DISTANCE LEARNING INSTITUTE

UNIVERSITY OF LAGOS

Author: MUTEEU A. OLOPADE (PhD)

QUANTUM MECHANICS I
(PHS 302)
MODULE

1
TABLE OF CONTENT S Page

GENERAL INTRODUCTION ....................................................................................................... 5

Study Session 1 VECTOR SPACES AND OPERATORS ....................................................... 6
1.0 Main Contents ............................................................................................................ 6
1.1 Vector Spaces ............................................................................................................. 6
1.2 Linear Independence ................................................................................................ 9
1.3 Basis Vector ................................................................................................................. 9
1.4 Inner or Scalar Product ................................................................................................. 8
1.5 Norm of a Vector………................................................................................................11

In-text questions .................................................................................................................................. 9

Summary of Study Session 1 ........................................................................................................... 12
Self- Assessment Questions for Study Session 1............................................................................ 12
References/Further Readings ........................................................................................................... 13
Study Session 2 ORTHOGONALITY AND ORTHONOMALITY.................................................. 14
2.1 Main Contents ............................................................................................................. 14
2.2 Bra and Ket (Dirac) Notation……………………………………………………………17

2.3 Orthogonal Functions……………………………………………………………………..18

2.4 Gram-Schmidt Orthogonalisation Procedure……………………………………………..19

In-text question ................................................................................................................................. 18

Summary of Study Session 2 ........................................................................................................... 23
Self- Assessment Questions for Study Session 2............................................................................ 23
References/Further Readings ........................................................................................................... 24
Study Session 3 SOME USEFUL MATHEMATICS ON MATRICES ............................................... 25
3.0 Main content………………………………………………………………………………27

3.1 Orthogonal Matrices………………………………………………………………………27

3.2 Symmetric Matrices...................................................................................................... 28

3.3 Hermitian Matrices ....................................................................................................... 28
3.4 UnitaryMatrices ............................................................................................................ 29
3.5 Normal Matrices ........................................................................................................... 29
In Text Questions .............................................................................................................................. 29

2
 Summary of Study Session 3 ........................................................................................................... 32
Self- Assessment Questions for Study Session 3............................................................................ 32
Study Session 4 OPERATORS AND RELATED TOPICS ................................................................ 34
4.0 Main Content........................................................................................................... 34
4.1 Linear Operators…………………………………………………………………….36

4.1.1 Eigenvalues of a linear Operator…….……………………………………….36

4.2 Operators in Quantum Mechanics…………………………………………………….38

4.3 Expectation Value…………………………………………………………………. ...39

4.4 Commutators and simultaneous eigenstates ................................................................ 41

Summary of Study Session 4 ........................................................................................................... 46
Self- Assessment Questions for Study Session 4............................................................................ 46
Study Session 5 INADEQUACIES OF CLASSICAL MECHANICS AND THE SCHRÖDINGER
EQUATION ........................................................................................................... 48
5.1 Main Content ................................................................................................................ 48
5.2 Blackbody radiation………………………………………………………………………50

5.3 Photoelectric effect ........................................................................................................... 50

5.4 Compton effect……………………………………………………………………………..54

5.5 Bohr’s Theory of the Hydrogen Atom……………………………………………………..56

5.6 Heisenberg’s Uncertainty Principle………………………………………………………..57

Summary of Study Session 5 ........................................................................................................... 55

Self- Assessment Questions for Study Session 5............................................................................ 55
Study Session 6 THE SCHROEDINGER EQUATION ....................................................................... 57
6.2 Main Content ................................................................................................................. 57
6.3 The Schrodinger Equation…………………………………………………………………59

6.4 Interpretation of the equation and its solution…………………………………………….62

Summary of Study Session 6 ........................................................................................................... 61

Self- Assessment Questions for Study Session 6............................................................................ 62
References/Further Readings ........................................................................................................... 62
Study Session 7 POSTULATES OF QUANTUM MECHANICS ....................................................... 63
7.1 Postulates of Quantum Mechanics ............................................................................... 63

3
 7.2 The Correspondence Principle……………………………………………………………69

Summary of Study Session 7 ........................................................................................................... 68

Self- Assessment Questions for Study Session 7............................................................................ 69
References/Further Readings ........................................................................................................... 70
Study Session 8 TIME-INDEPENDENT SCHRODINGER EQUATION IN ONE DIMENSION I…...73

8.1 Bound States ................................................................................................................. 71

8.2 Particle in aninfinite potential well ............................................................................. 71
8.3 The finite Potential well…………………………………………………………………..77

Summary of Study Session 8 ........................................................................................................... 79

Self- Assessment Questions for Study Session 8............................................................................ 80
Study Session 9 SCATTERING STATE ............................................................................................ 81
9.1 The Potential Step ........................................................................................................ 81
9.2 The Potential Barrier………………………………………………………………………90

Self- Assessment Questions for Study Session 9............................................................................ 94

Study Session 10 TIME-INDEPENDENT SCHRÖDINGER EQUATION IN ONE DIMENSION II ..... 96
10.1 The Simple harmonic oscillator ............................................................................... 96
Learning Outcomes for Study Session 10 ................................................................................... 96
Self- Assessment Questions for Study Session 10 ...................................................................104
SOLUTIONS TO SELF-ASSESSMENT QUESTIONS..................................................................... 106

4
 GENERAL INTRODUCTION
The dynamics of a quantum system is governed by the celebrated Schrödinger equation

𝑑
𝑖ħ |ψ⟩
𝑑𝑡

where i = √−1 and ħ = 1.05457266 × 10−34 J s is Planck’s h-bar constant. However, what is the
meaning of the symbols |ψ⟩ and H? The answers will be given in the first part of the course
(chapters 1-4), which reviews several physical and mathematical concepts that are needed to
formulate the theory of quantum mechanics. We will learn that |ψ⟩ in equation above represents
the ket-vector state of the system and H represents the Hamiltonian operator H represents the
Hamiltonian operator. The operator H is directly related to the Hamiltonian function in classical
physics, which will be defined in the first chapter. The ket-vector state and its physical meaning
will be introduced in the second chapter. Chapter 3 reviews the position and momentum operators,
whereas chapter 4 discusses dynamics of quantum systems. The second part of the course
(chapters 6-10) is devoted to some relatively simple quantum systems include solutions to
Schrodinger’s equation and Scattering states. Most of the material in these lecture notes is based
on the textbooks in the references.

5
Study Session 1 Vector Spaces and Operators

Introduction
In order to grasp Quantum Mechanics, you need to be conversant with Vector Spaces and other basic ideas
of mathematics. The vector space of twice integrable functions enable you to define a set of functions that
would form a set of ‘coordinates’ for the vector-like functions, such that as we expand a given vector in 2-
dimensional Euclidean space as a linear combination like ai + bj, we could also expand a given ‘quantum-
mechanical function’ as a linear combination of the set of functions. This Unit will teach you how to go
about setting up the set of functions that we shall call an orthonormal set. You shall learn to expand a given
function in terms of the orthonormal set, and get to know how to recover the coefficient of expansion of a
particular function.

