The Physics

A gate way to all Physics exams: GATE, CSIR-NET, JEST, IIT-JAM, M.Sc entrance

Topic Matrices

Short notes

MULTPLICATION OF MATRICES

If 𝐴 and 𝐵 are any two matrices, their product 𝐴𝐵 is defined only when the number of
columns in 𝐴 is equal to the number of rows in 𝐵. Let 𝐴 = [𝑎𝑖𝑗 ]𝑚×𝑛 and 𝐵 = [𝑏𝑗𝑘 ]𝑛×𝑝 .

Then, their product 𝐴𝐵 = 𝐶 = [𝐶𝑖𝑘 ]𝑚×𝑝 will be a matrix of order 𝑚 × 𝑝, where 𝐶𝑖𝑘 =
∑𝑛𝑗=1 𝑎𝑖𝑗 𝑏𝑗𝑘 .

Properties of matrix multiplication

If 𝐴, 𝐵 and 𝐶 are three matrices such that their product is defined, then

(a) 𝐴𝐵 ≠ 𝐵𝐴 (Generally not commutative)

(b) (𝐴𝐵)𝐶 = 𝐴(𝐵𝐶) (Associative Law)

(c) 𝐴(𝐵 + 𝐶) = 𝐴𝐵 + 𝐴𝐶 (Distributive Law)

(d) If 𝐴𝐵 = 𝐴𝐶 it does not mean that 𝐵 = 𝐶 (Cancellation/Law is not

applicable)

(e) If 𝐴𝐵 = 0. It does not mean that 𝐴 = 0 or 𝐵 = 0

(f) (𝐴𝐵)𝑇 = 𝐵𝑇 𝐴𝑇

[Date] 1
Matrices The Physics
TRANSPOSE OF A MATRIX

The transpose of a matrix 𝐴 is created by switching its rows with its columns. This
new matrix is denoted as 𝐴𝑇 . Consequently, if the original matrix 𝐴 has dimensions
𝑚 × 𝑛, the transpose 𝐴𝑇 will have dimensions 𝑛 × 𝑚.

Properties of transpose

(a) (𝐴𝑇 )𝑇 = 𝐴

(b) (𝐴 ± 𝐵)𝑇 = 𝐴𝑇 ± 𝐵𝑇

(c) (𝐴𝐵)𝑇 = 𝐵𝑇 𝐴𝑇

(d) (𝑘𝐴)𝑇 = 𝑘(𝐴)𝑇

(e) (𝐴1 𝐴2 𝐴3 … . 𝐴𝑛−1 𝐴𝑛 )𝑇 = 𝐴𝑇𝑛 𝐴𝑇𝑛−1 … 𝐴𝑇3 𝐴𝑇2 𝐴1𝑇

DETERMINANT OF A MATRIX

The determinant is a special number that can be calculated from a square matrix. For
a 3 × 3 matrix 𝐴, the determinant, denoted as |𝐴|, is a value that helps in
understanding various properties of the matrix, such as whether it is invertible. The
determinant of a 3 × 3 matrix 𝐴 is given by:

𝑎11 𝑎12 𝑎13

|𝐴| = |𝑎21 𝑎22 𝑎23 |
𝑎31 𝑎32 𝑎33

The formula to calculate the determinant of a 3 × 3 matrix is:

|𝐴| = 𝑎11 (𝑎22 𝑎33 − 𝑎23 𝑎32 ) − 𝑎12 (𝑎21 𝑎33 − 𝑎23 𝑎31 ) + 𝑎13 (𝑎21 𝑎32 − 𝑎22 𝑎31 )

This formula involves summing and subtracting products of elements from the matrix
in a specific pattern.

2
Matrices The Physics
Properties of the determinant of a matrix

(a) |𝐴| exists ⇔ 𝐴 is a square matrix

(b) |𝐴𝐵| = |𝐴||𝐵|

(c) |𝐴𝑇 | = |𝐴|

(d) |𝑘𝐴| = 𝑘 𝑛 |𝐴|, if 𝐴 is a square matrix of order 𝑛.

