑓(𝑡)} = 𝑘 𝑓 ̅(𝑠)

Hence 𝑳{𝒌 𝒇(𝒕)} = 𝒌 𝒇̅(𝒔)

* Property 4.2 (Linear property): If 𝑳{𝒇(𝒕)} = 𝒇̅(𝒔) and 𝑳{𝒈(𝒕)} = 𝒈(𝒔) then for any constants
𝒄𝟏 and 𝒄𝟐 , 𝑳{ 𝒄𝟏 𝒇(𝒕) + 𝒄𝟐 𝒈(𝒕)} = 𝒄𝟏 𝑳{𝒇(𝒕)} + 𝒄𝟐 𝑳{𝒈(𝒕)} = 𝒄𝟏 𝒇̅(𝒔) + 𝒄𝟐 𝒈
̅ (𝒔)

Proof:
∞
Given 𝐿{𝑓 (𝑡)} = 𝑓 ̅(𝑠) = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 and
∞
𝐿{𝑔(𝑡)} = 𝑔̅ (𝑠) = ∫0 𝑒 −𝑠𝑡 𝑔(𝑡) 𝑑𝑡

Then consider
∞
𝑳{ 𝒄𝟏 𝒇(𝒕) + 𝒄𝟐 𝒈(𝒕)} = ∫0 𝑒 −𝑠𝑡 [𝒄𝟏 𝒇(𝒕) + 𝒄𝟐 𝒈(𝒕)] 𝑑𝑡
∞
= ∫0 [ 𝒄𝟏 𝑒 −𝑠𝑡 𝒇(𝒕) + 𝒄𝟐 𝑒 −𝑠𝑡 𝒈(𝒕)] 𝑑𝑡
∞ ∞
= 𝑐1 ∫0 𝑒 −𝑠𝑡 𝑓(𝑡)𝑑𝑡 + 𝑐2 ∫0 𝑒 −𝑠𝑡 𝑔(𝑡)𝑑𝑡

= 𝑐1 𝐿{𝑓 (𝑡)} + 𝑐2 𝐿{𝑔(𝑡)}

= 𝑐1 𝑓 ̅(𝑠) + 𝑐2 𝑔̅ (𝑠)

Hence 𝑳{ 𝒄𝟏 𝒇(𝒕) + 𝒄𝟐 𝒈(𝒕)} = 𝒄𝟏 𝒇̅(𝒔) + 𝒄𝟐 𝒈

̅ (𝒔)

2
Laplace transforms

4.3 Laplace transforms of Some Elementary (standard) functions:

𝒌
1. 𝑳{𝒌} = 𝒔 (𝒔 > 𝟎), 𝒘𝒉𝒆𝒓𝒆 𝒌 𝒊𝒔 𝒂 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕.

∞
Proof: Since by definition, 𝐿{𝑓(𝑡)} = 𝑓(̅ 𝑠) = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡
∞ ∞ ∞
−𝑠𝑡 −𝑠𝑡
𝑒 −𝑠𝑡 𝑘 𝑘 𝑘
=> 𝐿{𝑘} = ∫ 𝑒 𝑘 𝑑𝑡 = 𝑘 ∫ 𝑒 𝑑𝑡 = 𝑘 [ ] = − [𝑒 −∞ − 𝑒 0 ] = − [0 − 1] = −
−𝑠 0 𝑠 𝑠 𝑠
0 0

𝒌
Hence 𝑳{𝒌} = 𝒔
𝟏 𝟐
Note: From the above result, we will observe 𝑳{𝟎} = 𝟎, 𝑳{𝟏} = 𝒔 , 𝑳{𝟐} = 𝒔 and so on…
𝟏
2. 𝑳{𝒕} = 𝒔𝟐

∞
Proof: Since by definition, 𝐿{𝑓(𝑡)} = 𝑓(̅ 𝑠) = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡
∞
∞ 𝑒 −𝑠𝑡 𝑒 −𝑠𝑡
=> 𝐿{𝑡} = ∫0 𝑒 −𝑠𝑡 𝑡 𝑑𝑡 = [𝑡 ( ) − 1 ( )] { 𝒔𝒊𝒏𝒄𝒆 ∫ 𝑼𝑽𝒅𝒙 = 𝑼𝑽𝟏 − 𝑼′ 𝑽𝟐 + 𝑼′′ 𝑽𝟑 − ⋯}
−𝑠 (−𝑠)2 0

𝟏
= [(𝟎 − 𝟎) − (𝟎 − 𝒔𝟐)]
𝟏
Hence 𝑳{𝒕} = 𝒔𝟐

Similarly, we get the Laplace transforms of some other elementary functions stated below.

S No. 𝒇(𝒕) 𝑳{𝒇(𝒕)}

1 𝒌 𝒌
𝑳{𝒌} = (𝒔 > 𝟎), 𝒘𝒉𝒆𝒓𝒆 𝒌 𝒊𝒔 𝒂 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕.
𝒔

2 𝒕 𝟏
𝑳{𝒕} =
𝒔𝟐
3 𝒕𝒏 𝒏!
𝑳{𝒕𝒏 } = 𝒔𝒏+𝟏 when 𝒏 > 𝟎

4 𝒕𝒏 𝑳{𝒕𝒏 } =
𝜞(𝒏+𝟏)
when 𝒏 > −𝟏 & 𝒏 𝒊𝒔 𝒏𝒐𝒕 𝒁+
𝒔𝒏+𝟏

5 𝒆−𝒂𝒕 𝟏
𝑳{ 𝒆−𝒂𝒕 } =
𝒔+𝒂
6 𝒆𝒂𝒕 𝟏
𝑳{ 𝒆𝒂𝒕 } =
𝒔−𝒂
7 𝒔𝒊𝒏 𝒂𝒕 𝒂
𝑳{𝒔𝒊𝒏 𝒂𝒕} = 𝟐
𝒔 + 𝒂𝟐
8 𝒄𝒐𝒔 𝒂𝒕 𝒔
𝑳{𝒄𝒐𝒔 𝒂𝒕} = 𝟐
𝒔 + 𝒂𝟐
9 𝒔𝒊𝒏𝒉 𝒂𝒕 𝒂
𝑳{𝒔𝒊𝒏𝒉 𝒂𝒕} = 𝟐
𝒔 − 𝒂𝟐
10 𝒄𝒐𝒔𝒉 𝒂𝒕 𝒔
𝑳{𝒄𝒐𝒔𝒉 𝒂𝒕} = 𝟐
𝒔 − 𝒂𝟐

3
Laplace transforms

Problems:
1,0<𝑡<2
2,2<𝑡<4
1. Find the Laplace transform of 𝑓 (𝑡) = {3,4<𝑡<6
0,𝑡>6
Sol:
∞
Since by definition, 𝐿{𝑓 (𝑡)} = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡
∞
 𝐿{𝑓(𝑡)} = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡
2 4 6 ∞
 𝐿{𝑓 (𝑡)} = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 + ∫2 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 + ∫4 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 + ∫6 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡
2 4 6 ∞
 𝐿{𝑓(𝑡)} = ∫0 1. 𝑒 −𝑠𝑡 𝑑𝑡 + ∫2 2. 𝑒 −𝑠𝑡 𝑑𝑡 + ∫4 3. 𝑒 −𝑠𝑡 𝑑𝑡 + ∫6 0. 𝑒 −𝑠𝑡 𝑑𝑡
2 4 6
𝑒 −𝑠𝑡 𝑒 −𝑠𝑡 𝑒 −𝑠𝑡
 𝐿{𝑓(𝑡)} = [ −𝑠 ] + 2. [ −𝑠 ] + 3. [ −𝑠 ] + 0
0 2 4
𝑒 −2𝑠 1 𝑒 −4𝑠 𝑒 −2𝑠 𝑒 −6𝑠 𝑒 −4𝑠
 𝐿{𝑓(𝑡)} = [ −𝑠
+ ] + 2. [
𝑠 −𝑠
+
𝑠
] + 3. [
−𝑠
+
𝑠
]
1
 𝐿{𝑓(𝑡)} = 𝑠 [−𝑒 −2𝑠
+ 1 − 2. 𝑒 −4𝑠 + 2. 𝑒 −2𝑠 − 3. 𝑒 −6𝑠 + 3. 𝑒 −4𝑠 ]
𝟏
 𝑳{𝒇(𝒕)} = 𝒔 [𝟏 + 𝒆−𝟐𝒔 + 𝒆−𝟒𝒔 − 𝟑. 𝒆−𝟔𝒔 ]

2. Determine the Laplace transform of

𝒆𝟑𝒕 − 𝟐𝒆−𝟐𝒕 + 𝐬𝐢𝐧 𝟐𝒕 + 𝐜𝐨𝐬 𝟑𝒕 + 𝐬𝐢𝐧𝐡 𝟑𝒕 − 𝟐 𝐜𝐨𝐬𝐡 𝟒𝒕 + 𝟗
Sol:
𝐿{𝑒3𝑡 − 2𝑒−2𝑡 + sin 2𝑡 + cos 3𝑡 + sinh 3𝑡 − 2 cosh 4𝑡 + 9}
= 𝐿{𝑒 3𝑡 } − 2. 𝐿{𝑒 −2𝑡 } + 𝐿{sin 2𝑡} + 𝐿{cos 3𝑡} + 𝐿{sinh 3𝑡} − 2. 𝐿{cosh 4𝑡} + 𝐿{9}
1 2 2 𝑠 3 2𝑠 1
= − + 2 2
+ 2 2
+ 2 2
− 2 2
+ 9.
𝑠−3 𝑠+2 𝑠 +2 𝑠 +3 𝑠 −3 𝑠 −4 𝑠
𝟏 𝟐 𝟐 𝒔 𝟑 𝟐𝒔 𝟗
= − + + + − +
𝒔 − 𝟑 𝒔 + 𝟐 𝒔𝟐 + 𝟒 𝒔𝟐 + 𝟗 𝒔𝟐 − 𝟗 𝒔𝟐 − 𝟏𝟔 𝒔

3. Find the Laplace transform of (𝒕𝟐 + 𝟏)𝟐

Sol:
4! 1 2!
𝐿{(𝒕𝟐 + 𝟏)𝟐 } = 𝐿{𝑡 4 + 1 + 2𝑡 2 } = 𝐿{𝑡 4 } + 𝐿{1} + 2. 𝐿{𝑡 2 } = + + 2. 2+1
𝑠 4+1 𝑠 𝑠

𝟐𝟐𝟒 𝟏 𝟒
=> 𝑳 {(𝒕𝟐 + 𝟏) } = + +.
𝒔𝟓 𝒔 𝒔𝟑
4. Find 𝑳{𝒔𝒊𝒏 (𝒘𝒕 + 𝜶)}
Sol:
𝑳{𝒔𝒊𝒏 (𝒘𝒕 + 𝜶)} = 𝑳{𝒔𝒊𝒏 𝒘𝒕 𝒄𝒐𝒔 𝜶 + 𝒄𝒐𝒔 𝒘𝒕 𝒔𝒊𝒏 𝜶}

= 𝐜𝐨 𝐬 𝜶 𝑳{𝒔𝒊𝒏 𝒘𝒕} + 𝐬𝐢 𝐧 𝜶 𝑳{𝒄𝒐𝒔 𝒘𝒕}

𝒘 𝒔
= 𝒄𝒐𝒔 𝜶 . + 𝒔𝒊𝒏 𝜶.
𝒔𝟐+𝒘𝟐 𝒔𝟐+𝒘𝟐

∴ 𝑳{𝒔𝒊𝒏 (𝒘𝒕 + 𝜶)} = 𝒔 +𝒘

𝟏
𝟐 𝟐
{𝒘 𝒄𝒐𝒔 𝜶 + 𝒔 𝒔𝒊𝒏𝜶}

4
Laplace transforms

5. Find 𝑳{𝒄𝒐𝒔𝒉𝟐 𝟐𝒕}

Sol:
𝟐
𝟑
𝒆𝟐𝒕 + 𝒆−𝟐𝒕 𝟏
{ }
𝑳 𝒄𝒐𝒔𝒉 𝟐𝒕 = 𝑳 {[ ] } = 𝑳{𝒆𝟒𝒕 + 𝒆−𝟒𝒕 + 𝟐}
𝟐 𝟒
𝟏
= 𝟒 [𝑳{𝒆𝟒𝒕 } + 𝑳{𝒆−𝟒𝒕 } + 𝑳{𝟐}]

𝟏 𝟏 𝟏 𝟐
∴ 𝑳{𝒄𝒐𝒔𝒉𝟐 𝟐𝒕} = { + + }
𝟒 𝒔−𝟒 𝒔+𝟒 𝒔

𝟏 𝟑
6. Determine the Laplace transform of (√𝒕 − )
√𝒕
Sol:
𝟑
𝟏 3 3 1 1
𝐿 {(√𝒕 − ) } = 𝐿 {𝑡 2 − 𝑡 −2 − 3. 𝑡 2 + 3𝑡 −2 }
√𝒕
3 3 1 1
= 𝐿 {𝑡 2 } − 𝐿 {𝑡 −2 } − 3. 𝐿 {𝑡 2 } + 3. 𝐿 {𝑡 −2 }

𝟑 𝟑 𝟏 𝟏
𝜞(𝟐+𝟏) 𝜞(−𝟐+𝟏) 𝜞(𝟐+𝟏) 𝜞(−𝟐+𝟏)
= 3
+1
− 3
− +1
− 3. 1
+1
+ 3. 1
− +1
{ ∵ 𝑳{𝒕𝒏} = 𝜞(𝒏+𝟏)
𝒔𝒏+𝟏
}
𝑠2 𝑠 2 𝑠2 𝑠 2
𝟏 𝟏
𝜞(− ) 𝜞( )
=
1 𝟑
5 . 𝜞( )−
𝟐
𝟑
𝟐 −
𝟐
1 −
3 𝟏
3
𝟐
𝟏
𝜞 ( ) + 𝟑.
𝟐
𝟐
1 { ∵ 𝜞(𝒏 + 𝟏) = 𝒏𝜞(𝒏) }
𝑠2 𝑠 2 𝑠2 𝑠2
1
= 5
3 𝟏
. 𝜞 (𝟐 + 𝟏) − 𝟐√𝝅 𝑠 2 − 3
3
√𝜋 + 𝟑.
√𝝅
1 { ∵ 𝜞 (𝟏𝟐) = √𝝅 }
2 𝑠2 2 𝑠2 𝑠2
1
= 5
3 𝟏
𝟐
𝟏
. 𝜞 ( ) − 𝟐√𝝅 𝑠 2 −
𝟐
3√𝜋
3 +
3√𝝅
1 { ∵ 𝜞 (− 𝟏𝟐) = −𝟐√𝝅 }
2 𝑠2 2 𝑠2 𝑠2