Learning Outcomes for Study Session 1

At the end of this study session, you should be able to:

1. Define the term Vector Spaces
2. Give examples of Vector Spaces
3. Define linear independence
4. Understand Inner or Scalar product of two vectors
5. Normalise any given vector

1.0 Main Contents

1.1 Vector Spaces

No doubt, you are quite familiar with the concept of a vector. With vector spaces, we are generalising this
basic idea. In other words, we shall have ‘vectors’ that are no longer just ordinary geometrical vectors, but
vectors of a different kind, but all having similar properties. We shall come across matrices that functions
that you could give the same treatment as you did geometrical vectors.

Definition
Given a set
v1 , v2 ,..., vn = S . If

vi  v j  S  i, j  1, 2, ..., n
(i) 1.1

(ii)
 vi  S  i,  1, 2, .., n ; 1.2

𝛼 𝜖 𝐾 , where 𝐾 is a field, e.g., the real number line ( 𝑅) or the complex plane ( 𝐶) Then, ( 𝑆)

is called a vector space or linear space. The vector space is a real vector space if 𝐾 ≡ 𝑅 , and a

Complex vector space if 𝐾 ≡ 𝐶

6
Condition (i) says that if you add any two vectors of the vector space you will get a member of the space.
Condition (ii) shows that a linear multiplication of any two vectors produces a vector also in the vector
space. That certainly makes sense, doesn’t it? You don’t want a situation where you add two vectors in
your space and get a vector not in the space. Moreover, you avoid a situation where multiplying by a
constant takes your vector away from the space. We are now safe to carry out either operation without
worrying whether the vector we get is a ‘sensible’ vector, because we are sure it is.

A way to remember these two conditions is: Additivity [condition (i)] + homogeneity [condition (ii)] =
linearity.

1.2 Linear Independence

Given a set{𝑣𝑖 }𝑛𝑖=1 . If we can write

a1 v1  a2 v 2      an v n = 0 1.9

and this implies the constants 𝑎1 = 𝑎2 = ⋯ = 𝑎𝑛 = 0, then we say{𝑣𝑖 }𝑛𝑖=1 is a linearly independent set.

If even just one of them is non-zero, then the set is linearly dependent. Think of it: a 3-dimensional
Cartesian vector will be a zero vector, 0, notice the boldface type (not zero scalar), if and only if the three
components are independently zero. Thus, for instance, i, j, and k, the traditional unit vectors in 3-
dimensional Cartesian space are linearly independent. Mathematically, this means that 𝜶𝒊 + 𝜷𝒋 + 𝜸𝒌 = 𝟎
if and only if 𝛼 = 𝛽 = 𝛾 = 0.

1.3 Basis Vector

Let V be an n - dimensional vector space. Any set of n linearly independent vectors 𝑒1 , 𝑒2 , … , 𝑒𝑛 . forms a
basis for V . Thus, any vector v  V can be expressed as a linear combination of the vectors𝑒1 , 𝑒2 , … , 𝑒𝑛 . ,
i.e.,

x  x1e1  x2 e 2    xn e n
1.10

Then we say that the vector space 𝑉 is spanned by the set of vectors{𝑒1 , 𝑒2 , … , 𝑒𝑛 } and that the set
{𝑒1 , 𝑒2 , … , 𝑒𝑛 }. It is said to be a basis for 𝑉.

If we wish to write any vector in 1 (say 𝑥.) direction, we need only one (if possible, a unit) vector. Any two
vectors in the x direction must be linearly dependent, for we can write one as 𝑎1 𝑖, and the other 𝑎2 𝑖, where
a1 and a 2 are scalars.

We form the linear combination

c1 (a1i)  c2 (a2i)  0 1.11

7
Where 𝑎1 , and 𝑎1 , are scalar constants.
𝑎2
Obviously,𝑐1 and 𝑐2 need not be zero for the expression to hold, for 𝑐1 = 𝑐2 would also satisfy
𝑎1
expression (1.11).

We conclude therefore that the vectors must be linearly dependent.

Can you then see that we can say that in general, any 𝑛 + 1 vectors in an 𝑛 −dimensional space must be
linearly dependent?

1.4 Inner or Scalar Product

Here, we shall expand your idea of the inner product of two vectors. In your first year in the
University, you came across the dot or inner product of two vectors. In this section, we shall
extend that idea, as mathematicians do, to other vector-like quantities. But first, let us take a look
at the properties of an inner product.

Properties of the Inner Product

Let 𝑉 be a vector space, real or complex. Then, the inner product of 𝑣, 𝑤 𝜖 𝑣 written as

(v,w), has the following properties:

(i) (v,v)  0 1.12

(ii) (v,v) = 0 if and only if v = 0 1.13

(iii) (v,w) = (w,v) (Symmetry) 1.14

(iv) (cv, w) = c * ( v, w ) ; ( v, cw) = c ( v , w ) 1.15

(v) ( v, w  z)  ( v, w)  ( v, z) 1.16

( v, w )  v w
(vi) 1.17

Where 𝑐 ∗ is the complex conjugate of the scalar c.

8
1.5 Norm of a Vector

Let X be a vector space over K , the real or complex number field. A real valued function ∥∥ on X is a
norm on X (i.e. ∥∥∶ 𝑋 ⟶ 𝑅) if and only if the following conditions are satisfied:

x 0
(i) 1.19

x 0
(ii) If and only if x  0 1.20

xy  x  y  x, y  X
(iii) (Triangle inequality) 1.21

x   x x  X and   C
(iv) (Absolute homogeneity) 1.22

The norm of a vector is its “distance” from the origin. Once again, you can see the basic idea of the
distance of a point from the origin being generalised to the case of the vectors in any vector space.

x
is called the norm of x. In the case where X  R , the real number line, the norm is the absolute value,
x
.If the norm of v in the vector space V is unity, such a vector is said to be normalised. In any case, even
if a vector is not normalised, we can normalise it by dividing by the norm.

In-text questions

1. The set of Cartesian vectors in 3-dimensions,

V3 a, b  V3 ,   R .

a  b  V3
(i) 1.3

(ii)  a  V3 1.4

Of course, you know that when two 3-dimensional vectors are added, you also get a 3-dimensional vector.
Moreover, multiplying a 3-dimensional vector by a real constant will give you a 3-dimensional vector.

2: A set of functions of 𝑥 {𝑓(𝑥), 𝑔(𝑥), … … . } = 𝐹

𝑓(𝑥), 𝑔(𝑥) ∈ 𝐹, 𝜆 𝜖 𝑅 or C

9
(i) f ( x)  g ( x)  F 1.7

(ii)  f ( x)  F 1.8

Adding two functions of x will result in a function of x. It just has to be. Also, multiplying a function of x
by a real number, you get a function of x.

3. Check if the set {𝒊, 2𝒊, 𝒋} is linearly independent.

Solution

We form the expression

c11  c2 2  c33  0

Where 𝜙1 = 𝒊 , 𝜙2 = 2𝒊 𝑎𝑛𝑑 𝜙3 = 𝒋

Thus, 𝒊𝒄1 + 2𝒊𝑐2 + 𝒋𝑐3 = 0

Or 𝒊(𝑐𝟏 + 2𝒊𝑐2 ) + 𝒋𝑐3 = 0

Which implies 𝑐1 + 2𝑐2 = 0 and 𝑐3 = 0, since i and j are non-zero vectors.

We see that 𝑐1 = −2𝑐2 , 𝑐3 = 0

𝑐1 and 𝑐2 do not necessarily have to be zero.