(e) If 𝐴 and 𝐵 are square matrices of same order then |𝐴𝐵| = |𝐵𝐴|

(f) If 𝐴 is a skew symmetric matrix of odd order then |𝐴| = 0

(g) If 𝐴 = diag⁡(𝑎1 , 𝑎2 … . 𝑎𝑛 ) then |𝐴| = 𝑎1 𝑎2 … 𝑎𝑛 .

(h) |𝐴|𝑛 = |𝐴𝑛 |, 𝑛 ∈ 𝑁.

(i) If |𝐴| = 0, then matrix is called singular.

RANK OF A MATRIX

The rank of an 𝑚 × 𝑛 matrix 𝐴 is the largest order of any non-zero minor in 𝐴. It is

denoted by 𝜌(𝐴). Specifically, the rank is defined as follows:

 A must have at least one non-zero minor of order 𝑟.

 A must not have any non-zero minors of order 𝑟 + 1.

From this definition, we can conclude the following:

(a) If a matrix 𝐴 does not possess any non-zero minor of order 𝑟 + 1, then 𝜌(𝐴) ≤ 𝑟.
(b) If at least one minor of order 𝑟 is non-zero, then 𝜌(𝐴) ≥ 𝑟.

Properties of the Rank of a Matrix

1 The rank of a zero matrix (a matrix with all elements equal to zero) is zero.

2 The rank of a matrix does not change under elementary row operations (row
swapping, row multiplication, and row addition).

3 The rank of a matrix 𝐴 is equal to the rank of its transpose 𝐴𝑇 . That is, 𝜌(𝐴) =
𝜌(𝐴𝑇 ).

3
Matrices The Physics
4 For any two matrices 𝐴 and 𝐵, the rank of their product satisfies 𝜌(𝐴𝐵) ≤
min(𝜌(𝐴), 𝜌(𝐵))

5 For any two matrices 𝐴 and 𝐵 of the same dimensions, 𝜌(𝐴 + 𝐵) ≤ 𝜌(𝐴) + 𝜌(𝐵).

6 The rank of a matrix is greater than or equal to the rank of any of its submatrices.

7 Full Rank:

 A matrix 𝐴 is said to have full row rank if its rank is equal to the number of its
rows.

 A matrix 𝐴 is said to have full column rank if its rank is equal to the number of its
columns.

 If 𝐴 is a square matrix and has full rank, it is invertible.

8 Rank and System of Linear Equations:

The rank of the augmented matrix of a system of linear equations is equal to the
rank of the coefficient matrix if and only if the system is consistent.

ADJOINT OF A MATRIX

The adjoint of a square matrix 𝐴 is obtained by replacing each element of 𝐴 with its
corresponding cofactor, and then taking the transpose of the resulting matrix. This
adjoint matrix is denoted as adj⁡𝐴.

Formally, if 𝐴 = [𝑎𝑖𝑗 ] is a square matrix, and 𝐹𝑖𝑗 is the cofactor of 𝑎𝑖𝑗 in the
determinant |𝐴|, then the adjoint of 𝐴 is given by:
𝑇
adj⁡𝐴 = [𝐹𝑖𝑗 ]

Here, [𝐹𝑖𝑗 ] is the matrix of cofactors of 𝐴, and the superscript 𝑇 denotes the
transpose operation.

Properties of adjoint matrix

If 𝐴, 𝐵 are square matrices of order 𝑛 and 𝐼 is corresponding unit matrix, then

(a) 𝐴(adj⁡𝐴) = |𝐴|𝐼𝑛 = (adj⁡𝐴)𝐴

(b) |adj⁡𝐴| = |𝐴|𝑛−1

4
Matrices The Physics
(c) adj⁡(adj⁡𝐴) = |𝐴|𝑛−2 𝐴

2
(d) |adj⁡(adj⁡𝐴)| = |𝐴|(𝑛−1)

(e) adj⁡(𝐴𝑇 ) = (adj⁡𝐴)𝑇

(f) adj⁡(𝐴𝐵) = (adj⁡𝐵)(adj⁡𝐴)