𝟏 𝟑 𝟑√𝝅 𝟏
𝟑√𝝅 𝟑√𝝅
∴ 𝑳 {(√𝒕 − ) }= 𝟓 − 𝟐√𝝅 𝒔 −
𝟐
𝟑 + 𝟏
√𝒕 𝟒 𝒔𝟐 𝟐 𝒔𝟐 𝒔𝟐

Practice Problems:
1. Find the Laplace transform of the following.
𝟏 𝟑 𝒔𝒊𝒏 𝒕, 𝟎 < 𝒕 < 𝝅
i. 𝒔𝒊𝒏𝒉𝟑 𝟐𝒕 ii. (√𝒕 + ) iii. 𝒇(𝒕) = {
√𝒕 𝟎, 𝒕 > 𝝅
1 1
2. Show that 𝐿 { }=
√𝜋𝑡 √𝑠

Answers:
𝟏 𝟏 𝟏 𝟑 𝟑
i. { − − + }
𝒔 𝒔−𝟖 𝒔+𝟖 𝒔−𝟐 𝒔+𝟐

𝟑 √𝝅 𝟏 𝟑 √𝝅 𝟑 √𝝅 𝟏+𝒆−𝝅𝒔
ii. 𝟓 − 𝟐√𝝅 𝒔 + + iii.
𝟐
𝟑 𝟏
𝟒 𝒔𝟐 𝟐 𝒔𝟐 𝒔𝟐 𝒔𝟐 +𝟏

5
Laplace transforms

4.4 First shifting theorem: If 𝐿{𝑓(𝑡)} = 𝑓 ̅ (𝑠) then

i. 𝐿{𝑒 𝑎𝑡 𝑓(𝑡)} = 𝑓 ̅ (𝑠 − 𝑎) = [𝐿{𝑓(𝑡)}]𝑠→𝑠−𝑎
ii. 𝐿{𝑒 −𝑎𝑡 𝑓(𝑡)} = 𝑓 ̅ (𝑠 + 𝑎) = [𝐿{𝑓(𝑡)}]𝑠→𝑠+𝑎
Proof:
∞
Since by definition, 𝑓(̅ 𝑠) = 𝐿{𝑓(𝑡)} = ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡

Then
∞ ∞
i. ̅ (𝑠 − 𝑎)
𝐿{𝑒 𝑎𝑡 𝑓(𝑡)} = ∫0 𝑒−𝑠𝑡 𝑒 𝑎𝑡 𝑓(𝑡) 𝑑𝑡 = ∫0 𝑒 −(𝑠−𝑎)𝑡 𝑓(𝑡) 𝑑𝑡 = 𝑓
∞ ∞
ii. ̅ (𝑠 + 𝑎)
𝐿{𝑒 −𝑎𝑡 𝑓(𝑡)} = ∫0 𝑒−𝑠𝑡 𝑒 −𝑎𝑡 𝑓(𝑡) 𝑑𝑡 = ∫0 𝑒 −(𝑠+𝑎)𝑡 𝑓(𝑡) 𝑑𝑡 = 𝑓

Laplace transforms based on first shifting theorem:

𝑎 𝑎
1. 𝐿{𝑒 𝑎𝑡 sin 𝑎𝑡 } = [𝐿{sin 𝑎𝑡 }]𝑠→𝑠−𝑎 = { } =
𝑠 2 +𝑎2 𝑠 →𝑠−𝑎 (𝑠−𝑎)2 +𝑎2
𝑠−𝑎 𝑠−𝑎
2. 𝐿{𝑒 −𝑎𝑡 cosh 𝑎𝑡 } = [𝐿 {cosh 𝑎𝑡 }]𝑠→𝑠−𝑎 = { } =
𝑠 2 +𝑎2 𝑠 →𝑠−𝑎 (𝑠−𝑎)2 +𝑎2

Similarly, we will get the following.

S No. Function Laplace transform

𝑎
1 𝑒 𝑎𝑡 sin 𝑎𝑡 𝐿{𝑒𝑎𝑡 sin 𝑎𝑡} = 2
(𝑠 − 𝑎) + 𝑎2
𝑠−𝑎
2 𝑒 𝑎𝑡 cos 𝑎𝑡 𝐿{𝑒𝑎𝑡 cos 𝑎𝑡} = 2
(𝑠 − 𝑎) + 𝑎2
𝑎
3 𝑒 −𝑎𝑡 sin 𝑎𝑡 𝐿{𝑒−𝑎𝑡 sin 𝑎𝑡} = 2
(𝑠 + 𝑎) + 𝑎2
4 𝑒 −𝑎𝑡 cos 𝑎𝑡 𝑠+𝑎
𝐿{𝑒−𝑎𝑡 cos 𝑎𝑡} = 2
(𝑠 + 𝑎) + 𝑎2
5 𝑒 𝑎𝑡 sinh 𝑏𝑡 𝐿{𝑒𝑎𝑡 sinh 𝑏𝑡} =
𝑏
2 2
(𝑠 − 𝑎) − 𝑏
𝑠−𝑎
6 𝑒 𝑎𝑡 cosh 𝑏𝑡 𝐿{𝑒𝑎𝑡 cosh 𝑏𝑡} = 2 2
(𝑠 − 𝑎) − 𝑏
7 𝑒 −𝑎𝑡 sinh 𝑏𝑡 𝐿{𝑒−𝑎𝑡 sinh 𝑏𝑡} =
𝑏
2 2
(𝑠 + 𝑎) − 𝑏
8 𝑒 −𝑎𝑡 cosh 𝑏𝑡 𝑠+𝑎
𝐿{𝑒−𝑎𝑡 cosh 𝑏𝑡} = 2 2
(𝑠 + 𝑎) − 𝑏
9 𝑒 𝑎𝑡 𝑡 𝑛 𝐿{𝑒 𝑎𝑡 𝑡 𝑛 } =
𝑛!
(𝑠 − 𝑎)𝑛+1
10 𝑒 −𝑎𝑡 𝑡 𝑛 𝐿{𝑒 −𝑎𝑡 𝑡 𝑛 } =
𝑛!
(𝑠 + 𝑎)𝑛+1
11 𝑒 𝑎𝑡 sin 𝑏𝑡 𝐿{𝑒𝑎𝑡 sin 𝑏𝑡} =
𝑏
2 2
(𝑠 − 𝑎) + 𝑏

6
Laplace transforms

Problems:
Find the Laplace transform of the following.
1. 𝑒 −3𝑡 (2 cos 5𝑡 − 3 sin 5𝑡) 2. 𝑒 −𝑡 𝑐𝑜𝑠 2 𝑡
3. (𝑡 + 3)2 . 𝑒 𝑡 4. 𝑒 4𝑡 sin 2𝑡 𝑐𝑜𝑠 𝑡

Sol:
1. 𝐿{𝑒 −3𝑡 (2 cos 5𝑡 − 3 sin 5𝑡)} = 2. 𝐿{𝑒 −3𝑡 . cos 5𝑡} − 3. 𝐿{𝑒 −3𝑡 . sin 5𝑡}
= 2. [𝐿{cos 5𝑡}]𝑠→𝑠+3 − 3. [𝐿{sin 5𝑡}]𝑠→𝑠+3
𝑠 5
= 2. {𝑠2+25 } − 3. {𝑠2+25 }
𝑠→𝑠+3 𝑠→𝑠+3

𝑠+3 5
= 2. { } − 3. { }
(𝑠+3)2 +25 (𝑠+3)2+25
2𝑠+6 15
= {𝑠2+6𝑠+9+25 } − {𝑠2+6𝑠+9+25 }

2𝑠+6−15
={ }
𝑠 2 +6𝑠+34
𝟐𝒔−𝟗
∴ 𝑳{𝒆−𝟑𝒕 (𝟐 𝐜𝐨𝐬 𝟓𝒕 − 𝟑 𝐬𝐢𝐧 𝟓𝒕)} = {𝒔𝟐 +𝟔𝒔+𝟑𝟒}

1+cos 2𝑡 1 1 𝑠
2. 𝐿{𝑒 −𝑡 . 𝑐𝑜𝑠 2 𝑡} = [𝐿{𝑐𝑜𝑠 2 𝑡}]𝑠→𝑠+1 = [𝐿 { }] = 2 { 𝑠 + 𝑠 2+4}
2 𝑠→𝑠+1 𝑠→𝑠+1
𝟏 𝟏 𝒔+𝟏
∴ 𝑳{𝒆−𝒕 . 𝒄𝒐𝒔𝟐 𝒕} = { + 𝟐 }
𝟐 𝒔 + 𝟏 𝒔 + 𝟐𝒔 + 𝟓

3. 𝐿{(𝑡 + 3)2 𝑒 𝑡 } = [𝐿{(𝑡 + 3)2 }]𝑠→𝑠−1 = [𝐿{𝑡 2 + 6𝑡 + 9}]𝑠→𝑠−1

𝟐! 𝟏 𝟏
={ + 𝟔. + 𝟗. }
𝒔𝟐+𝟏 𝒔𝟏+𝟏 𝒔 𝒔→𝒔−𝟏

𝟐 𝟔 𝟗
∴ 𝑳{(𝒕 + 𝟑)𝟐 𝒆𝒕 } = + +
(𝒔−𝟏)𝟑 (𝒔−𝟏)𝟐 𝒔−𝟏

4. 𝑳{𝑒 4𝑡 sin 2𝑡 cos 𝑡} = 𝑳{𝒔𝒊𝒏 𝟐𝒕. 𝒄𝒐𝒔 𝒕}𝒔→𝒔−𝟒

𝟏
{∵ 𝒔𝒊𝒏𝒂. 𝒄𝒐𝒔𝒃 = [𝒔𝒊𝒏(𝒂 + 𝒃) + 𝒔𝒊𝒏(𝒂 − 𝒃)]}
𝟐
𝒔𝒊𝒏 𝟑𝒕+𝒔𝒊𝒏 𝒕
∴ 𝑳{𝑒 4𝑡 sin 2𝑡 cos 𝑡} = 𝑳 { }
𝟐 𝒔→𝒔−𝟒

𝟏 𝟑 𝟏
= { 𝟐 + 𝟐 }
𝟐 𝒔 + 𝟗 𝒔 + 𝟏 𝒔→𝒔−𝟒
𝟏 𝟑 𝟏
= { + }
𝟐 (𝒔 − 𝟒)𝟐 + 𝟗 (𝒔 − 𝟒)𝟐 + 𝟏

𝟏 𝟑 𝟏
𝑯𝒆𝒏𝒄𝒆, 𝑳{𝒆𝟒𝒕 𝐬𝐢𝐧 𝟐𝒕 𝐜𝐨𝐬 𝒕} = { 𝟐 + 𝟐 }
𝟐 𝒔 − 𝟖𝒔 + 𝟐𝟓 𝒔 − 𝟖𝒔 + 𝟏𝟕

7
Laplace transforms

Practice problems:
Find the Laplace transform of the following.
1
1. 𝐿 {𝑒 𝑡 (𝑐𝑜𝑠 2𝑡 + 2 𝑠𝑖𝑛ℎ 2𝑡)} 2. 𝐿{𝑠𝑖𝑛ℎ 𝑎𝑡 𝑠𝑖𝑛𝑎𝑡}

Answers:
𝒔−𝟏 𝟏 𝟐𝒂𝟐 𝒔
1. + 2.
𝒔𝟐 −𝟐𝒔+𝟓 𝒔𝟐 −𝟐𝒔−𝟑 𝒔𝟒 +𝟒𝒂𝟒

Unit step function:

The unit step function can also be called as Heaviside Unit function. It can be denoted by
𝑢(𝑡 − 𝑎) or 𝐻(𝑡 − 𝑎) and the unit step function at ‘a’ is defined as
𝟎, 𝒊𝒇 𝒕 < 𝒂
𝒖(𝒕 − 𝒂) 𝒐𝒓 𝑯(𝒕 − 𝒂) = {
𝟏, 𝒊𝒇 𝒕 > 𝒂
𝒆−𝒂𝒔
Laplace transform of Unit step function: 𝑳{𝒖(𝒕 − 𝒂)} =
𝒔
∞ 𝑎 ∞ ∞
−𝑠𝑡 −𝑠𝑡 −𝑠𝑡
𝑒 −𝑠𝑡 𝒆−𝒂𝒔
𝐿{𝑢(𝑡 − 𝑎)} = ∫ 𝑒 𝑢(𝑡 − 𝑎)𝑑𝑡 = ∫ 𝑒 .0 𝑑𝑡 + ∫ 𝑒 .1𝑑𝑡 = { } =
0 0 𝑎 −𝑠 𝑎 𝒔

Based on Laplace transform of unit step function, we will observe the Second shifting
theorem of Laplace transform, which discussed below.
4.5 Second shifting theorem of Laplace transform:
 First form of Second shifting theorem:
𝑓(𝑡 − 𝑎), 𝑖𝑓 𝑡 > 𝑎
If 𝐿{𝑓(𝑡)} = 𝑓 ̅ (𝑠)and 𝑔(𝑡) = { then 𝑳{𝒈(𝒕)} = 𝒆−𝒂𝒔 𝒇 ̅ (𝒔)
0, 𝑖𝑓 𝑡 < 𝑎
 Second form of Second shifting theorem:
If 𝐿{𝐹(𝑡)} = 𝑓 ̅ (𝑠) and 𝑎 > 0 then 𝑳{𝑭(𝒕 − 𝒂)𝑯(𝒕 − 𝒂)} = 𝒆−𝒂𝒔 𝒇 ̅ (𝒔)
0, 𝑖𝑓 𝑡 < 0
Where 𝐻 (𝑡) = { is Heaviside unit step function at ‘0’.
1, 𝑖𝑓 𝑡 > 0

Problems:
𝝅 𝝅
𝒄𝒐𝒔 (𝒕 − 𝟑 ) , 𝒊𝒇 𝒕 > 𝟑
1. Find 𝑳{𝒈(𝒕)}, where 𝒈(𝒕) = { 𝝅
𝟎, 𝒊𝒇 𝒕 <
𝟑
Sol:
𝑓 (𝑡 − 𝑎), 𝑖𝑓 𝑡 > 𝑎 𝜋
Give 𝑔(𝑡) is of the form 𝑔(𝑡) = { where 𝑓 (𝑡) = cos 𝑡, 𝑎 = 3
0, 𝑖𝑓 𝑡 < 𝑎
Then by First form of Second shifting theorem, we have
𝑳{𝒈(𝒕)} = 𝒆−𝒂𝒔 𝒇 ̅ (𝒔) --------------> (*)
𝑠
For, 𝑓 ̅ (𝑠) = 𝐿{𝑓(𝑡)} = 𝐿{𝑐𝑜𝑠 𝑡} = 2 𝑠 +1
𝝅𝒔
𝜋𝑠 −
−3 𝒔 𝒔𝒆 𝟑
Thus, from 𝒆𝒒𝒏 (∗), 𝑳{𝒈(𝒕)} = 𝒆 ( )=
𝒔𝟐 +𝟏 𝒔𝟐 +𝟏