Conclusion: The set is not linearly independent.

4. Show that {𝒊. 2𝒌. 𝒋} is a linearly independent set.

Solution

ic1  2kc2  jc3  0

c1  0, c2  0, c3  0

The set is linearly independent

Note that we have made use of the fact that

xi  yj  zk  0 Implies x  0, y  0, z  0

5. You are quite familiar with the set of vectors (i, j) as the normal basis vectors in 2-dimensional space or
a plane. Show that (i  j, i  j) is also a set of basis vectors for the plane.

10
Solution
We check for linear independence.

 (i  j)   (i  j)  0

Then,

i(   )  j(   )  0

This means that 𝛼 + 𝛽 = 0 and 𝛼 − 𝛽 = 0

Adding the last two equations makes us conclude that.𝛼 = 0 Consequently, 𝛽 is also 0. We conclude that
the two vectors are linearly independent. Since these are two linearly independent vectors in two
dimensional (Euclidean) space (a plane), they form a basis for the plane.

6: Given the vectors a and b in 3-dimensions, i.e., 𝑉0 , we define the inner product as (a, b)  a b where
T

a T is the transpose of the column matrix representing a. This is the dot product you have always been
familiar with.

1 2
a = [0],and b =[1], 𝑎𝑇 = [1 0 1]
1 1

 2
(a, b)  a b  1 0 11  3
T

1

Do not mix

d x  c x 
(c, d)  c d  c x
T
 cy  
c z d y   c x d x  c y d y  c z d z  d x dy 
d z c y   (d, c)
 d z  c z 
, with

c x  c x d x cx d y cx d z 
T

cd  c y  d x dy  
d z  c y d x cyd y

c y d z   dc T
 c z  c z d x cz d y c z d z 


The former is a scalar, and the latter a matrix.

V
6. Given the vector a in 3 , the norm of a is

11
 a  (a, a)
1.23

1
0 
 
1
Thus, if a = , then

1
(a, a)  1 0 10  2
1

a  (a, a)  2

We see that a is not normalised.

1 
a 1  
c  0
a 2 
1
However, is normalised.

Summary of Study Session 1

In this study session 1, you have learnt:
1 Vector spaces are sets that contain some vector-like quantities that satisfy certain conditions.
2 How to check whether a set of vectors is linearly independent.
3 A set of linearly independent vectors is necessary to span a space.
4 N-dimensional vector space V is spanned by the set of n vectors.
5 The norm of a vector is its distance from the ‘origin.’
6 Dividing a vector by its norm normalises it, so that its length is unity.

Self- Assessment Questions for Study Session 1

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 1

Show that the following are vector spaces over the indicated field:

(i) The set of real numbers over the field of real numbers.
(ii) The set of complex numbers over the field of real numbers.
(iii) The set of quadratic polynomials over the complex field.

12
SAQ 2

Check whether the following vectors are linearly independent.

(i) 2i  3 j  k ,  i  j  3k and  3i  2 j  k

SAQ 3

1   1
 ,  
1 1
Show whether or not the set     is a basis for the two-dimensional Euclidean space.

SAQ 4

 1 2
 
Find the coordinates of the vector  2 i  with respect to the basis

1 0 0 1 0  i  1 0  
 ,  ,  ,  
0 1 1 0  i 0  0  1  .

References/Further Readings

1. Mathematical Physics – Butkov, E.

2. Mathematical Methods for Physics and Engineering – Riley, K. F., Hobson, M. P. and Bence,
S. J.

13
Study Session 2 ORTHOGONALITY AND ORTHONOMALITY

Introduction
Orthogonal functions play an important role in Quantum mechanics. This is because they afford us a set of
functions ‘which do not mix,’ just the way you could resolve a vector in two dimensions in the x and y
directions, respectively, with the unit vectors i and j. The dot product of the two unit vectors gives you
zero. We would also like to resolve our vectors in some ‘directions.’ Thus, you need to know about
orthogonal and orthonormal functions. The orthonormal functions would form the possible states you can
find a system. You know such states should not ‘mix.’ In this Unit, you will learn about Orthonormality
and Orthogonality; how to create an orthogonal and subsequently, an orthonormal set and expand a given
function in terms of an orthonormal set. This would naturally lead to an analysis of the probability of
finding a system in any of the states in the orthonormal set. This Unit also gives you an insight into some
elements of matrix algebra.

Learning Outcomes for Study Session 2

When you have studied this session, will equip you with the knowledge of :
 Orthogonal functions
 Orthonormal functions
 Expansion of a given function as a linear combination of a set of orthonormal functions (states).
 Recovering the coefficient of the expansion.
 Finding the probability of finding the system in a given state.
 Some elements of matrix algebra.

2.1 Main Content

Definitions:

(i) We say v 1 and v 2 in a vector space V are orthogonal if their inner product is zero, that is,
( v1 , v 2 )  0 .
(ii) Suppose there exists a linearly independent set,{∅𝑖 }𝑛𝑖=1 i.e., ,{∅1 , ∅2 , … , ∅𝑛 }. such that
( i ,  j )  0 i  j
,
 
n
, then, i i 1 is an orthogonal set.

(iii) If in addition to condition (ii) above,(∅𝑖 , ∅𝑗 ) = 1 then,

i in1 is an orthonormal set.

0 𝑖𝑓 𝑖 ≠ 𝑗
For an orthonormal set, therefore, we can write,(∅𝑖 , ∅𝑗 ) = 𝛿𝑖𝑗 where 𝛿𝑖𝑗 = {
1 𝑖𝑓 𝑖 = 𝑗 .

14
As we have seen earlier, if any vector in the vector space, V , can be written as a linear combination
n
v  a11  a 2 2      a n n   ai i
i 1 2.1

then we say the space is spanned by the complete orthonormal basis

i in1 , where

( m ,  n )   mn
2.2

If
i in1 is an orthonormal set, It follows that we can recover the coefficient of expansion as follows:
n n
( j , v)  ( j ,  ai i )   ai ( j ,  i )  a j
i 1 i 1 2.3

Moreover,
n n n n n
( v, v)  ( a k  k ,  ai i )   a k *  ai ( j ,  i )   ai
2

k 1 i 1 k 1 i 1 i 1 2.4

If, in addition, the vector v is normalised, then ( v, v)  1 and consequent upon equation (2.4),
n

a 1
2
i
i 1

2.5

Do you remember what you learnt about probability in Statistics? The sum of the probability for various
2
ai
possible events is unity. Thus, we can interpret the as the probability that the system which has n
2
a
possible states, assumes state i with probability i . In other words, the probability that the system is in
2
ai
state i is .

2.2 Bra and Ket (Dirac) Notation

We have written the inner product in the form (,) . We could also write it in the form of a bra, |   , and a
ket,   | . This is the Dirac notation. Putting the bra and the ket together forms a ‘bracket’   |   . The set
{ j }nj 1 {|  j }nj 1
of vectors can be seen as a set of bra vectors (space of vectors) . Then, we would need a
{  j |} j 1
n
dual set of vectors (dual space of vectors) to be able to write the inner product. Why? Recall
that we needed to change our column vectors to row vectors to be able to take the inner product of two

15
column vectors? If | B  is a column vector, then  B | is the dual vector, the row vector but with the
entries being the complex conjugate of what they were as | B  .