(g) adj⁡(𝐴𝑚 ) = (adj⁡𝐴)𝑚 , 𝑚 ∈ 𝑁

(h) adj⁡(𝑘𝐴) = 𝑘 𝑛−1 (adj⁡𝐴), 𝑘 ∈ 𝑅

INVERSE OF A MATRIX

If 𝐴 and 𝐵 are two matrices such that 𝐴𝐵 = 𝐼 = 𝐵𝐴, where 𝐼 is the identity matrix,
then 𝐵 is called the inverse of 𝐴 and is denoted by 𝐴−1 . Thus, we have:

𝐴−1 = 𝐵 ⟺ 𝐴𝐵 = 𝐼 = 𝐵𝐴

To find the inverse of a given matrix 𝐴, we use the following formula:

adj⁡𝐴
𝐴−1 =
|𝐴|

Here, adj⁡𝐴 is the adjoint of 𝐴, and |𝐴| is the determinant of 𝐴.

The inverse of 𝐴 exists if and only if |𝐴| ≠ 0. If |𝐴| ≠ 0, the matrix 𝐴 is called
invertible or nonsingular.

Properties of inverse matrix

Let 𝐴 and 𝐵 are two invertible matrices of the same order, then

(a) (𝐴𝑇 )−1 = (𝐴−1 )𝑇

(b) (𝐴𝐵)−1 = 𝐵−1 𝐴−1

−1
(c) (𝐴𝑘 ) = (𝐴−1 )𝑘 , 𝑘 ∈ 𝑁

(d) adj⁡(𝐴−1 ) = (adj⁡𝐴)−1

5
Matrices The Physics
1
(e) |𝐴−1 | = |𝐴| = |𝐴|−1

(f) If 𝐴 = diag⁡(𝑎1 , 𝑎2 , … … , 𝑎𝑛 ), then 𝐴−1 = diag⁡(𝑎1−1 , 𝑎2−1 , … . 𝑎𝑛−1 )

(g) 𝐴𝐵 = 𝐴𝐶 ⇒ 𝐵 = 𝐶 if |𝐴| ≠ 0,

SYSTEM OF LINEAR EQUATIONS

Consider the following system of equations:

𝑝11 𝑢 + 𝑝12 𝑣 + 𝑝13 𝑤 = 𝑞1

{𝑝21 𝑢 + 𝑝22 𝑣 + 𝑝23 𝑤 = 𝑞2
𝑝31 𝑢 + 𝑝32 𝑣 + 𝑝33 𝑤 = 𝑞3

This system can be represented in matrix form as 𝑷𝒖 = 𝑸, where:

𝑝11 𝑝12 𝑝13 𝑢 𝑞1

𝑃 = [𝑝21 𝑝22 𝑝23 ] , ⁡𝑢 = [𝑣 ] , ⁡𝑄 = [𝑞2 ]
𝑝31 𝑝32 𝑝33 𝑤 𝑞3

The system of equations is consistent if there is at least one solution.

Rank Method

The augmented matrix for this system is:

𝑝11 𝑝12 𝑝13 : 𝑞1

[𝑃: 𝑄] = [𝑝21 𝑝22 𝑝23 : 𝑞2 ]
𝑝31 𝑝32 𝑝33 : 𝑞3

The system 𝑷𝒖 = 𝑸 is consistent and has a solution (either unique or infinite) if and
only if the rank of matrix 𝑃 is equal to the rank of the augmented matrix [𝑃: 𝑄], i.e.,
𝜌([𝑃: 𝑄]) = 𝜌(𝑃). For a system with 𝑛 equations, the following conclusions can be
made:

 If 𝜌([𝑃: 𝑄]) = 𝜌(𝑃) = 𝑛 (where 𝑛 is the number of unknowns), the system has
a unique solution.

 If 𝜌([𝑃: 𝑄]) = 𝜌(𝑃) < 𝑛, then 𝑛 − 𝜌(𝑃) unknowns can have arbitrary values,
and the remaining unknowns are uniquely determined. In this case, the
system will have infinitely many solutions.

6
Matrices The Physics
EIGENVALUES AND EIGENVECTORS

Consider a square matrix 𝐴 of order 𝑛, and let 𝐼 be the identity matrix of the same
order. The expression |𝐴 − 𝜆𝐼|, where 𝜆 is a scalar, is known as the characteristic
matrix.

The determinant |𝐴 − 𝜆𝐼| set to zero forms the characteristic equation of the matrix
𝐴. The solutions to this equation are called the eigenvalues of 𝐴.