8
Laplace transforms

2. Find 𝑳{(𝒕 − 𝟐)𝟑 𝒖(𝒕 − 𝟐)}

Sol:
Given function {(𝑡 − 2)3 𝑢(𝑡 − 2)} is of the form 𝐹 (𝑡 − 𝑎)𝑢(𝑡 − 𝑎),
with 𝑎 = 2, 𝐹(𝑡 − 𝑎) = (𝑡 − 2)3 => 𝐹(𝑡) = 𝑡 3
3! 6
and 𝑓 ̅ (𝑠) = 𝐿{𝐹(𝑡)} = 𝐿{𝑡 3 } = 3+1 = 4 𝑠 𝑠
then by second form of second shifting theorem, we have
𝑳{𝑭(𝒕 − 𝒂)𝑯(𝒕 − 𝒂)} = 𝒆−𝒂𝒔 𝒇̅ (𝒔)

𝟑
𝟔𝒆−𝟐𝒔
∴ 𝑳{(𝒕 − 𝟐) 𝒖(𝒕 − 𝟐)} =
𝒔𝟒
3. Find the Laplace transform of 𝒆−𝟑𝒕 𝒖(𝒕 − 𝟐)
Sol:
Consider 𝐿{𝑒 −3𝑡 𝑢(𝑡 − 2)} = 𝐿{𝑢(𝑡 − 2)}𝑠→𝑠+3 {∵ 𝑏𝑦 𝐹𝑖𝑟𝑠𝑡 𝑠ℎ𝑖𝑓𝑡𝑖𝑛𝑔 𝑡ℎ𝑚}

𝒆−𝟐𝒔 𝒆−𝒂𝒔
={ } {∵ 𝑳{𝒖(𝒕 − 𝒂)} = }
𝒔 𝑠→𝑠+3 𝒔

𝒆−𝟐(𝒔+𝟑)
= 𝒔+𝟑

−𝟑𝒕
𝒆−𝟐𝒔−𝟔
∴ 𝑳{𝒆 𝒖(𝒕 − 𝟐)} =
𝒔+𝟑
Practice problems:
Find the Laplace transform of the following functions.
2𝜋 2𝜋
𝑐𝑜𝑠 (𝑡 − ) , 𝑖𝑓 𝑡 >
3 3
1. 𝑓(𝑡) = { 2𝜋
0, 𝑖𝑓 𝑡 < 3
2. 𝑒 −2𝑡 {1 − 𝑢(𝑡 − 1)}
𝜋 𝜋
𝑠𝑖𝑛 (𝑡 − 3 ) , 𝑖𝑓 𝑡 > 3
3. 𝑓(𝑡) = { 𝜋
0, 𝑖𝑓 𝑡 <
3

Answers:
2𝜋𝑠 𝜋𝑠
− −
𝑠𝑒 3 1−𝑒 −(𝑠+2) 𝑒 3
1. 2. 3.
𝑠 2 +1 𝑠+2 𝑠 2 +1

9
Laplace transforms

4.6 Change of scale property of Laplace transform:

1 𝑠 𝑡
If 𝐿{𝑓(𝑡)} = 𝑓 ̅ (𝑠) then 𝐿{𝑓(𝑎𝑡)} = 𝑎 𝑓 ̅ (𝑎) and 𝐿 {𝑓 (𝑎)} = 𝑎. 𝑓 ̅ (𝑎𝑠)

Problems:
3
−
1 −
1
−𝑡 𝑒 𝑠+1
1. If 𝐿{𝑓(𝑡)} = 𝑠 𝑒 𝑠 then prove that 𝐿{𝑒 𝑓(3𝑡)} =
𝑠+1
Sol:
Since by first shifting theorem, 𝐿{𝑒 −𝑡 𝑓(3𝑡)} = 𝐿{𝑓(3𝑡)}𝑠→𝑠+1 --------> (*)
For, by change of scale property,
1 𝑠 1 3 3 1 3
𝐿{𝑓(3𝑡)} = 𝑓̅ ( ) = . . 𝑒− 𝑠 = . 𝑒− 𝑠
3 3 3 𝑠 𝑠
3
1 −
Now, from 𝒆𝒒𝒏 (∗), 𝐿{𝑒 −𝑡 𝑓(3𝑡)} = 𝐿{𝑓(3𝑡)}𝑠→𝑠+1 = { . 𝑒 𝑠 }
𝑠 𝑠→𝑠+1
𝟑
−
𝒆 𝒔+𝟏
=> 𝑳{𝒆−𝒕 𝒇(𝟑𝒕)} = 𝒔+𝟏

Hence proved.
Practice Problems:
9𝑠 2−12𝑠+15
1. If 𝐿{𝑓(𝑡)} = then find 𝐿{𝑓(3𝑡)} by using change of scale property.
(𝑠−1)3
2. Find 𝐿{𝑠𝑖𝑛2 𝑎𝑡} by using change of scale property.

Answers:
9(𝑠 2 −4𝑠+5) 2 𝑎2
1. 2.
(𝑠−3)3 𝑠(𝑠 2 +4𝑎2 )

10
Laplace transforms

4.7 Laplace transform of derivatives:

If 𝐿{𝑓(𝑡)} = 𝑓 ̅ (𝑠) then

𝐿{𝑓 𝑛 (𝑡)} = 𝑠 𝑛 𝑓 ̅ (𝑠) − 𝑠 𝑛−1 𝑓 (0) − 𝑠 𝑛−2 𝑓 ′ (0) − 𝑠 𝑛−3 𝑓 ′′ (0) − ⋯ − 𝑓 (𝑛−1) (0)
*Note: From the above result we may observe that
 𝑳{𝒇′(𝒕)} = 𝒔𝑓 ̅ (𝑠) − 𝑓(0)
 𝑳{𝒇′′(𝒕)} = 𝒔𝟐 𝑓 ̅ (𝑠) − 𝑠𝑓(0) − 𝑓′(0)
𝑳{𝒇′′′(𝒕)} = 𝒔𝟑 𝑓 ̅ (𝑠) − 𝒔𝟐 𝑓 (0) − 𝑠𝑓 ′ (0) − 𝑓′′(0) and so on...

Problems:
Using the theorem on transforms of derivatives, find the Laplace transforms of the following
functions.
1. 𝑒 𝑎𝑡 2. 𝑡 𝑐𝑜𝑠 𝑎𝑡

Sol:
1. Given 𝑓(𝑡) = 𝑒 𝑎𝑡 then 𝑓 ′ (𝑡) = 𝑎 𝑒 𝑎𝑡 then by Laplace transform of derivatives,

𝑳{𝒇′(𝒕)} = 𝒔𝑓 ̅ (𝑠) − 𝑓(0)

=> 𝑳{𝑎 𝑒 𝑎𝑡 } = 𝑠. 𝐿{𝑒 𝑎𝑡 } − 𝑒 0.𝑡
=> 𝑎𝐿{𝑒 𝑎𝑡 } = 𝑠. 𝐿{𝑒 𝑎𝑡 } − 1
=> 𝑠. 𝐿{𝑒 𝑎𝑡 } − 𝑎𝑎𝐿{𝑒 𝑎𝑡 } = 1
𝟏
=> (𝑠 − 𝑎)𝐿{𝑒 𝑎𝑡 } = 1 => 𝑳{𝒆𝒂𝒕 } =
𝒔−𝒂
2. Let 𝑓(𝑡) = 𝑡 cos 𝑎𝑡 then 𝑓 ′ (𝑡) = cos 𝑎𝑡 − 𝑎𝑡 sin 𝑎𝑡
And 𝑓 ′′ (𝑡) = −𝑎 sin 𝑎𝑡 − 𝑎 sin 𝑎𝑡 − 𝑎2 𝑡 cos 𝑎𝑡 = −2𝑎 sin 𝑎𝑡 − 𝑎2 𝑡 cos 𝑎𝑡
Then by Laplace transform of derivatives,

𝐿{𝑓′′(𝑡)} = 𝑠 2 𝑓 ̅ (𝑠) − 𝑠𝑓 (0) − 𝑓′(0)

=> 𝐿{−2𝑎 sin 𝑎𝑡 − 𝑎2 𝑡 cos 𝑎𝑡} = 𝑠 2 𝐿{𝑡 𝑐𝑜𝑠 𝑎𝑡} − 𝑠(0) − cos 0
=> −2𝑎 𝐿{sin 𝑎𝑡} − 𝑎2 𝐿{𝑡 cos 𝑎𝑡} = 𝑠 2 𝐿{𝑡 𝑐𝑜𝑠 𝑎𝑡} − 1
=> 𝑠 2 𝐿{𝑡 𝑐𝑜𝑠 𝑎𝑡} + 𝑎2 𝐿{𝑡 cos 𝑎𝑡} = 1 − 2𝑎 𝐿{sin 𝑎𝑡}
𝑎
=> 𝐿{𝑡 𝑐𝑜𝑠 𝑎𝑡} (𝑠 2 + 𝑎2 ) = 1 − 2𝑎. 2
𝑠 + 𝑎2
𝟏 𝟐 𝒂𝟐
Hence, 𝑳{𝒕 𝒄𝒐𝒔 𝒂𝒕} = {𝟏 − }
(𝒔𝟐 +𝒂𝟐 ) 𝒔𝟐 +𝒂𝟐

Practice Problems:
Using the theorem on transforms of derivatives, find the Laplace transforms of the following
functions. 1. 𝑡 sin 𝑎𝑡 2. cos 𝑎𝑡
2𝑎𝑠 𝑠
Answers: 1. 2.
(𝑠 2 +𝑎2 )2 𝑠 2 +𝑎2

11
Laplace transforms

4.8 Laplace transform of integrals: ̅

If 𝐿{𝑓 (𝑡)} = 𝑓 (𝑠) then
𝑡 𝑡 𝑡 𝑓̅(𝑠)
𝐿 {∫0 ∫0 … ∫0 𝑓 (𝑢)𝑑𝑢 𝑑𝑢 … 𝑑𝑢( 𝑛 𝑡𝑖𝑚𝑒𝑠)} = for n = 1, 2, …
𝑠𝑛

*Note: From the above result we may observe that

𝒕 ̅ (𝑠)
𝑓
 𝑳 {∫𝟎 𝒇(𝒖)𝒅𝒖} = 𝒔
𝒕 𝒕 ̅ (𝑠)
𝑓
 𝑳 {∫𝟎 ∫𝟎 𝒇(𝒖)𝒅𝒖 𝒅𝒖} = 𝒔𝟐 and so on…

Problems:
𝒕 𝒕
1. Find 𝑳 {∫𝟎 ∫𝟎 𝒔𝒊𝒏𝒉 𝒂𝒕 𝒅𝒕 𝒅𝒕}
Sol:
Let 𝑓(𝑡) = 𝑠𝑖𝑛ℎ 𝑎𝑡 then by Laplace transform of integrals, we have
𝑡 𝑡 ̅ (𝑠)
𝑓
𝐿 {∫ ∫ 𝑓 (𝑢)𝑑𝑢 𝑑𝑢} = 2
0 0 𝑠
𝒕 𝒕
𝟏 𝟏 𝒂 𝒂
=> 𝑳 {∫ ∫ 𝒔𝒊𝒏𝒉 𝒂𝒕 𝒅𝒕 𝒅𝒕} = 𝟐
𝑳{𝒔𝒊𝒏𝒉 𝒂𝒕} = 𝟐 ( 𝟐 𝟐
)= 𝟐 𝟐
𝟎 𝟎 𝒔 𝒔 𝒔 −𝒂 𝒔 (𝒔 − 𝒂𝟐 )

Practice problems:
Find the following
𝑡 𝑡
1. 𝐿 {∫0 ∫0 𝑐𝑜𝑠ℎ 𝑎𝑡 𝑑𝑡 𝑑𝑡}
𝑡
2. 𝐿 {∫0 𝑒 −𝑡 𝑐𝑜𝑠 𝑡 𝑑𝑡}

Answers:
𝑠 𝑠+1
1. 2.
𝑠(𝑠 2 −𝑎2 ) 𝑠(𝑠 2 +2𝑠+2)

12
Laplace transforms

4.9 Laplace transform of a function multiplied by 𝒕𝒏 :

𝒅 𝒏
̅
If 𝐿{𝑓(𝑡)} = 𝑓 (𝑠) ̅
then 𝑳{𝒕𝒏 𝒇(𝒕)} = (−𝟏)𝒏 𝒅 𝒔𝒏 {𝑓 (𝑠)} 𝒇𝒐𝒓 𝒏 = 𝟏, 𝟐, …

𝑑
̅
i.e., 𝐿{𝑡 𝑓(𝑡)} = − 𝑑𝑠 {𝑓 (𝑠)} ̅
or 𝐿{𝑡 𝑓(𝑡)} = −{𝑓′(𝑠)}

𝑑2
̅
𝐿{𝑡 2 𝑓(𝑡)} = 𝑑 𝑠 2 {𝑓 (𝑠)} ̅
= 𝑓 ′′(𝑠) and so on…

Problems:
Find the following. 1. 𝑳{𝒕𝟐 𝒔𝒊𝒏 𝟐𝒕} 2. 𝑳{𝒕 𝒆𝟐𝒕 𝒔𝒊𝒏 𝟑𝒕}
Sol:
𝑑2
̅
1. Let 𝑓(𝑡) = 𝑠𝑖𝑛 2𝑡 then we know that 𝐿{𝑡 2 𝑓(𝑡)} = 𝑑 𝑠 2 {𝑓 (𝑠)} ̅
= 𝑓′′(𝑠)---->(*)
2
For, 𝑓 (̅ 𝑠) = 𝐿{𝑓(𝑡)} = 𝐿{𝑠𝑖𝑛 2𝑡} = 𝑠 2+4

𝑑2 2
∴ 𝑓𝑟𝑜𝑚 𝑒𝑞 𝑛 (∗), 𝐿{𝑡 2 𝑠𝑖𝑛 2𝑡} = 2
{ 2 }
𝑑𝑠 𝑠 +4
𝑑 𝑑 1
=2 { [ ]}
𝑑𝑠 𝑑𝑠 𝑠 2 +4

𝑑 −2𝑠
= 2. 𝑑𝑠 {(𝑠 2+4)2 }

𝑑 𝑠
= (−4) 𝑑𝑠 {(𝑠 2+4)2 }
2
(𝑠 2 +4) −2𝑠(𝑠 2+4).2𝑠
= (−4) { (𝑠 2+4)4
}

4−3𝑠 2
= (−4). {(𝑠 2+4)3 }

𝟒(𝟑 𝒔𝟐 −𝟒)
∴ 𝑳{𝒕𝟐 𝒔𝒊𝒏 𝟐𝒕} = (𝒔𝟐 +𝟒)𝟑
𝒅
̅ (𝒔)} ----> (*)
2. Let 𝒇(𝒕) = 𝒆𝟐𝒕 𝒔𝒊𝒏 𝟑𝒕 then we know that 𝑳{𝒕 𝒇(𝒕)} = − 𝒅𝒔 {𝒇