It follows from the foregoing, that we can write the expansion of a wave function
n
   c j j   cj | j 
j as j 1 2.6

( j , a j )  a( j ,  j ) (a j ,  j )  a * ( j ,  j )
Moreover, and .It follows that
a( j ,  j )  ( j , a j )  (a *  j ,  j )  (a*) * ( j ,  j )
. We can extract the following rule from this:

( j , a j )  (a *  j ,  j )
2.7

More generally, a could be an operator𝐴. Then,

( j , A j )  ( A  j ,  j )
2.8

We can write this in the form,

  j | A |  j  A  j |  j 

Equations 2.3 and 2.4 now become,

n n
  j , v   j |  ai | i    ai   j | i  a j
i 1 i 1 2.9
n n n n n
 v | v   a k  k |  aii    a k *  ai   j | i   ai
2

k 1 i 1 k 1 i 1 i 1 2.10

2.3 Orthogonal Functions

An even function is symmetrical about the y axis. In other words, a plane mirror placed on the axis will
produce an image that is exactly the function across the axis. An example is shown in Fig …a. An odd
function will need to be mirrored twice, once along the y axis, and once along the x axis to achieve the
same effect. Fig. … b is an example of an odd function.
2 n 1
A function f (x) of x is said to be an odd function if f ( x)  f ( x) , e.g., sin x, x , and a function
f (x) of x is said to be an even function if f ( x)  f ( x) , e.g., cos x, x where n  0,1, 2, ....
2n

16
 0 0

a) Odd Function b) Even Function

Some real-valued functions are odd, some are even; the rest are neither odd nor even. However, we can
write any real-valued function as a sum of an odd and an even function.

Let the function beℎ(𝑥), then we can write

ℎ(𝑥) = 𝑓(𝑥) + 𝑔(𝑥) 2.11

Where 𝑓(𝑥) is odd and 𝑔(𝑥) is even. Then, 𝑓(−𝑥) = −𝑓(𝑥) and 𝑔(−𝑥) = 𝑔(𝑥)

h(  x )  f (  x )  g (  x )   f ( x )  g ( x ) 2.12

Adding equations (2.11) and (2.12) gives

h( x)  h( x)  2 g ( x)

Subtracting equation (2.12) from equation (2.11) gives

h( x )  h(  x )  2 f ( x )

It follows, therefore, that

h( x)  h( x)
f ( x) 
2 2.13

and

h( x)  h( x)
g ( x) 
2 2.14

17
2.4 Gram-Schmidt Orthogonalisation Procedure

This provides a method of constructing an orthogonal set from a given set. Normalising each member of
the set then provides an orthonormal set. The method entails setting up the first vector, and then
constructing the next member of the orthogonal set by making it orthogonal to the first member of the set
under construction. Then the next member of the set is constructed in a way to be orthogonal to
the two preceding members. This procedure can be continued until the last member of the set is
constructed.

In-text question
1. Write the function h( x)  e sin x as a sum of odd and even functions
2x

Solution

h( x)  e 2 x sin x , h( x)  e 2 x sin(  x)  e 2 x sin x Therefore, the odd function is

h( x)  h( x) e 2 x sin x  e 2 x sin x e 2 x  e 2 x

f ( x)    sin x
2 2 2

 cosh 2 x sin x

The even function is

h( x)  h( x) e 2 x sin x  e 2 x sin x e 2 x  e 2 x

g ( x)    sin x
2 2 2

 sinh 2 x sin x

It is obvious that the odd function is a product of an odd function and an even function. Likewise, the even
function is a product of two odd functions. We conclude, therefore, that the following rules apply:

Even  Even = Even 2.15

Even  Odd = Odd 2.16

Odd  Odd = Even 2.17

The integral

18
 𝑎
∫−𝑎 𝑓(𝑥)𝑑𝑥 = 0 If 𝑓(𝑥) is odd 2.18

𝑎 𝑎
∫−𝑎 𝑓(𝑥)𝑑𝑥 = 2 ∫0 𝑓(𝑥)𝑑𝑥 If 𝑓(𝑥) is even 2.19

Recall that the inner product in the space of twice integrable complex valued functions of two complex
valued functions 𝑓(𝑥) and 𝑔(𝑥) over the interval a  x  b is defined as (𝑓, 𝑔) = ∫ 𝑓 ∗ (𝑥)𝑔(𝑥)𝑑𝑥.Two
𝑏
𝑎
functions 𝑓(𝑥)and 𝑔(𝑥)are said to be orthogonal over an interval 𝑎 ≤ 𝑥 ≥ 𝑏 if their inner product is zero.

2. Example from function vector space

 2

Construct an orthonormal set from the set 1, x, x ,... over the interval  1  x  1 . Thus, given the set
 f1 , f 2 , f 3 ,..., we want to construct an orthogonal set 1 , 2 ,3 ,..., i.e.
1
1
i ( x) j ( x)dx
= 0, if i  j , then we normalise each member of the set.

Let 1  f1  1 , and  2  f 2  1  x  

Then, we determine  , subject to

(1 ,  2 )  0

1 1
1 x2
11  ( x   ) dx 
2
 x 0
1 1

2.20

  0

Thus,  2  x

Let
3  f 3  2  1  x 2  x  

(1 , 3 )  0 ( 2 ,  3 )  0
subject to and

The first condition gives:

19
 1
1
1  (x 2  x   )dx
=0 2.21

or
1 1 1
x3 x 2 2x 2
  x 1   2 x
1

3 1
2 1
3 0 =0 2.22

2
 2  0
3 2.23

or

1
 
3 2.24

The second condition gives

1 1
1
x  (x 2  x   )dx   ( x 3  x 2  x)dx
1

Or
1 1 1
x4 x 3 x 2
  0
4 1
3 1
2 1 2.25
1
2x 3
0
3 0

or

 0 2.26

Putting the values of  and  from equations (2.24) and (2.26) into the expression
3  f 3  2  1  x 2  x   , we arrive at

1
3  x 2 
3 2.27

 4 , 5 , etc., can be got in a similar fashion.

j
To normalise , we multiply the function by a normalisation constant, A , say, and invoke the relation

20
 1
 A 2 j ( x)dx  1
2
1 2.28

For 1 , this becomes

1 1
 A 1 dx  2 A dx = 1
2 2 2
1 0

from which

2 A2  1

1
A
or 2

1
1 
The normalised function 2 2.29

Similarly,
1
 1
A 2 x 2 dx  1

3 1
1 1 
2
x A
A 2
A   2
2
1
3 1
3 3 3

3
A2 
Thus, 2.

Hence, the normalised function,

3
2  x
2

In like manner,
2
1  2 1 1
2 4 2 2 1
1 A  x  3  dx  20 A  x  3 x  9 dx
2

=1

from which
1
x5 2x3 x
2A  
2
1
5 9 90

21
or

1 2 1
2 A2      1
5 9 9

80 2
A 1
Therefore, 45

45  2 1 
2  x  
The normalised function 8  3

n
3. Example from the n-dimensional Euclidean space, R

Suppose we are given two vectors u and v. The projection of v onto u is indeed,

𝜃 𝒖

∥ 𝒗 ∥ cos 𝜃

Pru v  ( v cos )uˆ

Where  is the angle between the two vectors, and û is the unit vector in the direction of u, and.
u uv
uˆ  cos  
u u  v  u v cos  u v
Therefore, . But . So,

