A Course on

Vector space and Matrix

For Group theory

 For application of symmetry group in chemical system require concept of vector space

So far we deal with three dimensional vector space: We specify any vector by
three mutually perpendicular unit vectors

v  a x iˆ  a y ˆj  a z kˆ

Where, iˆ, ˆj , kˆ mutually perpendicular unit vector [chosen along three Cartesian axes [X, Y & Z]
And ax, ay and az are components along three Cartesian axes

Lets consider another vector, w  bx iˆ  b y ˆj  bz kˆ

Each component of the resultant
 Addition of these two vector: vector is the sum of the
components of individual vectors
v  w  ( a x  bx )iˆ  ( a y  b y ) ˆj  ( a z  bz ) kˆ
 A vectors is uniquely specified by its components, thus they can be
represented by 3-tuples of the components

v  ( a x , a y , a z ); w  (bx , b y , bz )

Unit vectors in terms of 3-tuples are:

iˆ  (1, 0, 0 ); ˆj  ( 0, 1, 0 ); kˆ  ( 0, 0, 1)
We can summarize the properties of the vectors,

(1) Sum of any two vector is also a vector

αβ  λ
Now consider collection of all vectors {V} in three dimension, then it is
closed under addition
(2) Vector addition is commutative and associative

u  v  v  u ; u  ( v  w )  (u  v )  w
 (3) There exist a null vector with the following properties
0 v  v
This equation is satisfied for every vector of the space. Note that every
components of null vector is zero

(4) multiple of vector by a scalar is also a vector belong to the same vector space {V}

cv  ca x iˆ  ca y ˆj  ca z kˆ

(5) Scalar multiplication is commutative, associative and distributive with

respect to vector addition
ab u  ba u; a b u   ba u; b u  v   b u  b v ;
Distributive properties also hold for scalar addition
v a  b   a v  b v ;
(6) For every vector there is an associated inverse vector such that ,

v  ( v )  0

(7) 0 v  0; 1v  v

Now generalize the idea of 3 dimensional vector space by considering

(i) Dimension n, where n can be any integer number
(ii) Scalars are complex number
(iii) They satisfy the listed 7 properties

A vector space consists of a set of vectors which satisfy the

above listed properties.
A set of vector v1 , v 2 , v 3 ..... v N Any one of them cannot be represented by the linear
combination of the rest
(We use bra and ket notation for vector)
Will be called linearly independent when For example, in three dimensions the unit vector k is linearly
a1 v1  a2 v2  a3 v3  ..... aN v N  0 independent of i and j, but any vector in the xy-plane is
linearly dependent on i and j.
At the condition a1  a 2  a3  .....  a N  0

A collection of vectors is said to span the space if Lets basis: e1 , e 2 , e 3 ..... e N

every vector can be written as a linear combination
of the members of this set. Any vector in the space can be written as,

A set of linearly independent vectors that spans   a1 e1  a 2 e 2  a3 e 3  .....  a N e N

the space is called a basis.
Thus a vector uniquely represented by the (ordered) n-tuple of
The number of vectors in any basis is called the its components:
dimension of the space.

It is often easier to work with the components than with

the abstract vectors themselves. To add vectors, you add
their corresponding components:
 To multiply by a scalar you multiply each component:
The null vector is represented by a string of zeroes:

The components of the inverse vector have their signs reversed:

The only disadvantage of working with components is that you have to commit
yourself to a particular basis, and the same manipulations will look very different to
someone working in a different basis.
Consider the ordinary vectors in three dimensions a x iˆ  a y ˆj  a z kˆ
with complex components.

(a) Does the subset of all vectors with az = 0 constitute a vector space? If so, what
is its dimension; if not, why not?

(b) What about the subset of all vectors whose z component is I?

(c) How about the subset of vectors whose components are all equal?

(a) Yes. It follow all the criteria listed of vector space.

Dimension will be 2.

(b) No. Sum of two such vector will produce a vector with z-component 2. Which
does belong to the set.

(c) Yes. It follow all the criteria listed of vector space.