𝟑 𝟑
̅ (𝒔) = 𝑳{𝒇(𝒕)} = 𝑳{𝒆𝟐𝒕 𝒔𝒊𝒏 𝟑𝒕} = 𝑳{𝒔𝒊𝒏 𝟑𝒕}𝒔 →𝒔−𝟐 =
For, 𝒇 = 𝒔𝟐 −𝟒𝒔+𝟏𝟑
((𝒔−𝟐)𝟐 +𝟗)

𝒅 𝟑
∴ 𝒇𝒓𝒐𝒎 𝒆𝒒𝒏 (∗), 𝑳{𝒕 𝒆𝟐𝒕 𝒔𝒊𝒏 𝟑𝒕} = − { 𝟐 }
𝒅𝒔 𝒔 − 𝟒𝒔 + 𝟏𝟑
𝒅
= (−𝟑). 𝒅𝒔 (𝒔𝟐 − 𝟒𝒔 + 𝟏𝟑)−𝟏

= (−𝟑). (−𝟏). (𝒔𝟐 − 𝟒𝒔 + 𝟏𝟑)−𝟐 (𝟐𝒔 − 𝟒)

𝟔(𝒔−𝟐)
∴ 𝑳{𝒕 𝒆𝟐𝒕 𝒔𝒊𝒏 𝟑𝒕} = (𝒔𝟐
−𝟒𝒔+𝟏𝟑)𝟐

13
Laplace transforms

4.10 Laplace transform of a function divided by 𝒕𝒏 :

𝑓(𝑡) ∞
If 𝑓(̅ 𝑠) = 𝐿{𝑓(𝑡)} 𝑡hen 𝐿 { } = ∫𝑠 𝑓(̅ 𝑠)𝑑𝑠
𝑡

𝑓(𝑡) ∞ ∞
𝐿{ 𝟐 } = ∫𝑠 ∫𝑠 𝑓(̅ 𝑠)𝑑𝑠 𝑑𝑠 and so on…
𝒕

Problems:
𝒆𝒂𝒕 −𝒆𝒃𝒕
1. Find 𝑳 { }
𝒕

𝑓(𝑡) ∞
Sol: Let 𝑓 (𝑡) = 𝑒 𝑎𝑡 − 𝑒 𝑏𝑡 then we have 𝐿 { } = ∫𝑠 𝑓(̅ 𝑠)𝑑𝑠 -------> (*)
𝑡
1 1
For, 𝑓 (̅ 𝑠) = 𝐿{𝑒 𝑎𝑡 − 𝑒 𝑏𝑡 } = 𝑠−𝑎 − 𝑠−𝑏

𝑒 𝑎𝑡 −𝑒 𝑏𝑡 ∞ 1 1
Now from 𝑒𝑞 𝑛 (∗), 𝐿 { } = ∫𝑠 { − 𝑠−𝑏 } 𝑑𝑠
𝑡 𝑠−𝑎

= {log(𝑠 − 𝑎) − log(𝑠 − 𝑏)}∞

𝑠

𝑠−𝑎 ∞
= {log (𝑠−𝑏 )}
𝑠

𝑎 ∞
𝑠[1− ]
𝑠
= {log ( 𝑏 )}
𝑠[1− ]
𝑠 𝑠

𝑎 ∞
[1− ]
𝑠
= {log ( 𝑏 )}
[1− ]
𝑠 𝑠
𝑎
[1− ]
𝑠
= log 1 − log ( 𝑏 )
[1− ]
𝑠

𝑠−𝑎
= − log {𝑠−𝑏 }

= −{log(𝑠 − 𝑎) − log(𝑠 − 𝑏)}

𝑒 𝑎𝑡 −𝑒𝑏𝑡 𝑠−𝑏
∴ 𝐿{ } = log { }
𝑡 𝑠−𝑎

Practice problems:
Find the Laplace transforms of the following.
sin 𝑡 𝑒 −𝑎𝑡 −𝑒−𝑏𝑡 cos 2𝑡−cos 3𝑡
1. 2. 3. 4. 𝑡 2 𝑒 −2𝑡
𝑡 𝑡 𝑡

Answers:
𝑠+𝑏 1 𝑠 2 +9 2
1. 𝑐𝑜𝑡 −1𝑠 2. log { } 3. 𝑙𝑜𝑔 { } 4.
𝑠+𝑎 2 𝑠 2 +4 (𝑠+2)3

14
Laplace transforms

4.11 Evaluation of Integrals by Laplace transforms:

The integral of a function 𝑓(𝑡) defined in (0, ∞) can be easily evaluated by the definition of
Laplace transform, which is illustrated in the following examples.
Problems:
∞ ∞ sin 2𝑡
Evaluate 1. ∫0 𝑡 𝑒 −3𝑡 𝑑𝑡 2. ∫0 𝑑𝑡
𝑡

Sol:
∞ ∞
1. Given ∫0 𝑡 𝑒 −3𝑡 𝑑𝑡 is of the form ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 with 𝑠 = 3 and 𝑓(𝑡) = 𝑡
∞
And since ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 = 𝐿{𝑓(𝑡)}
∞ 𝟏 𝟏 𝟏
Then ∫
𝟎
𝒕 𝒆−𝟑𝒕 𝒅𝒕 = 𝑳{𝒕} = = =
𝒔𝟐 𝟑𝟐 𝟗
∞ sin 2𝑡 ∞ sin 2𝑡
2. Given ∫0 𝑑𝑡 is of the form ∫0 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 with 𝑠 = 0 and 𝑓 (𝑡) =
𝑡 𝑡

∴ ∫0∞ 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡 = 𝐿{𝑓(𝑡)}

∞ sin 2𝑡 sin 2𝑡
=> ∫0
𝑡
𝑑𝑡 = 𝐿 { 𝑡
}
∞
= ∫𝑠 𝐿{sin 2𝑡} 𝑑𝑠
∞ 2
= ∫𝑠 𝑑𝑠
𝑠 2+22

1 𝑠 ∞
= 2. 2 {𝑡𝑎𝑛 −1 (2)}
𝑠

𝑠
= 𝑡𝑎𝑛−1 ∞ − 𝑡𝑎𝑛−1 (2)

∞ 𝐬𝐢𝐧 𝟐𝒕 𝝅 𝒔
∴ ∫𝟎 𝒅𝒕 = − 𝒕𝒂𝒏−𝟏 ( )
𝒕 𝟐 𝟐

Practice problems:
Evaluate the following integrals by using Laplace transforms.
∞
1. ∫0 𝑡 𝑒 −3𝑡 𝑠𝑖𝑛 𝑡 𝑑𝑡
∞ 𝑒−𝑎𝑡 −𝑒−𝑏𝑡
2. ∫0 𝑑𝑡
𝑡

Answers:
3 𝑏
1. 2. 𝑙𝑜𝑔 { }
50 𝑎

15
Laplace transforms

4.12 Laplace transforms of Special functions:

 4.12(a) Periodic function:
A function 𝑓(𝑥) is said to be a periodic function of a period 𝑇 > 0 if 𝑓 (𝑥 ) = 𝑓(𝑥 + 𝑇)

Ex: sin 𝑥, cos 𝑥 are periodic functions of periodicity 2𝜋 and tan 𝑥, cot 𝑥 are periodic
functions of periodicity 𝜋.
Note: 1. If 𝑓(𝑥) is a periodic function of a period 𝑇 > 0 then
𝑓 (𝑥 ) = 𝑓 (𝑥 + 𝑇) = 𝑓 (𝑥 + 2𝑇) = 𝑓 (𝑥 + 3𝑇) = ⋯ and the least periodicity 𝑇 is called
as Primitive period.
2. If 𝑇 is a periodicity of a function 𝑓(𝑥 ) then the periodicity of the function
𝑇
𝑓 (𝑎𝑥 + 𝑏) is 𝑎.
 Laplace transform of a periodic function:
The Laplace transform of a periodic function 𝑓(𝑡) having period 𝑇 > 0 is given by
𝑻
𝟏
𝑳{𝒇(𝒕)} = ∫ 𝒆−𝒔𝒕 𝒇(𝒕)𝒅𝒕
𝟏 − 𝒆−𝒔𝑻 𝟎

Problems:
1. Find the Laplace transform of the square wave function of period 𝟐𝒂 defined as
𝒌, 𝒘𝒉𝒆𝒏 𝟎 < 𝒕 < 𝒂
𝒇(𝒕) = {
−𝒌, 𝒘𝒉𝒆𝒏 𝒂 < 𝒕 < 𝟐𝒂
Sol:
Given 𝑓(𝑡) is a periodic function with period 𝑇 = 2𝑎 then by the Laplace transform of a
periodic function of period 𝑇 > 0, we have
𝑇
1
𝐿{𝑓(𝑡)} = ∫ 𝑒 −𝑠𝑡 𝑓 (𝑡)𝑑𝑡
1 − 𝑒 −𝑠𝑇 0
1 2𝑎
=
1−𝑒 −2𝑎𝑠
∫0 𝑒 −𝑠𝑡 𝑓(𝑡)𝑑𝑡
1 𝑎 2𝑎
= 1−𝑒−2𝑎𝑠 {∫0 𝑒−𝑠𝑡 𝑘𝑑𝑡 + ∫𝑎 𝑒−𝑠𝑡 (−𝑘)𝑑𝑡}

𝑎 2𝑎
1 𝑒−𝑠𝑡 𝑒−𝑠𝑡
= 1−𝑒−2𝑎𝑠 [𝑘 ( −𝑠 ) − 𝑘 ( −𝑠 ) ]
0 𝑎

1 𝑘
= 1−𝑒−2𝑎𝑠 . 𝑠 {−𝑒−𝑎𝑠 + 1 + 𝑒−2𝑎𝑠 − 𝑒−𝑎𝑠 }

1 𝑘
= 1−𝑒−2𝑎𝑠 . 𝑠 {1 + 𝑒−2𝑎𝑠 − 2𝑒−𝑎𝑠 }

1 𝑘
= 1−𝑒−2𝑎𝑠 . 𝑠 (1 − 𝑒−𝑎𝑠 )2
1 𝑘
= 1−(𝑒 −𝑎𝑠)2 . 𝑠 (1 − 𝑒−𝑎𝑠 )2

1 𝑘 𝒌(𝟏−𝒆−𝒂𝒔 )
= (1+𝑒 −𝑎𝑠)(1−𝑒 −𝑎𝑠 ) . 𝑠 (1 − 𝑒−𝑎𝑠 )2 =
𝒔(𝟏+𝒆−𝒂𝒔 )

16
Laplace transforms

2. Find 𝑳{|𝐬𝐢𝐧 𝒕|}

Sol: Since |sin 𝑡| is a periodic function with period 𝜋.

𝑇
1
∴ 𝐿{𝑓(𝑡)} = −𝑠𝑇
∫ 𝑒−𝑠𝑡 𝑓(𝑡)𝑑𝑡
1−𝑒 0
𝜋
1
=> 𝐿{𝑓(𝑡)} = −𝜋𝑠
∫ 𝑒−𝑠𝑡 |𝑠𝑖𝑛 𝑡|𝑑𝑡
1−𝑒 0
1 𝜋 −𝑠𝑡
= −𝜋𝑠 ∫0
𝑒 𝑠𝑖𝑛 𝑡𝑑𝑡 {∵ |𝑠𝑖𝑛 𝑡| = 𝑠𝑖𝑛 𝑡 𝑖𝑛 0 < 𝑡 < 𝜋}
1−𝑒
𝜋
1 𝑒 −𝑠𝑡
= [ (−𝑠 𝑠𝑖𝑛 𝑡 − 𝑐𝑜𝑠 𝑡)]
1−𝑒 −𝜋𝑠 1+𝑠 2 0
𝑒 𝑎𝑥
{∵ ∫(𝑒 𝑎𝑥 sin 𝑏𝑥 ) 𝑑𝑥 = (asin 𝑏𝑥 − 𝑏 cos 𝑏𝑥 ) }
𝑎 2 +𝑏2
1 𝑒 −𝜋𝑠 1
= [ (0 + 1) − (0 − 1)]
1−𝑒 −𝜋𝑠 1+𝑠 2 1+𝑠 2
𝟏+𝒆−𝝅𝒔
=
(𝟏+𝒔𝟐 )(𝟏−𝒆−𝝅𝒔 )
 4.12(b) Unit impulse or Dirac delta function:
The Dirac delta function can be used to deal the situations in which a very large force
acting for a very short time like earthquake, collision of two bodies, etc.
And the Dirac delta function is defined as 𝛿 (𝑡 − 𝑎) = lim 𝑓∈ (𝑡 − 𝑎)
𝜖→0
1
,𝑎 ≤ 𝑡 ≤ 𝑎 + 𝜖
Where 𝑓∈ (𝑡 − 𝑎) = {𝜖
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
 Laplace transform of Dirac delta function:
𝐿{𝛿 (𝑡 − 𝑎)} = 𝐿 {lim 𝑓∈ (𝑡 − 𝑎)} = lim 𝐿{𝑓∈ (𝑡 − 𝑎)}
𝜖→0 𝜖→0
∞ −𝑠𝑡
= lim{∫0 𝑒 𝑓∈ (𝑡 − 𝑎) 𝑑𝑡}
𝜖→0
𝑎 𝑎+𝜖 1 ∞
= lim {∫0 𝑒 −𝑠𝑡 (0)𝑑𝑡 + ∫𝑎 𝑒 −𝑠𝑡 ( ) 𝑑𝑡 + ∫𝑎+𝜖 𝑒 −𝑠𝑡 (0)𝑑𝑡}
𝜖→0 𝜖
−𝑠𝑡 𝑎+𝜖
1 𝑒
= lim { } { }
𝜖→0 𝜖 −𝑠 𝑎
1
= lim { { −𝑠𝑎−𝑠𝜖
} 𝑒 − 𝑒 −𝑎𝑠 }
𝜖→0 −𝑠𝜖
1
= lim { } 𝑒 −𝑎𝑠 {𝑒 −𝑠𝜖 − 1}
𝜖→0 −𝑠𝜖
1−𝑒 −𝑠𝜖
= 𝑒 −𝑎𝑠 lim { }
𝜖→0 𝑠𝜖
−𝑎𝑠 𝑠𝑒 −𝑠𝜖
=𝑒 lim { } {∵ 𝑏𝑦 𝐿 𝐻𝑜𝑠𝑝𝑖𝑡𝑎𝑙 𝑟𝑢𝑙𝑒}
𝑠𝜖
𝜖→0
=𝑒 −𝑎𝑠 lim 𝑒 −𝑠𝜖
𝜖→0
= 𝑒−𝑎𝑠 (1)
∴ 𝑳{𝛿 (𝑡 − 𝑎)} = 𝒆−𝒂𝒔

17
Laplace transforms

 4.12 (c) Unit step function:

The unit step function can also be called as Heaviside Unit function. It can be denoted
by 𝑢(𝑡 − 𝑎) or 𝐻(𝑡 − 𝑎) and the unit step function at ‘a’ is defined as
𝟎, 𝒊𝒇 𝒕 < 𝒂
𝒖(𝒕 − 𝒂) 𝒐𝒓 𝑯(𝒕 − 𝒂) = {
𝟏, 𝒊𝒇 𝒕 > 𝒂

 Laplace transform of Unit step function:

𝒆−𝒂𝒔
𝑳{𝒖(𝒕 − 𝒂)} =
𝒔
∞ 𝑎 ∞ ∞
−𝑠𝑡 −𝑠𝑡 −𝑠𝑡
𝑒 −𝑠𝑡 𝒆−𝒂𝒔
𝐿{𝑢(𝑡 − 𝑎)} = ∫ 𝑒 𝑢(𝑡 − 𝑎)𝑑𝑡 = ∫ 𝑒 .0 𝑑𝑡 + ∫ 𝑒 .1𝑑𝑡 = { } =
0 0 𝑎 −𝑠 𝑎 𝒔

Problems:
1. Find 𝑳{𝒆𝒕−𝟑 𝒖(𝒕 − 𝟑)}

Sol:
𝐿{𝑒 𝑡−3 𝑢(𝑡 − 3)} = 𝑒 −3 𝐿{𝑒 𝑡 𝑢(𝑡 − 3)}
= 𝑒 −3 𝐿{𝑢(𝑡 − 3)}𝑠→𝑠−1 {∵ 𝑏𝑦 𝐹𝑖𝑟𝑠𝑡 𝑠ℎ𝑖𝑓𝑡𝑖𝑛𝑔 𝑡ℎ𝑒𝑜𝑟𝑒𝑚}

𝑒−3𝑠 𝑒 −𝑎𝑠
= 𝑒 −3 { 𝑠 )} {∵ 𝐿 {𝑢(𝑡 − 𝑎)} = 𝑎𝑛𝑑 𝑎 = 3}
𝑠→𝑠−1 𝑠

𝑒−3(𝑠−1)
= 𝑒 −3 { 𝑠−1 }

𝑒 −3𝑒 −3𝑠 𝑒 3
= 𝑠−1

𝒕−𝟑
𝒆−𝟑𝒔
∴ 𝑳{𝒆 𝒖(𝒕 − 𝟑)} =
𝒔−𝟏

18
Laplace transforms

Inverse Laplace transforms

Inverse Laplace transform (Definition):

If 𝐿{𝑓(𝑡)} = 𝑓 ̅(𝑠) then 𝑓(𝑡) = 𝐿−1 {𝑓(̅ 𝑠)} is called the inverse Laplace transform of 𝑓 ̅(𝑠) and
𝐿−1 known as Inverse Laplace transform operator.

4.13 Inverse Laplace transform of fundamental functions:

S No. 𝑳−𝟏 { 𝒇̅(𝒔)} = 𝒇(𝒕)

1 𝒌
𝐿−1 { } = 𝒌, 𝒘𝒉𝒆𝒓𝒆 𝒌 𝒊𝒔 𝒂 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕.
𝒔

2 𝟏
𝐿−1 { 𝟐 } = 𝑡
𝒔

3 𝒏! 𝟏 𝒕𝒏
𝐿−1 {𝒔𝒏+𝟏 } = 𝒕𝒏 or 𝐿−1 {𝒔𝒏+𝟏 } = 𝒏! when 𝒏 > 𝟎

4 𝟏 𝒕𝒏
𝐿−1 { } = 𝒘𝒉𝒆𝒏 𝒏 > −𝟏 & 𝒏 ∉ 𝑍+
𝒔𝒏+𝟏 𝜞(𝒏 + 𝟏)

5 𝟏
𝐿−1 {𝒔+𝒂} = 𝒆−𝒂𝒕

6 𝟏
𝐿−1 { } = 𝒆𝒂𝒕
𝒔−𝒂

7 𝒂 𝟏 𝒔𝒊𝒏 𝒂𝒕
𝐿−1 {𝒔𝟐+𝒂𝟐 } = 𝒔𝒊𝒏 𝒂𝒕 or 𝐿−1 {𝒔𝟐+𝒂𝟐 } = 𝒂

8 𝒔
𝐿−1 { } = 𝒄𝒐𝒔 𝒂𝒕
𝒔𝟐 + 𝒂𝟐

9 𝒂 𝟏 𝒔𝒊𝒏𝒉 𝒂𝒕
𝐿−1 {𝒔𝟐−𝒂𝟐 } = 𝒔𝒊𝒏𝒉 𝒂𝒕 or 𝐿−1 {𝒔𝟐−𝒂𝟐 } = 𝒂

10 𝒔
𝐿−1 { } = 𝒄𝒐𝒔𝒉 𝒂𝒕
𝒔𝟐 − 𝒂𝟐

19
Laplace transforms

4.14 Linear property of Inverse Laplace transform:

If 𝑳−𝟏 { 𝒇̅(𝒔)} = 𝒇(𝒕)and 𝑳−𝟏 { 𝒈
̅ (𝒔)} = 𝒈(𝒕) then for any constants 𝒄𝟏 and 𝒄𝟐 ,
𝑳−𝟏 { 𝒄𝟏 𝒇̅(𝒔) + 𝒄𝟐 𝒈
̅ (𝒔)} = 𝒄𝟏 𝑳−𝟏 { 𝒇̅(𝒔)} + 𝒄𝟐 𝑳−𝟏 { 𝒈
̅ (𝒔)} = 𝒄𝟏 𝒇(𝒕) + 𝒄𝟐 𝒈(𝒕).

Problems:
Find the Inverse Laplace transform of the following.
𝒔𝟐−𝟑𝒔+𝟒
1. 𝑳−𝟏 { }
𝒔𝟑

𝟐𝒔+𝟏
2. 𝑳−𝟏 { }
𝒔(𝒔+𝟏)

Sol:
𝑠 2−3𝑠+4 𝑠2 𝑠 1
1. Consider 𝐿−1 { 𝑠3
} = 𝐿−1 { 3 } − 3𝐿−1 { 3 } + 4𝐿−1 { 3 }
𝑠 𝑠 𝑠

1 1 1
= 𝐿−1 { 𝑠 } − 3𝐿−1 {𝑠2 } + 4𝐿−1 {𝑠3 }

1 2!
= 1 − 3𝑡 + 4𝐿−1 {𝑠2+1 ∗ 2!}

4𝑡 2
= 1 − 3𝑡 +
2!
𝒔𝟐 −𝟑𝒔+𝟒
∴ 𝑳−𝟏 { } = 𝟏 − 𝟑𝒕 + 𝟐𝒕𝟐
𝒔𝟑

2𝑠+1 𝑠+𝑠+1 𝑠 𝑠+1 1 1

2. Consider = = + = +
𝑠(𝑠+1) 𝑠(𝑠+1) 𝑠(𝑠+1) 𝑠(𝑠+1) (𝑠+1) 𝑠

𝟐𝒔 + 𝟏 1 1
=> 𝑳−𝟏 { } = 𝐿−1 { + }
𝒔(𝒔 + 𝟏) (𝑠 + 1) 𝑠
1 1
= 𝐿−1 {(𝑠+1)} + 𝐿−1 { 𝑠 }

= 𝒆−𝒂𝒕 + 𝟏

20
Laplace transforms

4.15 Inverse Laplace transform by partial fractions:

𝑔̅(𝑠)
If 𝑓 ̅(𝑠) = ℎ̅(𝑠), where 𝑔̅ (𝑠) and ℎ̅ (𝑠) are the polynomials of s then the inverse Laplace transform
𝑔̅(𝑠)
of 𝑓(̅ 𝑠) can be find by decomposing ℎ̅(𝑠) into partial fractions. This can be illustrated in the
following examples.
Problems:
Find the Inverse Laplace transform of the following by using partial fractions.
𝒔𝟐+𝒔−𝟐 𝟏
1. 𝑳−𝟏 {𝒔(𝒔+𝟑)(𝒔−𝟐)} 2. 𝑳−𝟏 {(𝒔+𝟏)(𝒔𝟐+𝟏)}

Sol:
𝒔𝟐+𝒔−𝟐 𝒔𝟐+𝒔−𝟐
1. To find 𝑳−𝟏 {𝒔(𝒔+𝟑)(𝒔−𝟐)}, first we resolve 𝒔(𝒔+𝟑)(𝒔−𝟐) into partial fractions as following

𝒔𝟐 + 𝒔 − 𝟐 𝑨 𝑩 𝑪
= + + − − − −→ (∗)
𝒔(𝒔 + 𝟑)(𝒔 − 𝟐) 𝒔 𝒔 + 𝟑 𝒔 − 𝟐

=> 𝒔𝟐 + 𝒔 − 𝟐 = 𝑨(𝒔 + 𝟑)(𝒔 − 𝟐) + 𝑩(𝒔)(𝒔 − 𝟐) + 𝑪(𝒔)(𝒔 + 𝟑) − −−→ (𝟏)

𝑁𝑜𝑤, 𝑡𝑜 𝑓𝑖𝑛𝑑 𝐴, 𝐵 𝑎𝑛𝑑 𝐶, 𝑝𝑢𝑡 𝑠 = 0 𝑖𝑛 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (1),

−6𝐴 = −2 => 𝑨 = 𝟏/𝟑

𝑛𝑒𝑥𝑡 𝑝𝑢𝑡 𝑠 = −3 𝑖𝑛 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (1), 15𝐵𝑠 = 4 => 𝑩 = 𝟒/𝟏𝟓
𝑛𝑜𝑤 𝑎𝑔𝑎𝑖𝑛, 𝑝𝑢𝑡 𝑠 = 2 𝑖𝑛 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (1), 10𝑐𝑠 = 4 => 𝑪 = 𝟐/𝟓
Thus, from equation (*),

𝒔𝟐 +𝒔−𝟐 𝑨 𝑩 𝑪 𝟏 𝟒 𝟐
= + + = + +
𝒔(𝒔+𝟑)(𝒔−𝟐) 𝒔 𝒔+𝟑 𝒔−𝟐 𝟑𝒔 𝟏𝟓(𝒔+𝟑) 𝟓(𝒔−𝟐)

𝒔𝟐 +𝒔−𝟐 1 4 2
=> 𝑳−𝟏 { } = 𝐿−1 { + + }
𝒔(𝒔+𝟑)(𝒔−𝟐) 3𝑠 15(𝑠+3) 5(𝑠−2)

1 1 4 1 2 1
= 𝐿−1 { } + 𝐿−1 { } + 𝐿−1 { }
3 𝑠 15 𝑠+3 5 𝑠−2
𝟏 𝟒 𝟐
= + 𝒆−𝟑𝒕 + 𝒆𝟐𝒕
𝟑 𝟏𝟓 𝟓

𝟏 𝟏
2. To find 𝑳−𝟏 {(𝒔+𝟏)(𝒔𝟐+𝟏)}, first we decompose (𝒔+𝟏)(𝒔𝟐+𝟏) into partial fractions.
𝟏 𝑨 𝑩𝒔 + 𝑪
= + 𝟐 − −−→ (∗)
(𝒔 + 𝟏)(𝒔 + 𝟏) 𝒔 + 𝟏
𝟐 𝒔 +𝟏

=> 1 = 𝐴(𝒔𝟐 + 𝟏) + (𝑩𝒔 + 𝑪)(𝒔 + 𝟏)

=> 1 = 𝐴𝑠 2 + 𝐴 + 𝐵𝑠 2 + 𝐵𝑠 + 𝐶𝑠 + 𝐶
𝑂𝑛 𝐶𝑜𝑚𝑝𝑎𝑟𝑖𝑠𝑜𝑛,

21
Laplace transforms

𝐴 + 𝐵 = 0 − −−→ (1)
𝐵 + 𝐶 = 0 − −−→ (2)
𝐴 + 𝐶 = 1 => 𝐶 = 1 − 𝐴 − −−→ (3)
1 1 1
By solving these equations, we get 𝐴 = 2 , 𝐵 = − 2 𝑎𝑛𝑑 𝐶 = 2
𝑛𝑜𝑤, 𝑓𝑟𝑜𝑚 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (∗),
(− )𝒔 + 𝟏 𝟏
𝟏 𝟏
𝑳−𝟏 { −𝟏
}=𝑳 { + 𝟐𝟐 𝟐}
(𝒔 + 𝟏)(𝒔𝟐 + 𝟏) 𝟐(𝒔 + 𝟏) 𝒔 + 𝟏

𝟏 𝟏 𝟏 𝒔 𝟏 𝟏
= 𝑳−𝟏 { } − 𝑳−𝟏 { } + 𝑳−𝟏 { }
𝟐 𝒔+𝟏 𝟐 𝒔𝟐+𝟏 𝟐 𝒔𝟐+𝟏
𝟏
= 𝟐 {𝒆−𝒕 − 𝐜𝐨𝐬 𝒕 + 𝐬𝐢𝐧 𝒕}

Practice problems:
Find the Inverse Laplace transform of the following by using partial fractions.
4 𝑠−2
1. 𝐿−1 {(𝑠+1)(𝑠+2)} 2. 𝐿−1 {(𝑠+2)(𝑠+3)}

Answers:

1. 4(𝑒 −𝑡 − 𝑒 −2𝑡 ) 2. −4𝑒 −2𝑡 + 5𝑒 −3𝑡

4.16 First shifting theorem of Inverse Laplace transform:

If 𝑳−𝟏 { 𝒇̅(𝒔)} = 𝒇(𝒕) then 𝑳−𝟏 { 𝒇̅(𝒔 − 𝒂)} = 𝒆𝒂𝒕 𝒇(𝒕) and 𝑳−𝟏 { 𝒇̅(𝒔 + 𝒂)} = 𝒆−𝒂𝒕 𝒇(𝒕).
Problems:
Find the following by applying first shifting theorem.
𝟏 𝒔
1. 𝑳−𝟏 { } 2. 𝑳−𝟏 { }
(𝒔+𝟐)𝟐+𝟏𝟔 𝒔𝟐+𝟒𝒔+𝟓

Sol:
1 1
1. Consider 𝐿−1 { } = 𝐿−1 { }
(𝑠+2)2 +16 (𝑠+2)2+42

1
= 𝑒 −2𝑡 𝐿−1 {𝑠 2+42 } {∵ 𝑏𝑦 𝐹𝑖𝑟𝑠𝑡 𝑠ℎ𝑖𝑓𝑡𝑖𝑛𝑔 𝑡ℎ𝑒𝑜𝑟𝑒𝑚}
1 4 𝟏
= 𝑒 −2𝑡 𝐿−1 { } = 𝒆−𝟐𝒕 𝒔𝒊𝒏 𝟒𝒕
4 𝑠 2 +42 𝟒
𝒔 𝒔
2. Consider 𝑳−𝟏 { } = 𝑳−𝟏 { }
𝒔𝟐+𝟒𝒔+𝟓 𝒔𝟐 +𝟐.𝟐𝒔+𝟒+𝟏
(𝑠 + 2) − 2
= 𝐿−1 { }
(𝑠 + 2)2 + 1
𝑠−2
= 𝑒 −2𝑡 𝐿−1 {𝑠 2+1} {∵ 𝑏𝑦 𝐹𝑖𝑟𝑠𝑡 𝑠ℎ𝑖𝑓𝑡𝑖𝑛𝑔 𝑡ℎ𝑒𝑜𝑟𝑒𝑚}