Hence,

 uv  u (u, v )
Pru v   v 
 u  u2 u
(u, v )
u
 u v  (u, u) 2.30

since (u, u) | u | .
2

We proceed as we did in the case of the example from function space.

u1  v1 2.31

The projection of 𝒗𝟐 onto 𝒖𝟏 is 𝑃𝑟𝑢1 𝒗𝟐. This is the component of 𝒗𝟐 in the direction of 𝒖𝟏. The

component of v 2 perpendicular (i.e., orthogonal) to 𝒖𝟏 is (vector subtraction):

22
u 2  v 2  Pru1 v 2
2.32

v3
Similarly the components of perpendicular to both 𝒖𝟏 and 𝒖𝟐 is,

u 3  v 3  Pru1 v 3  Pru2 v 3
2.33

In general then,
n 1
u n  v n   Prui v n
i 1 2.34

We can then normalise each vector

uk
ek 
|| u k || 2.35

Note that
Pru v projects vector v orthogonally onto vector u.

Summary of Study Session 2

In this study session 2, you have learnt:
1 The inner product of a pair orthogonal vector is zero.
2 A basis that consists of orthogonal vectors only is an orthogonal basis.
3 With an orthogonal basis, we can define the probabilities of measurement.
4 The Gram- Schmidt orthonormalisation scheme can be used to create an orthogonal basis.

Self- Assessment Questions for Study Session 2

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 2.1

Which of the following functions are even and which ones are odd?

(i) x 2 sin x cosh x

(ii) | e x | cosh 2 x (iii) sec x

23
SAQ 2.2

Show that

(i) sin 𝑚𝑥 and cos 𝑛𝑥 are orthogonal,    x   .

(ii) sin 𝑚𝑥 and sin 𝑛𝑥 are orthogonal, 𝑚 ≠ 𝑛,    x   .

References/Further Readings

1. Mathematical Physics – Butkov, E.

2. Mathematical Methods for Physics and Engineering – Riley, K. F., Hobson, M. P. and Bence,
S. J.

24
Study Session 3 SOME USEFUL MATHEMATICS ON MATRICES

Introduction

You shall need the following because we often represent an operator in quantum mechanics by a matrix.
e
We shall take as the usual basis in 3-dimensional space, { e1 , e 2 , 3 }. You may also see this basis as {i, j,
k}.

Objectives

This Unit will equip you with the knowledge of:

 Orthonormal functions
 Expansion of a given function as a linear combination of a set of orthonormal functions (states).
 Recovering the coefficient of the expansion.
 Finding the probability of finding the system in a given state.
 Some elements of matrix algebra.

3.0 Main content

3.1 Orthogonal Matrices

A matrix Q such that (Q a )  (Qb)  a  b 

a, b  E is called an orthogonal matrix, i.e., orthogonal
matrices preserve the inner product.

Since (Qa)  (Qb)  b  {Q (Qa)}  b  {(Q Q)a} , a necessary and sufficient condition for Q to be
T T

orthogonal is

QQ T  I
3.1

or equivalently,

Q 1  Q T
3.2

Note that

det(QQ T )  det(Q) det(QT )

 det(Q) det(Q)

25
  det( Q)   1
2

 det(Q)  1 3.3

Q is said to be a proper orthogonal matrix if det(Q)  1 and an improper orthogonal matrix if

det(Q)  1 .

If det (Q)  1 , then

det(Q  I )  det(Q  I ) det(QT )

 det(QQT  QT ) ( det( A) det( B)  det( AB) for any 2 square matrices)

 det( I  Q T ) ( QQ  I for an orthogonal matrix Q)

T

 det( I T  Q TT ) ( det A  det A for any square matrix A.)

T

  det( I  Q) ( I  I and Q  I )
TT
T

  det(Q  I ) ( det( A)   det( A) for any square matrix A.)

(if a number is equal to its negative, it must be zero)

 e 3  Qe 3  e 3
Therefore, 1 is an eigenvalue so that .

e
Choose e1 , e 2 to be orthonormal to 3 . In terms of this basis,

 a b 0
Q   c d 0
 0 0 1
3.4

 a c 0
Q  b d 0
T

0 0 1
3.5
1 0 0 a  b ac  bd 0
2 2

 
QQ T  0 1 0  ca  bd c2  d 2 0
0 0 1  0 0 1
3.6

a2  b2  1  c2  d 2 3.7
26
ac  bd  0  ca  bd 3.8

Also,

det(Q)  1  ad  bc 3.9

ac
b
From equation 3.8, d

Putting this in 3.9 gives

ac 2
ad  1
d 3.10

 a(c 2  d 2 )  d  a  d

Use equation 3.8 in equation 3.7 to get c  b .

Therefore,

 a b 0
Q   b a 0
 0 0 1
3.11

with a  b  1 .
2 2

Thus,   ,  a  cos  , b  sin  ,

so

 cos sin  0
Q   sin  cos 0
 0 0 1
3.12

If you represent the three unit vectors in 3-dimensional Euclidean space by i, j, k, this corresponds to a
rotation about an axis perpendicular to k .

27
3.2 Symmetric Matrices

For a symmetric matrix A, A  A

T

Choose
e1 , e 2 , e 3  , ,
as eigenvectors of A , with eigenvalues 1 2 3 .

Ae k   k e k 3.13

 k (e k  e j )  Ae k  e j
3.14

 e k  AT e j

 e k  Ae j

  j (e k  e j )
.

 j  k ek  e j  0
This means that if , then

e1 , e2 , e3 e i  e j   ij
Choose to be unit vectors, then, .

Aii   i
This means that we could represent a symmetric matrix as a diagonal matrix with only the entries :

 1 0 0
 
A0 2 0
0 0  3 
 3.15

This result is referred to as the spectral representation of a symmetric matrix.

3.3 Hermitian Matrices

The Adjoint (or Hermitian conjugate) of a matrix A is given by

Adj( A)  A  ( A)T *   3.16

A Hermitian matrix is the complex equivalent of a real symmetric matrix, satisfying

A  A 3.17

28
3.4 Unitary Matrices

The complex analogue of a real orthogonal matrix is a unitary matrix, i.e., AA  I or, equivalently,

A  A1 3.18

A unitary matrix is the complex analogue of an orthogonal matrix. As such, we conclude that every
orthogonal matrix and every unitary matrix is diagonalisable (its rows and columns are linearly independent,
meaning that the matrix is non-singular; we can find its inverse). Conversely, we expect that orthogonal and
unitary matrices would diagonalise some matrices, being composed of linearly independent (normalised)
vectors, and hence non-singular.

Real symmetric matrices can be diagonalised by orthogonal matrices. As such, real symmetric matrices with
n distinct eigenvalues are orthogonally diagonalisable.

3.5 Normal Matrices

A normal matrix is one that commutes with its Hermitian conjugate.

i.e., AA   A  A 3.19

In Text Questions
1. Show that sin mx and sin nx are orthogonal, m  n ,    x   .