Properties of Eigenvalues
(a) The eigenvalues of a matrix 𝐴 and its transpose 𝐴𝑇 are identical.
(b) The determinant of 𝐴 is equal to the product of its eigenvalues.
(c) If 𝜆 is an eigenvalue of 𝐴, then 𝑘 + 𝜆 is an eigenvalue of 𝐴 + 𝑘𝐼, where 𝑘 is a scalar.
(d) The sum of the eigenvalues of 𝐴 equals the sum of the elements on its main
diagonal (the trace of 𝐴 ).
(e) Matrices 𝐴 and 𝐵−1 𝐴𝐵 share the same eigenvalues.
(f) For an invertible matrix 𝐴, the eigenvalues of 𝐴−1 𝐵 and 𝐵𝐴−1 are the same.

|𝐴|
(g) If |𝐴| ≠ 0 and 𝜆 is an eigenvalue of 𝐴, then is an eigenvalue of 𝐴−1 .
𝜆

(h) Given eigenvalues 𝜆1 , 𝜆2 , … , 𝜆𝑛 of 𝐴 :

1 1 1
 The eigenvalues of 𝐴−1 are 𝜆 , 𝜆 , … , 𝜆 .
1 2 𝑛

 The eigenvalues of 𝑘𝐴 are 𝑘𝜆1 , 𝑘𝜆2 , … , 𝑘𝜆𝑛 for any scalar 𝑘.

 The eigenvalues of 𝐴𝑚 are 𝜆1𝑚 , 𝜆𝑚 𝑚

2 , … , 𝜆𝑛 .

CAYLEY-HAMILTON THEOREM

Every square matrix satisfies its own characteristic equation. In other words, if 𝐴 is a
square matrix of order 𝑛, the characteristic equation |𝐴 − 𝜆𝐼| = 0 is also satisfied by
substituting 𝜆 with the matrix 𝐴 itself.

Formally, if the characteristic equation of 𝐴 is given by |𝐴 − 𝜆𝐼| = 0, then 𝐴 satisfies

this equation when 𝜆 is replaced by 𝐴.

7
Matrices The Physics
Properties of the Cayley-Hamilton Theorem

1 Every square matrix 𝐴 satisfies its characteristic polynomial. If the characteristic

polynomial of 𝐴 is 𝑝(𝜆) = det⁡(𝐴 − 𝜆𝐼), then 𝑝(𝐴) = 0.

2 The characteristic polynomial of a matrix 𝐴 is a polynomial whose roots are the

eigenvalues of A. If 𝑝(𝜆) = 𝜆𝑛 + 𝑐𝑛−1 𝜆𝑛−1 + ⋯ + 𝑐1 𝜆 + 𝑐0 , then substituting 𝐴 into
this polynomial results in 𝑝(𝐴) = 𝐴𝑛 + 𝑐𝑛−1 𝐴𝑛−1 + ⋯ + 𝑐1 𝐴 + 𝑐0 𝐼 = 0.

3 The degree of the characteristic polynomial of an 𝑛 × 𝑛 matrix 𝐴 is 𝑛, and it is

satisfied by the matrix 𝐴.

4 The Cayley-Hamilton theorem confirms that the eigenvalues of 𝐴 satisfy the

characteristic equation. If 𝜆 is an eigenvalue of 𝐴, then 𝑝(𝜆) = 0.

5 The theorem can be used to express higher powers of 𝐴 in terms of lower powers.
This is particularly useful for simplifying expressions involving 𝐴𝑘 where 𝑘 > 𝑛.

6 If 𝐴 is invertible, the Cayley-Hamilton theorem can help in finding 𝐴−1 by

expressing it in terms of 𝐴 and its powers. For example, if

𝑝(𝜆) = 𝜆𝑛 + 𝑐𝑛−1 𝜆𝑛−1 + ⋯ + 𝑐0 , then:

0 = 𝐴𝑛 + 𝑐𝑛−1 𝐴𝑛−1 + ⋯ + 𝑐1 𝐴 + 𝑐0 𝐼

Rearranging this equation can provide an expression for 𝐴−1 .

For more materials…join the course..

8
Matrices The Physics