𝑠 1
= 𝑒 −2𝑡 𝐿−1 {𝑠 2+1} − 2𝑒 −2𝑡 𝐿−1 {𝑠 2+1}
= 𝑒 −2𝑡 cos 𝑡 − 2𝑒 −2𝑡 sin 𝑡 = 𝒆−𝟐𝒕 (𝐜𝐨𝐬 𝒕 − 𝟐 𝐬𝐢𝐧 𝒕)

22
Laplace transforms

Practice problem:
𝟏 𝟏
1. Find 𝑳−𝟏 { } by applying first shifting theorem. {𝑨𝒏𝒔: 𝟐 𝒆−𝟐𝒕 𝒔𝒊𝒏 𝟐𝒕}
𝒔𝟐 +𝟒𝒔+𝟓

4.17 Second shifting theorem of Inverse Laplace transform:

Form I: If 𝑳−𝟏 { 𝒇̅(𝒔)} = 𝒇(𝒕) then 𝑳−𝟏 {𝒆−𝒂𝒔 𝒇̅(𝒔)} = 𝒇(𝒕 − 𝒂). 𝑯(𝒕 − 𝒂)
Where 𝑯(𝒕 − 𝒂) is Heaviside unit function.
𝒇(𝒕 − 𝒂), 𝒇𝒐𝒓 𝒕 > 𝒂
Form II: If 𝑳−𝟏 { 𝒇̅(𝒔)} = 𝒇(𝒕) and 𝒈(𝒕) = {
𝟎, 𝒇𝒐𝒓 𝒕 < 𝒂
then 𝑳−𝟏 {𝒆−𝒂𝒔 𝒇̅(𝒔)} = 𝒈(𝒕).
Problems:
𝑒 −3𝑠
Find: 1. 𝐿−1 { }
(𝑠−4)2
1+𝑒−𝜋𝑠
2. 𝐿−1 { }
𝑠 2 +1
Sol:
𝑒 −3𝑠 1
1. Given 𝐿−1 { } is of the form 𝑳−𝟏 {𝒆−𝒂𝒔 𝒇̅(𝒔)} with 𝒂 = 𝟑, 𝒇̅(𝒔) = { 2}
(𝑠−4)2 (𝑠−4)

And by second shifting theorem of Inverse Laplace transform, we know that

̅ (𝒔)} = 𝒇(𝒕 − 𝒂). 𝑯(𝒕 − 𝒂)
𝑳−𝟏 {𝒆−𝒂𝒔 𝒇
𝑒 −3𝑠
=> 𝐿−1 { } = 𝑓 (𝑡 − 3). 𝐻 (𝑡 − 3) − − − −→ (∗)
(𝑠 − 4)2
1 1
̅ (𝒔)} = 𝑳−𝟏 {
𝑓𝑜𝑟, 𝑓 (𝑡) = 𝑳−𝟏 { 𝒇 } = 𝑒 4𝑡 { }
= 𝑒 4𝑡 . 𝑡
(𝑠 − 4)2 𝑠2
=> 𝑓 (𝑡 − 3) = (𝑡 − 3). 𝑒4(𝑡−3)
𝑒 −3𝑠
ℎ𝑒𝑛𝑐𝑒, 𝑓𝑟𝑜𝑚 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (∗), 𝐿−1 { } = (𝒕 − 𝟑). 𝒆𝟒(𝒕−𝟑) . 𝑯(𝒕 − 𝟑)
(𝑠 − 4)2

1+𝑒 −𝜋𝑠 1 𝑒 −𝜋𝑠

2. Consider 𝐿−1 { } = 𝐿−1 { } + 𝐿−1 { }
𝑠 2+1 𝑠 2+1 𝑠 2 +1

𝟏 + 𝒆−𝝅𝒔 𝒆−𝝅𝒔
=> 𝑳−𝟏 { } = 𝐬𝐢𝐧 𝒕 + 𝑳−𝟏 { } − −→ (𝟏)
𝒔𝟐 + 𝟏 𝒔𝟐 + 𝟏
𝒆−𝝅𝒔 1
Now, 𝑳−𝟏 {𝒔𝟐 +𝟏} is of the form 𝑳−𝟏 {𝒆−𝒂𝒔 𝒇̅(𝒔)} with 𝒂 = 𝝅, 𝒇̅(𝒔) = {𝒔𝟐+𝟏}

̅ (𝒔)} = 𝑳−𝟏 { 𝟐1 } = sin 𝑡

Then 𝑓 (𝑡) = 𝑳−𝟏 { 𝒇 𝒔 +𝟏

=> 𝑓 (𝑡 − 𝝅) = 𝒔𝒊𝒏(𝒕 − 𝝅) − −−→ (𝟐)

23
Laplace transforms

And by second shifting theorem of Inverse Laplace transform, we know that

̅ (𝒔)} = 𝒇(𝒕 − 𝒂). 𝑯(𝒕 − 𝒂)
𝑳−𝟏 {𝒆−𝒂𝒔 𝒇
𝒆−𝝅𝒔
=> 𝑳−𝟏 {𝒔𝟐+𝟏} = 𝒔𝒊𝒏(𝒕 − 𝝅). 𝑯(𝒕 − 𝝅) − −−→ (𝟑) {∵ 𝑏𝑦 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛 (2)}

hence, from equations (1) & (3), we get

−𝟏
𝟏 + 𝒆−𝝅𝒔
𝑳 { 𝟐 } = 𝐬𝐢𝐧 𝒕 + 𝒔𝒊𝒏(𝒕 − 𝝅). 𝑯(𝒕 − 𝝅)
𝒔 +𝟏
Practice Problems:
Find the following.
𝑒 −2𝑠 3+5𝑠
1. . 𝐿−1 { } 2. 𝐿−1 { }
𝑠 2 +4𝑠+5 𝑠 2 𝑒2𝑠

Answers: 1. 𝑒 −2(𝑡−2) . sin(𝑡 − 2). 𝐻(𝑡 − 2) 2. (3𝑡 − 1)𝐻(𝑡 − 2)

4.18 Inverse Laplace transform of derivatives:
𝑑 𝑛
If 𝐿−1 { 𝑓 ̅(𝑠)} = 𝑓(𝑡) then 𝐿−1 { 𝑑𝑠𝑛 𝑓 (̅ 𝑠)} = (−1)𝑛 . 𝑡 𝑛 . 𝑓(𝑡) or

𝐿−1 { 𝑓 ̅(𝑛) (𝑠)} = (−1)𝑛 . 𝑡 𝑛 . 𝑓(𝑡)

𝒅 −𝟏 𝒅
*Note: put 𝒏 = 𝟏, 𝑳−𝟏 { 𝒇̅(𝒔)} = (−𝟏). 𝒕. 𝒇(𝒕) => 𝒇(𝒕) = 𝑳−𝟏 { 𝒇̅(𝒔)} −→ (∗)
𝒅𝒔 𝒕 𝒅𝒔

Problems: Find the following by using derivatives.

𝒔+𝟏
1. 𝑳−𝟏 {𝒍𝒐𝒈 (𝒔−𝟏)} 2. 𝐿−1 {𝑡𝑎𝑛−1 (𝑠 + 1)}

Sol: 1. since by the Inverse Laplace transform of derivatives,

−1 −1 𝑑
𝑓(𝑡 ) = ̅
𝐿 { 𝑓 (𝑠)} − −−→ (∗)
𝑡 𝑑𝑠
𝑠+1
For, 𝑓 (̅ 𝑠) = 𝑙𝑜𝑔 ( ) = 𝑙𝑜𝑔(𝑠 + 1) − 𝑙𝑜𝑔(𝑠 − 1)
𝑠−1

𝑑
̅ (𝑠) = 1 1
=> 𝑑𝑠 𝑓 𝑠+1
− 𝑠−1
−𝟏 𝟏 𝟏 −𝟏
Thus, from equation (*), 𝒇(𝒕) = 𝑳−𝟏 { − }= {𝒆−𝒕 − 𝒆𝒕 }
𝒕 𝒔+𝟏 𝒔−𝟏 𝒕
𝑑 1
2. 𝑓 (̅ 𝑠) = 𝑡𝑎𝑛−1 (𝑠 + 1) => 𝑓 ̅(𝑠) =
𝑑𝑠 1+(𝑠+1)2
−1 −1 1 −1 −𝑡 −1 1 −𝟏
Then 𝑓 (𝑡) = 𝐿 { } = 𝑒 𝐿 {2 } = 𝒆−𝒕 𝒔𝒊𝒏𝒕
𝑡 (𝑠+1)2 +1 𝑡 𝑠 +1 𝒕

Practice problems: Find the following by using derivatives.

𝒔+𝟑 𝟏
1. 𝑳−𝟏 {𝒍𝒐𝒈 (𝒔+𝟒)} 2. 𝑳−𝟏 {𝒍𝒐𝒈 (𝟏 + 𝟐)}
𝒔

1 2
Answers: 1. {𝑒 −4𝑡 − 𝑒 −3𝑡 } 2. (1 − 𝑐𝑜𝑠 𝑡)
𝑡 𝑡

24
Laplace transforms

4.19 Inverse Laplace transform of integrals:

∞ 𝑓(𝑡) ∞
If 𝐿−1 { 𝑓 ̅(𝑠)} = 𝑓(𝑡) then 𝐿−1{∫𝑠 𝑓 ̅(𝑠)𝑑𝑠} = or 𝑓(𝑡) = 𝑡. 𝐿−1{∫𝑠 𝑓 ̅(𝑠)𝑑𝑠}
𝑡
Problems:
Find the following by using integrals.
𝟏 𝟐𝒔
1. 𝑳−𝟏 {(𝒔+𝟏)𝟐} 2. 𝑳−𝟏 {(𝒔𝟐−𝟒)𝟐 }

Sol: 1. Since by Inverse Laplace transform of integrals, we have

∞ ∞ 1
𝑓 (𝑡) = 𝑡. 𝐿−1{∫𝑠 𝑓 ̅(𝑠)𝑑𝑠} = 𝑡. 𝐿−1 {∫𝑠 𝑑𝑠}
(𝑠+1)2

1 ∞
= 𝑡. 𝐿−1 {− ( ) }
𝑠+1 𝑠
1
= 𝑡. 𝐿−1 {𝑠+1} = 𝒕. 𝒆−𝒕

∞ ∞ 2𝑠
2. 𝑓(𝑡) = 𝑡. 𝐿−1 {∫𝑠 𝑓(̅ 𝑠)𝑑𝑠} = 𝑡. 𝐿−1 {∫𝑠 (𝑠 2−4)2
𝑑𝑠}

−1
−1 ∞
= 𝑡. 𝐿 {( 2 ) }
𝑠 −4 𝑠
1 𝟏
= 𝑡. 𝐿−1 {𝑠 2−22 } = 𝒕. 𝟐 𝒔𝒊𝒏𝒉 𝟐𝒕

Practice problems:
𝟐
1. Find 𝑳−𝟏 {(𝒔−𝒂) }
𝟑 by using integrals.
{𝐴𝑛𝑠𝑤𝑒𝑟: 𝑡 2 𝑒 𝑎𝑡 }

4.20 Inverse Laplace transform of a function multiplied by s:

If 𝐿−1 { 𝑓 ̅(𝑠)} = 𝑓(𝑡) and 𝑓 (0) = 0 then 𝐿−1 { 𝑠. 𝑓(̅ 𝑠)} = 𝑓 ′ (𝑡)

𝒅𝒏 𝑑𝑛
In general, then 𝑳−𝟏 { 𝒔𝒏 . 𝒇̅(𝒔)} = 𝒏
𝒇(𝒕) , 𝑖𝑓 𝑓 (0) = 0
𝒅𝒕 𝑑𝑡 𝑛
Problems:
𝒔
Find: 1. 𝑳−𝟏 {𝟐𝒔𝟐 −𝟏}

𝑠
Sol: Given 𝐿−1 {2𝑠 2−1} is of the form then 𝐿−1 { 𝑠 𝑛 . 𝑓(̅ 𝑠)} with 𝑛 = 1 and
1
𝒇̅(𝒔) =
2𝑠2 −1

And we know that, 𝐿−1{ 𝑠. 𝑓 (̅ 𝑠)} = 𝑓 ′(𝑡) − −→ (∗)

25
Laplace transforms

1
1 1 1 1 1
√
−1 −1 −1 −1 2
For, 𝑓(𝑡) = 𝐿 { }=𝐿 { 1 }= 𝐿 { 2 }=√ 𝐿 { 2 }
2𝑠2 −1 2(𝑠2 − ) 2 1 2 1
2 𝑠2 −(√ ) 𝑠2 −(√ )
2 2

1 𝑡
=> 𝑓 (𝑡) = 𝑠𝑖𝑛ℎ ( 2)
√2 √

And 𝑓 (0) = 0
𝑠 𝑑 1 𝑡 1 𝑡 1
Thus, from equation (*), 𝐿−1 {2𝑠 2−1} = 𝑑𝑡 { 𝑠𝑖𝑛ℎ ( 2) } = 𝑐𝑜𝑠ℎ ( 2) ( 2)
√2 √ √2 √ √

𝒔 𝟏 𝒕
Hence, 𝑳−𝟏 { }= 𝒄𝒐𝒔𝒉 ( )
𝟐𝒔𝟐 −𝟏 𝟐 √𝟐

𝑠
Practice problem: 1. Find 𝐿−1 {(𝑠+2)2 } {𝐴𝑛𝑠𝑤𝑒𝑟: 𝑒 −2𝑡 (1 − 2𝑡)}

4.21 Inverse Laplace transform of a function divided by s:

𝒇̅(𝒔) 𝒕
If 𝐿−1 { 𝑓 ̅(𝑠)} = 𝑓(𝑡) and 𝑓 (0) = 0 then 𝑳−𝟏 { } = ∫𝟎 𝒇(𝒕)𝒅𝒕
𝒔
𝒇̅(𝒔) 𝒕 𝒕 𝒕
In general, 𝑳−𝟏 { 𝒔𝒏 } = ∫𝟎 ∫𝟎 … ∫𝟎 𝒇(𝒕)𝒅𝒕 𝒅𝒕 … 𝒅𝒕 (𝒏 𝒕𝒊𝒎𝒆𝒔)
1 1
Problems: 1. Find 𝐿−1 {𝑠(𝑠 2+𝑎2 )} 2. 𝐿−1 {𝑠(𝑠+1)}
1
1 (𝑠2 +𝑎2) 𝑓̅(𝑠)
} is of the form 𝐿−1 {
1
Sol: 𝐿−1 { }
2) = 𝐿
−1
{ } where 𝑓 ̅(𝑠) = (𝑠2+𝑎2)
𝑠(𝑠 2+𝑎 𝑠 𝑠