Solution

 1 
 sin mx sin nxdx  2   [cos( m  n) x  cos(m  n) x]dx
The inner product is   

 
1 1 1 
  sin( m  n ) x  sin( m  n ) x 
2  m  n  mn  
 =0

2. Show that the following is an orthogonal matrix.

29
  2/2 6 / 6  3 / 3
 
Q 0 6 /3 3 /3 
 
 2 /2  6 /6 3 / 3 

 2/2 0 2/2 
 
QT   6 / 6 6 / 3  6 / 6
 
 3 / 3 3/3 3 / 3 

 2/2 6 / 6  3 / 3  2 / 2 0 2/2 
  
QQ   0
T
6 /3 3 / 3  6 / 6 6 / 3  6 / 6
  
 2 / 2  6 / 6 3 / 3   3 / 3 3 /3 3 / 3 

1 0 0
 
  0 1 0
0 0 1
 

Hence, Q is an orthogonal matrix.

0 0 1
  0 0 0
1 0 0
3. Diagonalise the symmetric matrix using the appropriate orthogonal matrix;

Solution
0 0 1
  0 0 0
1 0 0
The characteristic equation is formed by

 0 1
0  0  0
1 0 

 3    0

Eigenvalues are 0, 1 and 1 .

30
 0 0 1  a1   a1  0
 0 0 0   a   0  a   0 
  2  2  
1 0 0  a3   a3  0
For  = 0, the eigenvector is given by 

or

 a1  0
 a   1 
 2  
 a 3  0

For   1 :

 1 0 1   a1 
   0  1 0  a 2   0
 1 0  1  a 3 

or

 a1  1
a   1 0 
 2 2 
 a 3  1
(Normalised)

For : 𝜆 = −1

1 0 1  a1 
  0 1 0 a 2   0
1 0 1  a 3 

or

 a1   1
a   1  0 
 2 2 
 a 3   1 
(Normalised)

The appropriate orthogonal matrix is,

0 1 / 2  1/ 2 
 
1 0 0 
0 1 / 2 1 / 2 
Q= 
31
 and

 0 1 0 
Q   1 / 2 0 1 / 2 
T 
 1 / 2 0 1 / 2 

 0 1 0  0 0 1  0 1 / 2 1 / 2 
 1/ 2  
 0 1/ 2  0 0 0 1 0 0 
 1 / 2 0 1/ 
2  1 0 0 0 1 / 2 1 / 2 
D  Q Q
T
=

 0 1 0  0 1 / 2 1/ 2 
 1/ 2  
 0 1 / 2  1 0 0 
 1 / 2 0 1 / 2  0 1 / 2  1 / 2 
= 

0 0 0 
0 1 0 
 
0 0 1
=

Note that you can only diagonalise a matrix the columns (or rows) of which are linearly independent. The
columns (as well as the rows) of an orthogonal matrix are linearly independent.

Summary of Study Session 3

In this study session, you have learnt that:
5 The inner product of a pair orthogonal vectors is zero.
6 A basis that consists of orthogonal vectors only is an orthogonal basis.
7 With an orthogonal basis, we can define the probabilities of measurement.
8 The Gram- Schmidt orthonormalisation scheme can be used to create an orthogonal basis.

Self- Assessment Questions for Study Session 3

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

32
SAQ 3.1

Write the following as a sum of odd and even functions.

(i) e  x cosh x
(ii) x ln x

SAQ 3.2

Evaluate the following integrals

a

a

(i) a
x 2 n 1 dx n  0,1, 2, ....
, (ii) a
x 2 n dx n  0,1, 2, ....
,

SAQ 3.3

3 x 
 
If the matrix 1 2 is a proper orthogonal matrix, find x.

References/Further Readings
1. Mathematical Physics – Butkov, E.

2. Mathematical Methods for Physics and Engineering – Riley, K. F., Hobson, M. P. and Bence, S. J.

33
Study Session 4: OPERATORS AND RELATED TOPICS

Introduction
Operators are quite important in Quantum mechanics because every observable is represented by a
Hermitian operator. The eigenvalues of the operator are the possible values the physical observable can
take, and the expectation value of the observable in any particular state is the average value it takes in that
particular state. Commuting operators indicate that the corresponding physical observables can have the
same eigenstates, or equivalently, they can both be measured simultaneously with infinite accuracy. You
shall get to learn about all these in this Unit.

Learning Outcomes for Study Session 4

When you have studied this session, you should be able to:
 Define a linear operator.
 Find the eigenvalues of a linear operator.
 Calculate the expectation value of a physical observable in a given state.
 Do commutator algebra.
 Find the matrix elements of a linear operator.
 Write the matrix for a change from one basis to another.

4.0 Main Content

4.1 Linear Operators

A linear map, or linear transformation or linear operator, is a function f : X  Y between vector spaces
X and Y which preserves vector addition and scalar multiplication, i.e.,

𝑓(𝑥1 + 𝑥2 ) = 𝑓𝑥1 + 𝑓𝑥2 4.1

𝑓(𝜆𝑥) = 𝜆 𝑓(𝑥) For 𝜆 𝜖 𝐾, a constant, and 𝑥1 , 𝑥2 𝜖 𝑋 4.2

Equivalently, .
f (ax1  bx 2 )  af ( x1 )  bf ( x 2 ) 4.3

As an example, the differential operator is a linear operator.

d d d
( f1 ( x)   f 2 ( x))   f 1 ( x)   f 2 ( x)
dx dx dx 4.4

Where 𝛼 𝑎𝑛𝑑 𝛽 are constants (scalars) in the underlying field.

4.1.1 Eigenvalues of a Linear Operator

Let A be an operator and 𝜆 the associated eigenvalue corresponding to an eigenvector 𝜓. Then, we can
write

A   4.5

34
Frequently, the operator A is a matrix, and the eigenvector  a column matrix. It follows that

( A   I )  0 4.6

Where I is the appropriate identity matrix, that is, a square matrix that has 1 along its main diagonal and
zero elsewhere.

For a non-trivial solution, we require that the determinant vanish, that is,

A I  0
4.7

Solving the resulting characteristic (or secular) equation, we obtain the possible values of, called the
eigenvalues. Then armed with the eigenvalues, we can then obtain the associated eigenfunctions.

3  2
 
Given the matrix 1 2  , find the corresponding eigenvectors and the eigenvalues.

Solution
u  3  2  u1  u 
u   1       1
Let the eigenvector be u 2  , and the corresponding eigenvalue be  . Then, 1 2  u 2  u 2 

OR

 3  2 1 0   u1 
 
 1 2    0 1  u   0
     2 

Which implies

3 2
0
1 2

OR  2  5  6  2  0

 2  5  8  0

5  25  32 5 i 7
  
2 2 2

5 7
1 i
Let =2 2 . Then, the corresponding eigenvector can be found:

35
 3  2  u1   5 7   u1 
1 2  u    2  i 2  u 
  2   2 

5 7
3u1  2u 2    i u1

 2 2  (i)
5 7
u1  2u 2    i u 2

 2 2 

5 7 1 7
2u 2  3u1    i u1    i
 2
u1

 2 2   2 
From (i),

11 7
 i 
2  2 2 
Thus, choosing u1 = 1, we get u 2 =  

 1 
 
Hence, an eigenfunction for the matrix is  1  i 7 

v 
v   1
Similarly, choosing v2  as the other eigenvector with a corresponding eigenvalue

v 
 2
5
i
7 v   1
2 2 , we can get the eigenvector v 2  .