Dimension will be 1.
 An (N-1)-degree polynomial a 0  a1 x  a 2 x 2  .... a N 1 x N 1

Can be written as N- tuples: ( a 0 , a1 , a 2 ..a N 1 )

Addition of such polynomial can be written as

2 N 1 2 N 1
( a 0  a1 x  a 2 x  .... a N 1 x )  (b0  b1 x  b2 x  ....bN 1 x )

( a 0  b0 )  ( a1  b1 ) x  ( a 2  b2 ) x 2  ....( a N 1  bN 1 ) x N 1
Component-wise addition rule.
Thus polynomials of degree N-1 behave like a vector
multiplication of a polynomial by a Scalar amount to multiplying each components in the N tuples
k ( a 0  a1 x  a 2 x 2  .... a N 1 x N 1 )  ( ka 0  ka1 x  ka 2 x 2  .... ka N 1 x N 1 )
k ( a 0 , a1 , a 2 . . a N 1 )  ( ka 0 , ka1 , ka 2 . . ka N 1 )
Hence collection of all polynomials form a vector space of dimension N, with convenient basis 1, x , x 2 .... x N 1
 dy y1  A1e ix  B1e  ix  ( A1 , B1 ) Solutions can be characterized by
 2y parameters A and B. Thus, can be
dx expressed as 2-tuples form.
y 2  A2 e ix  B2 e  ix  ( A2 , B2 )
General solution,
i x  i x
.
y  Ae  Be .
Where, A and B are y k  Ak e ix  Bk e  ix  ( Ak , Bk )
arbitrary constant

i x  i x
Sum of two solutions: y1  y 2  A1e ix  B1e  ix  A2 e ix  B2 e  ix  ( A1  A2 ) e  ( B1  B2 ) e

Thus, sum can be conveniently expressed as, ( A1 , B1 )  ( A2 , B2 )  ( A1  A2 , B1  B2 )

Collection of these solution satisfied properties (1-7) , thus they form a two dimensional vector space with basis:

( e i x , e  i x )
 Collections of all the vectors of the form
 Solution of Schrödinger equations
a1 p x  a 2 p y  a 3 p z
H  E Form a vector space, basis are three p-orbials
Lets linearly independent degenerate solutions, ( px, py, pz )
H  1  E 1; H  2  E 2

Collections of all the vectors of the form

a 1 1  a 2 2

Their linear combination still is a solution of the

a1d 1  a 2 d 2  a 3 d 3  a 4 d 4  a 5 d 5
Schrödinger equation. Thus, collections these
solution form a 2- dimensional vector space Form a vector space, basis are five d-orbials
with basis: (d1, d 2 , d 3 , d 4 , d 5 )

( 1 , 2 )

Similarly, n-fold degenerate solution form a n-

dimensional vector space
 Inner Products:
In three dimensions we encounter two kinds of vector products: the dot product and the cross product.
 Dot product in n-dimentional vector space is usually called the inner product.
 Cross-product: There is no natural way for n-dimensional vector spaces,
v The inner product of two vectors (|>) and |b> ) is a with the following properties:
complex number which we write as,

b    b *  d
all space

A vector space with an inner product is called an inner product space.

v The inner product of any vector with itself is a Two vectors whose inner product is zero are called
nonnegative number its square root is real-we call this orthogonal. A collection of mutually orthogonal
the norm of the vector: normalized vectors,

|  |  
is called an orthonormal set.
v Vectors whose norm is 1, is said to be normalized
v It is always possible and almost always convenient, to choose an orthonormal basis; in that
case the inner product of two vectors can be written very neatly in terms of their components

These results generalize the familiar formulas

  a1 e1  a 2 e 2  a3 e 3  .....  a N e N
a  a x iˆ  a y ˆj  a z kˆ; b  bx iˆ  b y ˆj  bz kˆ
b  b1 e1  b2 e 2  b3 e 3  .....  bN e N

The norm (squared) becomes In three dimensional vector space

In three dimensional vector space

The components themselves are

A set of lineally independent vectors which span the space can be converted to
orthonormal basis set -- The Gram-Schmidt procedure
Eigenvalue of Hermitian matrix is real

 Eigenvectors of Hermitian matrix with different eigenvaules are orthogonal