1 1
=> 𝑓 (𝑡) = 𝐿−1 {(𝑠 2+𝑎2 )} = 𝑎 sin 𝑎𝑡

𝑓̅(𝑠) 𝑡
And we know that, 𝐿−1 { } = ∫0 𝑓(𝑡)𝑑𝑡
𝑠

1 𝑡1 1 −𝑐𝑜𝑠 𝑎𝑡 𝑡 𝟏
=> 𝐿−1 { ( 2 } = ∫0 sin 𝑎𝑡 𝑑𝑡 = [ ] = (−𝒄𝒐𝒔 𝒂𝒕 + 𝟏)
𝑠 𝑠 +𝑎 2) 𝑎 𝑎 𝑎 0 𝒂𝟐

1
1 1
2. 𝐿−1 {𝑠(𝑠+1) } = 𝐿−1 { 𝑠+1 } then let 𝑓(̅ 𝑠) = (𝑠+1) => 𝑓(𝑡) = 𝑒−𝑡
𝑠

1 𝑡
And thus, 𝐿−1 {𝑠(𝑠+1)} = ∫0 𝑒 −𝑡 𝑑𝑡=[−𝑒−𝑡 ]𝑡0 = −𝑒−𝑡 + 1 = 1−𝑒−𝑡

Practice problem:
1 1
1. Find 𝐿−1 {𝑠(𝑠+2)} {𝐴𝑛𝑠𝑤𝑒𝑟: (1 − 𝑒 −2𝑡 )}
2

26
Laplace transforms

Convolution product:
If 𝑓(𝑡)and 𝑔(𝑡) be two functions defined for 𝑡 > 0 then the convolution product of
𝑓 (𝑡)and 𝑔(𝑡) can be denoted by 𝑓 (𝑡) ∗ 𝑔(𝑡) and is defined as
𝒕
𝒇(𝒕) ∗ 𝒈(𝒕) = ∫𝟎 𝒇(𝒖)𝒈(𝒕 − 𝒖)𝒅𝒖

4.22 Convolution theorem:

̅
If 𝐿{𝑓 (𝑡)} = 𝑓 (𝑠) ̅ (𝒔). 𝒈(𝒔)
and 𝐿 {𝑔(𝑡)} = 𝑔̅ (𝑠) then 𝑳{𝒇(𝒕) ∗ 𝒈(𝒕)} = 𝒇
Or
𝒕
𝑳−𝟏 { 𝒇̅(𝒔). 𝒈
̅ (𝒔)} = 𝒇(𝒕) ∗ 𝒈(𝒕) = ∫𝟎 𝒇(𝒖)𝒈(𝒕 − 𝒖)𝒅𝒖

*Note:
1. The convolution theorem is useful to obtain the Inverse Laplace transform of the
product of two transforms and helpful to solve ordinary differential equations.
2. 𝑓 (𝑡) ∗ 𝑔(𝑡) = 𝑔(𝑡) ∗ 𝑓(𝑡) (Commutative property)

3. 𝑓 (𝑡) ∗ (𝑔(𝑡) ∗ ℎ (𝑡)) = (𝑓(𝑡) ∗ 𝑔(𝑡)) ∗ ℎ(𝑡) (Associative property)

4. 𝑓 (𝑡) ∗ 0 = 0 = 0 ∗ 𝑓(𝑡)
5. 𝑓 (𝑡) ∗ 1 ≠ 𝑓(𝑡)

Problems on Convolution theorem:

𝟏
1. Find 𝑳−𝟏 {𝒔(𝒔+𝟏)(𝒔+𝟐)} by applying the convolution theorem.

Sol:
𝟏 𝟏
̅ (𝒔) =
Let 𝒇 ̅ (𝒔) =
𝒂𝒏𝒅 𝒈
𝒔 (𝒔+𝟏)(𝒔+𝟐)
𝟏
then 𝒇(𝒕) = 𝐿−1{ 𝑓 ̅(𝑠)} = 𝐿−1 { } = 1 => 𝒇(𝒕) = 𝟏 − −→ (𝟏)
𝒔
𝟏
And 𝒈(𝒕) = 𝐿−1{𝑔(𝑠)} = 𝐿−1 { } {∵ 𝑏𝑦 𝑢𝑠𝑖𝑛𝑔 𝑝𝑎𝑟𝑡𝑖𝑎𝑙 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛𝑠}
(𝒔+𝟏)(𝒔+𝟐)
𝟏 𝟏
= 𝐿−1 {𝒔+𝟏 − 𝒔+𝟐} => 𝒈(𝒕) = 𝒆−𝒕 − 𝒆−𝟐𝒕 − −→ (𝟐)

Now by convolution theorem,

𝑡
𝐿−1 { 𝑓(̅ 𝑠). 𝑔̅ (𝑠)} = 𝑓(𝑡 ) ∗ 𝑔(𝑡 ) = ∫0 𝑓(𝑢)𝑔(𝑡 − 𝑢)𝑑𝑢
1 𝑡
=> 𝐿−1 { } = ∫0 1. (𝑒 −(𝑡−𝑢) − 𝑒 −2(𝑡−𝑢) ) 𝑑𝑢
𝑠(𝑠+1)(𝑠+2)

𝑡
= ∫0 (𝑒 −𝑡 𝑒 𝑢 − 𝑒 −2𝑡 𝑒 2𝑢 ) 𝑑𝑢
𝑒 −2𝑡
= 𝑒 −𝑡 [𝑒 𝑢 ]𝑡0 − [𝑒 2𝑢 ]𝑡0
2

27
Laplace transforms

𝑒 −2𝑡
= 𝑒 −𝑡 (𝑒 𝑡 − 1) − (𝑒 2𝑡 − 1)
2

1 𝑒 −2𝑡
= 1 − 𝑒 −𝑡 − +
2 2

𝟏 𝟏 𝒆−𝟐𝒕
∴ 𝑳−𝟏 { } = − 𝒆−𝒕 +
𝒔(𝒔+𝟏)(𝒔+𝟐) 𝟐 𝟐

1
2. Find 𝐿−1 { } by using convolution theorem.
𝑠(𝑠 2 +4)

1 1 1 𝟏 1
Sol: 𝐿−1 { } = 𝐿−1 { . } and let 𝒇̅(𝒔) = 𝒔 𝒂𝒏𝒅 𝒈
̅ (𝒔) = 2
𝑠(𝑠 2 +4) 𝑠 𝑠 2 +4 𝑠 +4

𝟏
Then 𝒇(𝒕) = 𝐿−1 { 𝑓(̅ 𝑠)} = 𝐿−1 { } = 1 => 𝒇(𝒕) = 𝟏 − −−→ (𝟏)
𝒔
1 1
And 𝒈(𝒕) = 𝐿−1{𝑔(𝑠)} = 𝐿−1 { } = sin 2𝑡 − − − − − − − −−→ (𝟐)
𝑠2 +4 2

Now by convolution theorem,

𝑡
𝐿−1 { 𝑓(̅ 𝑠). 𝑔̅ (𝑠)} = 𝑓(𝑡 ) ∗ 𝑔(𝑡 ) = ∫0 𝑓(𝑢)𝑔(𝑡 − 𝑢)𝑑𝑢
𝑡
−1
1 1
=> 𝐿 { 2 } = ∫ 𝑓 sin 2(𝑡 − 𝑢) 𝑑𝑢
𝑠 (𝑠 + 4) 0 2

𝟏 −𝒄𝒐𝒔 𝟐(𝒕−𝒖) 𝒕 𝟏
= [ ] = [𝒄𝒐𝒔 𝟎 − 𝒄𝒐𝒔 𝟐𝒕]
𝟐 −𝟐 𝟎 𝟒

𝟏 𝟏
∴ 𝑳−𝟏 { } = [𝟏 − 𝒄𝒐𝒔 𝟐𝒕]
𝒔(𝒔𝟐 + 𝟒) 𝟒

Practice problems:
Find the following by applying the convolution theorem.
𝑠2
1. 𝐿−1 {(𝑠 2+4)(𝑠 2+9)}

1
2. 𝐿−1 {𝑠(𝑠 2+2𝑠+2)}

1
3. 𝐿−1 {(𝑠+𝑎)(𝑠+𝑏)}

Answers:
1
1. (5 sin 5𝑡 − 2 sin 2𝑡)
21

1
2. {1 − 𝑒 𝑡 (sin 𝑡 + cos 𝑡)}
2

1
3. (𝑒 −𝑎𝑡 − 𝑒 −𝑏𝑡 )
𝑏−𝑎

28
Laplace transforms

4.23 Applications of Laplace transforms:

 Solution of Ordinary differential equations by Laplace transform:

Laplace transform can be applied to solve Ordinary differential equations with initial
condition (Initial value problem) as in the following working rule.

Working Rule:

Step 1: Consider the given differential equation and take the Laplace transform on both
sides.
Step 2: Write the following relations in step 1.
𝐿{𝑦′} = 𝑠𝐿{𝑦} − 𝑦(0)
𝐿{𝑦′′} = 𝑠 2 𝐿{𝑦} − 𝑠𝑦(0) − 𝑦′(0)
𝐿{𝑦′′′} = 𝑠 3 𝐿{𝑦} − 𝑠 2 𝑦(0) − 𝑠𝑦 ′ (0) − 𝑦′′(0)
And so on…
Step 3: Substitute the given initial conditions in the equation of step 2 and write the
relation for 𝐿{𝑦}.
Step 4: Find 𝑦 by applying Inverse Laplace transform to the equation of step 3.

*Note: In this concept, we consider 𝒙 = 𝒙(𝒕) and 𝒚 = 𝒚(𝒕). i.e., 𝒙 and 𝒚 are functions of 𝒕.

Problems:
𝒅𝟐 𝒚 𝒅𝒚
1. Using Laplace transform, solve +𝟐 + 𝟓𝒚 = 𝒆−𝒕𝒔𝒊𝒏 𝒕,
𝒅𝒕𝟐 𝒅𝒕
given 𝒚(𝟎) = 𝟎, 𝒚′ (𝟎) = 𝟏.
Sol:
Given 𝑦 ′′ + 2𝑦 ′ + 5𝑦 = 𝑒 −𝑡 𝑠𝑖𝑛𝑡
By applying Laplace transform on both sides,

𝐿{𝑦 ′′ + 2𝑦 ′ + 5𝑦} = 𝐿{𝑒 −𝑡 𝑠𝑖𝑛𝑡}

=> 𝐿{𝑦 ′′ } + 2𝑙{𝑦 ′ } + 5𝐿{𝑦} = 𝐿{sin 𝑡}𝑠 →𝑠+1
1
=> 𝑠 2 𝐿{𝑦} − 𝑠𝑦(0) − 𝑦 ′ (0) + 2(𝑠𝐿{𝑦} − 𝑦(0)) + 5𝐿{𝑦} = {𝑠 2+1}
𝑠 →𝑠+1
2 1
=> 𝑠 𝐿{𝑦} − 1 + 2𝑠𝐿{𝑦} + 5𝐿{𝑦} =
𝑠 2 +2𝑠+2
1
=> 𝐿 {𝑦}(𝑠 2 + 2𝑠 + 5) − 1 =
𝑠 2 +2𝑠+2
2 1
=> 𝐿{𝑦}(𝑠 + 2𝑠 + 5) = 1 +
𝑠 2 +2𝑠+2
𝑠 2 +2𝑠+3
=> 𝐿{𝑦}(𝑠 2 + 2𝑠 + 5) =
𝑠 2 +2𝑠+2
𝑠 2 +2𝑠+3
=> 𝐿{𝑦} = (𝑠2
+2𝑠+5)(𝑠 2 +2𝑠+2)
𝑠 2 +2𝑠+3
=> 𝑦 = 𝐿−1 {(𝑠2 } − − − −−→ (∗)
+2𝑠+5)(𝑠 2 +2𝑠+2)
𝑠 2 +2𝑠+3 𝐴𝑠+𝐵 𝐶𝑠+𝐷
Now, consider (𝑠2 = + − −−→ (∗∗)
+2𝑠+5)(𝑠 2 +2𝑠+2) 𝑠 2 +2𝑠+5 𝑠 2 +2𝑠+2

=> 𝑠 2 + 2𝑠 + 3 = (𝐴𝑠 + 𝐵)( 𝑠 2 + 2𝑠 + 2) + (𝐶𝑠 + 𝐷)(𝑠 2 + 2𝑠 + 5)

29
Laplace transforms

=> 𝑠 2 + 2𝑠 + 3
= 𝐴𝑠 3 + 2𝐴𝑠 2 + 2𝐴𝑠 + 𝐵𝑠 2 + 2𝐵𝑠 + 2𝐵 + 𝐶𝑠 3 + 2𝐶𝑠 2 + 5𝐶𝑠 + 𝐷𝑠 2
+ 2𝐷𝑠 + 5𝐷
On comparison,
𝑠 3 → 𝐴 + 𝐶 = 0 − − − − − − − − − −→ (1)
𝑠 2 → 2𝐴 + 𝐵 + 2𝐶 + 𝐷 = 1 − − − −−→ (2)
𝑠 → 2𝐴 + 2𝐵 + 5𝐶 + 2𝐷 = 2 − − − −→ (3)
𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 → 2𝐴 + 5𝐷 = 3 − − − − − −→ (4)
2 1
By solving, 𝐴 = 0, 𝐵 = 3 , 𝐶 = 0 𝑎𝑛𝑑 𝐷 = 3
Then from equation (*) and (**),

𝑠 2 + 2𝑠 + 3
𝑦 = 𝐿−1 { }
(𝑠 2 + 2𝑠 + 5)(𝑠 2 + 2𝑠 + 2)

2 1 1 1
= 𝐿−1 { ((𝑠2 ) + ((𝑠2 )}
3 +2𝑠+5) 3 +2𝑠+2)

2 1 1 1
= 𝐿−1 { ( )+ ( )}
3 (𝑠+1)2 +22 3 (𝑠+1)2 +12

2 sin 2𝑡 1
= [𝑒 −𝑡 ] + [𝑒 −𝑡 sin 𝑡]
3 2 3

𝟏 −𝒕
∴𝒚= 𝒆 {𝒔𝒊𝒏 𝟐𝒕 + 𝒔𝒊𝒏 𝒕}
𝟑
𝒅𝟐 𝒙 𝒅𝒙
2. Using Laplace transform, solve −𝟒 + 𝟖𝒙 = 𝒆𝟐𝒕 ,
𝒅𝒕𝟐 𝒅𝒕
given 𝒙(𝟎) = 𝟐, 𝒙′ (𝟎) = 𝟐.
Sol:
Consider the given Differential equation 𝑥 ′′ − 4𝑦𝑥 ′ + 8𝑥 = 𝑒 2𝑡
By applying Laplace transform on both sides,