4.2 Operators in Quantum Mechanics

Central to the theory of quantum mechanics is the idea of an operator (as we have seen earlier). We have
indeed come across some operators. Recall

Hˆ  ( x)  E ( x) 4.8

Where Ĥ is an operator For the time-independent Schrödinger equation:

 2 d 2 ( x)
Hˆ  ( x)    V ( x) ( x)
2m dx 2 4.9

Ĥ is the total energy operator or Hamiltonian.

36
We identify some other operators:

(i) The kinetic energy operator Tˆ

2 d 2
Tˆ  
2m dx 2 4.10

(ii) The linear momentum operator p̂

d
pˆ  i
dx 4.11

(iii) The position operator x̂

xˆ  x 4.12

4.3 Expectation value

The expectation value of a quantity is the statistical predicted mean value of all measurements. The


n
1
x
xi
x x x 
n
(statistical) average value of the numbers 1 , 2 , …, n is i 1 . However, if there is a
f x
distribution, such that there are i of the value i , i = 1, 2, …, n , then the average becomes
m

fx i i
1 n
x i 1
m
  f i xi m

 f
n i 1
fi i n
i 1 , since i 1 4.13

since n is the total number of observations.

In the case of quantum mechanics, the average value, or expectation value, of an operator is



    * ( x)() ( x)dx

4.14

Thus, the expectation value of x is




 x   * ( x) x ( x)dx

4.15

37
 2  2x 
 sin  
Thus, if L  L  , with n = 2, and 0  x  L ,

2x 2x

L
2
x sin x sin dx
L 0 L L 4.16

2x

L
2
x sin 2 dx
=L 0 L

2 L2 L
 
= L 4 2

The expectation value of the momentum for the same case above is

 d 
 p 

 * ( x)  i

 ( x)dx
dx  4.17

2x d  2x 

2i L
sin  sin dx
= L 0 L dx  L 

2i 2 2x 2x


L
 sin cos dx
= L L 0 L L

2i 2 4x

L
1
 sin dx
= L L 0 2 L

2i L 4x
L

 cos
L2
4 L 0
= =0

The energy expectation value of for the ground state of the simple harmonic oscillator:
  
1 
     dx  2 
1 1
 E   0* Hˆ  0 dx   0*   0  0   0 *
0 0 0
  2  2 
4.18

0
since is normalised.

This is a special case of the general result

38
  

  
    * ˆ dx    *dx  

4.19

Thus, we see that for any eigenstate of an operator, the expectation value of the observable represented by
that operator is the eigenvalue. More generally, we would write the expectation value of an operator, A, in
a certain state  , as

  | A |  .

4.4 Commutators and simultaneous eigenstates

Consider an operator P̂ that represents a physical observable of a system, e.g., energy or momentum.
ˆ
Suppose that the state  has a particular value p of this observable, i.e., P  p . Suppose further that
ˆ
the same state also has the value q of a second observable represented by the operator Q̂ , i.e., Q  q
. Then p and q are called simultaneous eigenvalues. Then,

Qˆ Pˆ   Qˆ p  pQˆ   pq 4.20

Similarly,

Pˆ Qˆ   Pˆ q  qPˆ   qp 4.21

Since p and q are just real numbers, then qp  pq . Thus, the condition for simultaneous eigenstates is
ˆˆ ˆˆ
that PQ  QP or

Pˆ Qˆ  Qˆ Pˆ  0 4.22

Pˆ Qˆ  Qˆ Pˆ is said to be the commutator of P̂ and Q̂ and operators that satisfy the condition
Pˆ Qˆ  Qˆ Pˆ  0 are said to commute. The commutator is normally written [ Pˆ , Qˆ ] .

When two operators commute, then they can have the same eigenstates as we have seen in this subsection
(notice that Q and P have the same eigenstate ).

39
In-text question

 1 2  1  2i 
   
2 1 1  1
1 3 2   1
1. The expectation value of a matrix operator,  in state   is

 1 2  1 2i   2i  3   4  6i 
      
 2i 1  1 2  1 1  1    2i 1  1 4i  2    4i  2 
 1 3 2   1  2i  1    2i  1
      

ˆ
2. Show that [T , pˆ ] .

Solution

2 d 2  d   d  2 d 2 
   i     i  
[Tˆ , pˆ ] = 2m dx 2  dx   dx   2m dx 2 

3 3
3 3
𝑖ℏ 𝑑 𝜓 − 𝑖ℏ 𝑑 𝜓 = 0
=
2𝑚 𝑑𝑥3 2𝑚 𝑑𝑥3

ˆ ˆ
3. Calculate [ x, p] .

Solution

 d   d 
[ xˆ, pˆ ]  x  i     i ( x )
 dx   dx 

d d
 ix  i  ix
= dx dx

= i

ˆ ˆ
Thus, we can write [ x, p] = i

Indeed, when two operators do not commute, it means that the two associated observables cannot be
measured with infinite accuracy simultaneously. Thus, an attempt to measure the momentum of a particle

x 
with infinite accuracy will cause an infinite error in the position as is easily seen in the equation, p
40
. On the other hand, the momentum and the energy of such a system can be measured simultaneously with
infinite accuracy. Other non-commutating operators include Ê and tˆ , i.e., the energy operator and the
time operator.

3. Write the function h( x)  e sin x as a sum of odd and even functions.

2x

Solution

h( x)  e 2 x sin x , h( x)  e 2 x sin(  x)  e 2 x sin x Therefore, the odd function is

h( x)  h( x) e 2 x sin x  e 2 x sin x e 2 x  e 2 x

f ( x)    sin x
2 2 2

 cosh 2 x sin x

The even function is

h( x)  h( x) e 2 x sin x  e 2 x sin x e 2 x  e 2 x

g ( x)    sin x
2 2 2

 sinh 2 x sin x

It is obvious that the odd function is a product of an odd function and an even function. Likewise, the even
function is a product of two odd functions. We conclude, therefore, that the following rules apply:

Even  Even = Even 4.23

Even  Odd = Odd 4.24

Odd  Odd = Even 4.25

The integral
𝑎
∫−𝑎 𝑓(𝑥)𝑑𝑥 = 0 If 𝑓(𝑥) is odd 4.26
𝑎 𝑎
∫−𝑎 𝑓(𝑥)𝑑𝑥 = 2 ∫0 𝑓(𝑥)𝑑𝑥 If 𝑓(𝑥) is even 4.27

Recall that the inner product in the space of twice integrable complex valued functions of two complex
b
( f , g )   f * ( x) g ( x)dx
valued functions f (x) and g (x) over the interval a  x  b is defined as a .

41
 Two functions f (x) and g (x) are said to be orthogonal over an interval a  x  b if their inner
product is zero.
2
 
4. Construct an orthonormal set from the set 1, x, x ,... over the interval  1  x  1 . Thus, given the set
 f1 , f 2 , f 3 ,..., we want to construct an orthogonal set 1 , 2 ,3 ,..., i.e.,
1
1
i ( x) j ( x)dx
= 0, if i  j , then we normalise each member of the set.