𝐿{𝑥 ′′ − 4𝑦𝑥 ′ + 8𝑥} = 𝐿{𝑒 2𝑡 }

1
=> 𝐿{𝑥 } − 4𝐿{𝑥 ′ } + 8𝐿{𝑥 } =
𝑠−2
2 { } ′ 1
=> 𝑠 𝐿 𝑥 − 𝑠𝑥 (0) − 𝑥 (0) − 4(𝑠𝐿{𝑥 } − 𝑦(0)) + 8𝐿{𝑥 } = 𝑠−2
1
=> 𝑠 2 𝐿{𝑥 } − 2𝑠 − 2 − 4𝑠𝐿{𝑥 } + 8 + 8𝐿{𝑥 } = 𝑠−2
1
=> 𝐿{𝑥 }(𝑠 2 − 4𝑠 + 8) − 2𝑠 + 6 = 𝑠−2
1
=> 𝐿{𝑥 }(𝑠 2 − 4𝑠 + 8) = 𝑠−2 + 2𝑠 − 6

1 2𝑠 6
=> 𝐿{𝑥 } = + (𝑠2 − (𝑠2
(𝑠−2)(𝑠 2 −4𝑠+8) −4𝑠+8) −4𝑠+8)
−1 { 1 −1 2𝑠 1
=> 𝑥 = 𝐿 }+𝐿 {(𝑠 2 } − 6𝐿−1 {(𝑠 2 } − −→ (∗)
(𝑠−2)(𝑠 2−4𝑠+8) −4𝑠+8) −4𝑠+8)

30
Laplace transforms

Now, from equation (*),

Consider first term of RHS of equation (*)

1 𝐴 𝐵𝑠+𝐶
= + − −−→ (∗∗)
(𝑠−2)(𝑠 2 −4𝑠+8) 𝑠−2 𝑠 2 −4𝑠+8

=> 1 = 𝐴𝑠 2 − 4𝐴𝑠 + 8𝐴 + 𝐵𝑠 2 − 2𝐵𝑠 + 𝐶𝑠 − 2𝐶

On comparison,

𝑠 2 → 𝐴 + 𝐵 = 0 − − − − − − − − − −→ (1)
𝑠 → −4𝐴 − 2𝐵 + 𝐶 = 0 − − − − − −→ (2)
𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 → 8𝐴 − 2𝐶 = 1 − − − − − −→ (3)
1 1 1
By solving, 𝐴 = 4 , 𝐵 = − 4 𝑎𝑛𝑑 𝐶 = 2
Then from equation (**),
1
𝐿−1 { 2
}
(𝑠 − 2)(𝑠 − 4𝑠 + 8)
1 1 1 1 1 1
= 𝐿−1 { } − 𝐿−1 { 2 } + 𝐿−1 { 2 }
4 𝑠−2 4 𝑠 − 4𝑠 + 8 2 𝑠 − 4𝑠 + 8
1 1 1 1 1
= 𝑒 2𝑡 − 𝐿−1 { } + 𝐿−1 { }
4 4 (𝑠−2)2 +22
2 (𝑠−2)2 +22
1 2𝑡 1 −1 (𝑠−2)+2 1 1
= 𝑒 − 𝐿 { 2 2
} + 𝑒 2𝑡 𝐿−1 { 2 2 }
4 4 (𝑠−2) +2 2 (𝑠 +2 )

1 1 𝑠+2 1 1
= 𝑒 2𝑡 − 𝑒 2𝑡 𝐿−1 { } + 𝑒 2𝑡 𝐿−1 { }
4 4 𝑠 2 +22 2 (𝑠 2 +22 )

𝟏 𝟏 𝟏 𝟏
=> 𝑳−𝟏 { } = 𝒆𝟐𝒕 − 𝒆𝟐𝒕 (𝐜𝐨𝐬 𝟐𝒕 + 𝐬𝐢𝐧 𝟐𝒕) + 𝒆𝟐𝒕 𝐬𝐢𝐧 𝟐𝒕 −→ (𝑰)
(𝒔 − 𝟐)(𝒔𝟐 − 𝟒𝒔 + 𝟖) 𝟒 𝟒 𝟒
Next, from RHS of equation (*),
2𝑠
Consider the second term 𝐿−1 {(𝑠 2−4𝑠+8)} = 𝟐𝒆𝟐𝒕 (𝐜𝐨𝐬 𝟐𝒕 + 𝐬𝐢𝐧 𝟐𝒕) − −→ (𝑰𝑰)

And the third term

1 1 1 𝟏 𝟐𝒕
𝐿−1 { } = 𝐿−1 { } = 𝑒 2𝑡 −1 {
𝐿 } = 𝒆 𝐬𝐢𝐧 𝟐𝒕 −→ (𝑰𝑰𝑰)
(𝑠 2 − 4𝑠 + 8) (𝑠 − 2)2 + 22 𝑠 2 + 22 𝟐

Finally, from equations (∗), (𝐼 ), (𝐼𝐼 ) & (𝐼𝐼𝐼)

1 1 1 1
𝑥 = 4 𝑒 2𝑡 − 4 𝑒 2𝑡 (cos 2𝑡 + sin 2𝑡) + 4 𝑒 2𝑡 sin 2𝑡 + 2𝑒 2𝑡 (cos 2𝑡 + sin 2𝑡) + 2 𝑒 2𝑡 sin 2𝑡

𝟏 𝟕 𝟓
𝒙 = 𝒆𝟐𝒕 ( + 𝐜𝐨𝐬 𝟐𝒕 + 𝐬𝐢𝐧 𝟐𝒕)
𝟒 𝟒 𝟐

31
Laplace transforms

3. Using Laplace transform method, solve (𝑫𝟐 + 𝟏)𝒚 = 𝟔 𝒄𝒐𝒔 𝟐𝒕, 𝒕 > 𝟎,
If 𝒚 = 𝟑, 𝑫𝒚 = 𝟏 𝒘𝒉𝒆𝒏 𝒕 = 𝟎.
Sol:
Consider the given Differential equation (𝐷2 + 1)𝑦 = 6 𝑐𝑜𝑠 2𝑡
=> 𝑦 ′′ + 𝑦 = 6 𝑐𝑜𝑠 2𝑡
Applying Laplace transform on both sides,
𝐿{𝑦′′} + 𝐿{𝑦} = 6 𝐿{𝑐𝑜𝑠 2𝑡}
6𝑠
=> 𝑠 2 𝐿{𝑦} − 𝑠𝑦(0) − 𝑦 ′ (0) + 𝐿{𝑦} = 𝑠 2 +4

But, given that 𝒚(𝟎) = 𝟑, 𝒚′ (𝟎) = 𝟏, then we write

6𝑠
=> (𝒔𝟐 + 𝟏)𝐿{𝑦} − 3𝑠 − 1 = 𝑠 2+4
6𝑠 3𝑠 1
=> 𝐿{𝑦} = (𝒔𝟐 +𝟏)(𝑠 2+4) + 𝑠 2+1 + 𝑠 2+1

6𝑠 𝑠 1
=> 𝑦 = 𝐿−1 {(𝒔𝟐 +𝟏)(𝑠 2+4)} + 3 𝐿−1 {𝑠 2+1} + 𝐿−1 {𝑠 2+1} − −−→ (∗)

Now, consider,
6𝑠 𝐴𝑠 + 𝐵 𝐶𝑠 + 𝐷
= 2 +
(𝒔𝟐 + 𝟏)(𝑠 + 4) 𝑠 + 1 𝑠 2 + 4
2

=> 6𝑠 == 𝐴𝑠 3 + 4𝐴𝑠 + 𝐵𝑠 2 + 4𝐵 + 𝐶𝑠 3 + 𝐶𝑠 + 𝐷𝑠 2 + 𝐷
On comparison,

𝑠 3 → 𝐴 + 𝐶 = 0 − − − −→ (1)
𝑠 2 → 𝐵 + 𝐷 = 0 − − − −→ (2)
𝑠 → 4𝐴 + 𝐶 = 0 − − − −→ (3)
𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 → 4𝐵 + 𝐷 = 0 − − − −→ (4)
𝐵𝑦 𝑠𝑜𝑙𝑣𝑖𝑛𝑔 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛𝑠 (1)&(3), 𝐴 = 2, 𝐶 = −2
And 𝐵𝑦 𝑠𝑜𝑙𝑣𝑖𝑛𝑔 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛𝑠 (2)&(4), 𝐵 = 0, 𝐷 = 0
6𝑠 2𝑠 2𝑠
∴ = −
(𝒔𝟐 + 𝟏)(𝑠2 + 4) 𝑠2 + 1 𝑠2 + 4

And from equation (*),

2𝑠 2𝑠 𝑠 1
𝑦 = 𝐿−1 {𝑠 2+1 − 𝑠 2+4} + 3 𝐿−1 {𝑠 2+1} + 𝐿−1 {𝑠 2+1} = 2 𝑐𝑜𝑠𝑡 − 2 cos 2𝑡 + 3𝑐𝑜𝑠𝑡 + 𝑠𝑖𝑛𝑡

∴ 𝒚 = 𝟓 𝒄𝒐𝒔 𝒕 − 𝟐 𝒄𝒐𝒔 𝟐𝒕 + 𝒔𝒊𝒏 𝒕

32
Laplace transforms

Practice problems:
Solve the following differential equations by using Laplace transform method.
𝜋
1. 𝑥 ′′ + 9𝑥 = 𝑠𝑖𝑛 𝑡, 𝑥 (0) = 1 & 𝑥 ( ) = 1 [𝐻𝑖𝑛𝑡 𝑢𝑠𝑒 𝑥 ′ (0) = 𝑘 ]
2
2. (𝐷 2 + 4𝐷 + 5)𝑦 = 5, 𝑔𝑖𝑣𝑒𝑛 𝑦(0) = 0 = 𝑦′(0)
3. (𝐷 2 + 2𝐷 + 1)𝑦 = 3𝑡𝑒 −𝑡 , 𝑔𝑖𝑣𝑒𝑛 𝑦(0) = 4, 𝑦 ′ (0) = 0

Answers:
1 1 5
1. 𝑥 = (𝑠𝑖𝑛 𝑡 − 𝑠𝑖𝑛 3𝑡) + 𝑐𝑜𝑠 3𝑡 − 𝑠𝑖𝑛 3𝑡
8 3 6

2. 𝑦 = 1 − 𝑒 −2𝑡 (𝑐𝑜𝑠 𝑡 + 2 𝑠𝑖𝑛 𝑡)

𝑡3
3. 𝑦 = 𝑒 −𝑡 ( + 4𝑡 + 4)
2

-oo0oo-

33
 UNIT-5

Applications of First Order Ordinary Differential Equations

Ordinary Differential Equations have lot of applications in real life. First order differential
equations arise in many practical problems. In this section, some practical problems are
discussed such as Newton’s law of cooling, Law of Natural growth and decay etc.,

Newton’s law of cooling

According to Newton’s law of cooling, the temperature of a body changes as a rate which is
proportional to the difference of temperature between the surrounding medium and that of a
body itself.

If is the temperature of the surrounding medium and be the temperature of a body at any
time t, then where k is a constant.

To solve the above first order differential equation, separation of variables is used.

Separating the variables, we get

Integrating both sides,

where C is a integration constant.

which is the required solution.

Problems

1. A body originally at cools down to in 20 minutes, the temperature of the air

being What will be the temperature of the body after 40 minutes from the original?

Solution: If be the temperature of the body at any time t,

then where k is a constant.

Separating the variables, we get

Integrating both sides,

where C is a integration constant.

when t=0, and when t=20,

then C = 40 and and

so and when t = 40 minutes

2. The temperature of a cup of coffee was , when freshly poured the temperature of the
surrounding was . In one minute it was cooled to How long a period must elapse,
before the temperature of the cup becomes ?

Solution:

Let T be the temperature of a cup of coffee at any time t and be the temperature of the
surrounding room.
Practice Problems:

1. An object cools from in half an hour when surrounded by air whose

temperature is Find its temperature at the end of another half an hour.

2. A cooper ball is heater to a temperature of then at time t = 0, if it is placed in water

which is maintained at If at t = 3 minutes, the temperature of the ball is reduced to
Find the time at which the temperature of the ball is

Law of Natural growth and decay

If the rate of change of a quantity is proportion to the quantity present at the given time is
known as Law of Natural growth and decay.

The natural growth equation is given where k is positive constant and k is known as
growth constant, or natural growth rate.

The natural decay equation is given where k is positive constant and k is known as
decay constant, or natural decay rate.

1. Apparently in a sample radioactive material, a fixed fraction of the material will

spontaneously decay in each unit of time. Write a differential equation that describes the
mass of material at time t.

Solution: If is the mass at time t, then at time t, the rate of change of M is -kM where k
is a positive constant. Thus, we have

2. If a population grows exponentially, and the population is 100 at 1pm, while it is 144 at
3pm, find the population at 4pm.
Solution: Let denote the population at a time t hours afternoon.

3.

Solution:
Applications of First order Ordinary Differential Equations: Electrical Circuits
Applications of Higher order Ordinary Differential Equations:
Electrical circuits
Basic Relations Between Elements of Electric Circuits
𝒅𝒒
(i) 𝒊 = or 𝑞 = 𝑖 𝑑𝑡 [ since current is the rate of flow of electricity]
𝒅𝒕
(ii) Voltage drop across resistance R=Ri
𝑑𝑖
(iii)Voltage drop across inductance 𝐿 = 𝐿 𝑑𝑡
𝑞
(iv) Voltage drop across capacitance 𝐶 = 𝐶

Kirchhoff's laws: The formulation of differential equations for an electrical circuit depends on
the following two Kirchhoff's laws which are of cardinal importance:
1. Voltage law. The algebraic sum of the voltage drops around any closed circuit is equal to
the resultant electromotive force in the circuit.
2. Current law. The algebraic sum of the currents flowing into (or from) any node is zero.

Let i be the current flowing in the circuit at any time t. Then by Kirchhoff's first law, we have
sum of voltage drops across R and L=E
Applications of Higher order Ordinary Differential Equations: Simple
harmonic motion
Example1. In the case of a stretched elastic horizontal string which has one end fixed and a
particle of mass m attached to the other, find the equation of motion of the particle given that
l is the natural length of the string and e is its elongation due to weight mg. Also find the
displacement s of particle when initially s=o v=0.

Example 2. A particle of mass m executes S.H.M. in the line joining the points A and B, on a
smooth table and is connected with these points by elastic strings whose tensions in
equilibrium are each T. If l, l' be the extensions of the strings beyond their natural lengths,
find the time of an oscillation.
Applications of Higher order Ordinary Differential Equations:
Bending of Beams