Let 1  f1  1 , and  2  f 2  1  x  

Then, we determine  , subject to

(1 ,  2 )  0

1 1
1 x2
11  ( x   ) dx 
2
 x 0
1 1

4.28

  0

Thus,  2  x

Let
3  f 3  2  1  x 2  x  

(1 , 3 )  0 ( 2 ,  3 )  0
subject to and

The first condition gives:

1
1
1  (x 2  x   )dx
=0 4.29

or
1 1 1
x3 x 2 2x 2
  x 1   2 x
1

3 1
2 1
3 0 =0 4.30

2
 2  0
3 4.31

42
or

1
 
3 4.32

The second condition gives

1 1
1
x  (x 2  x   )dx   ( x 3  x 2  x)dx
1

Or
1 1 1
x4 x 3 x 2
  0
4 1
3 1
2 1 4.33
1
2x 3
0
3 0

or

 0 4.44

Putting the values of  and  from equations (2.24) and (2.26) into the expression
3  f 3  2  1  x 2  x   , we arrive at

1
3  x 2 
3 4.45

 4 , 5 , etc., can be got in a similar fashion.

j
To normalise , we multiply the function by a normalisation constant, A , say, and invoke the relation
1
 A 2 j ( x)dx  1
2
1 4.46

For 1 , this becomes

1 1
 A 1 dx  2 A dx = 1
2 2 2
1 0

from which

2 A2  1

43
 1
A
or 2

1
1 
The normalised function 2 4.47

Similarly,
1
 1
A 2 x 2 dx  1

3 1
1 1 
2
x A
A 2
 A2     2 1
3 1
3 3 3

3
A2 
Thus, 2.

Hence, the normalised function,

3
2  x
2

In like manner,

2
1  2 1 1
2 4 2 2 1
1  3 
    0 A  x  3 x  9 dx
2
A x dx 2
=1

from which
1
x5 2x3 x
2A  
2
1
5 9 90

or

1 2 1
2 A2      1
5 9 9

80 2
A 1
Therefore, 45

45  2 1 
2  x  
The normalised function 8  3

44
3. Suppose we are given two vectors u and v. The projection of v onto u is indeed,

Pru v  ( v cos )uˆ

Where  is the angle between the two vectors, and û is the unit vector in the direction of u, and.
u uv
uˆ  cos  
u u  v  u v cos  u v
Therefore, . But . So,

Hence,

 uv  u (u, v )
Pru v   v 
 u  u2 u
(u, v )
u
 u v  (u, u) 4.48

since (u, u) | u | .
2

We proceed as we did in the case of the example from function space.

u1  v1 4.49

Pru1 v 2
The projection of v 2 onto u1 is . This is the component of v 2 in the direction of u1 . The
component of
v 2 perpendicular (i.e., orthogonal) to u1 is (vector subtraction):

u 2  v 2  Pru1 v 2
4.50

v3 u
Similarly the components of perpendicular to both u1 and 2 is,

u 3  v 3  Pru1 v 3  Pru2 v 3
4.51

In general then,
n 1
u n  v n   Prui v n
i 1 4.52

We can then normalise each vector

uk
ek 
|| u k || 4.53

Note that
Pru v projects vector v orthogonally onto vector u.

45
Summary of Study Session 4
In this study session, you have learnt:
1. The inner product of a pair orthogonal vectors is zero.
2. A basis that consists of orthogonal vectors only is an orthogonal basis.
3. With an orthogonal basis, we can define the probabilities of measurement.
4. The Gram- Schmidt orthonormalisation scheme can be used to create an orthogonal basis.

Self- Assessment Questions for Study Session 4

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 4.1

 y i
 
If the matrix  i 2 is Hermitian, find the value of y.

SAQ 4.2

 2   1   3 
     
  1,  0 ,  7 
 0    1   1
Show that the set       is a basis for the 3-dimensional Euclidean space. Show that the set
is not orthogonal. Hence, construct an orthogonal set from the vectors.

SAQ 4.3

Show that the set you obtained in SAQ 4.2 is not an orthonormal set, even though it is an orthogonal set.
Normalise each vector in the set. Hence, expand the normalised vector  1 2 1 / 6 in terms of this
basis. Calculate the probability that the system represented by the vector is in each of these states.

SAQ 4.4

A particle trapped in the well

0, 0  x  a
V 
, elsewhere

is found to have a wavefunction

46
 i 2  x  2  2 x  1  3 x 
sin   sin   sin  
2 a  a  3a  a  2a  a 

2 n x
sin
Given that the allowable wave functions are of the form a a , what is the probability of obtaining
each result?

SAQ 4.5

A particle in a one-dimensional box 0  x  a is in state:

1 x A 2 x 3 3 x
 ( x)  sin  sin  sin
5a a a a 6a a

(a) Find A so that  (x) is normalized.

(b) What are the possible results of measurements of the energy, and what are the respective
probabilities of obtaining each result?

47
Study Session 5 Inadequacies of Classical Mechanics and the Schrödinger
Equation

Introduction
If classical mechanics had no inadequacies, there would have been no need for a new theory. Up until the
turn of the century, it was thought that Newton’s laws could account for all physical phenomena,
irrespective of the size of the particle involved, and for any particle travelling at whatever speed.

By now, you are familiar with the basic ideas of classical mechanics, based on Newton’s laws of motion. It
would appear that once you know the equation of motion of a body, you can simultaneously and accurately
predict its position and linear momentum at any other time. Moreover, you would expect that an electron
confined within the walls of a finite potential well, provided the energy is less than the height of the well,
and would have no effect outside the borders of the well. Of course, the harmonic oscillator you came
across could have zero energy.

In this unit, you will get to know that matter behaves like wave or like a particle; that the highest velocity
with which photoelectrons emitted are from a photometal is independent of the intensity of the incident
radiation.

5.0 Objectives of Study session 5

1 After going through this Unit, you should be able to:

2 Discuss the phenomena that pointed to the wave picture and the ones that required a particle nature
of matter.
3 Appreciate the dual nature of matter: the wave and the particle pictures.

5.1 Main Content

5.2 Blackbody radiation

Everybody at a temperature higher than absolute zero radiates energy as the internal energy (thermal) is
converted to electromagnetic radiation. The spectrum of frequencies emitted by the body depends on its
temperature. Thus, at low energies, lower frequencies (or longer wavelengths) are predominant. At higher
temperatures, the higher frequencies (shorter wavelengths) are more prominent. It would be seen that the
peak of the spectrum of radiation moves towards the higher frequencies as the temperature of the body
increases. A blackbody is the perfect absorber and emitter of radiation. An example of a blackbody would
be a tiny hole in a heated cavity. Radiation falling on such a hole would be bounced up and down inside the
cavity, due to reflection, until all the radiation is absorbed. Before the time of Planck, Stefan and
Boltzmann came up with a formula for finding the total radiant energy emitted by a blackbody per unit
surface area per unit time, as

j*  T 4 5.1

Where  is the Stefan-Boltzmann constant, equal to 5.6704  10 Js m K , and T is the absolute

8 1 2 4

temperature of the body. With this law, Stefan was able to determine the temperature of the sun.

48
Rayleigh and Jeans predicted, based on the principle of equipartition of energy that the radiation emitted by
a blackbody per unit time, per unit area is given as

8 v 2 kT
u
c3 5.2

Where k is the Boltzmann constant, v is the frequency of the radiation and T is the temperature of the
body. However, neither the Stefan-Boltzmann nor the Rayleigh-Jeans formula could give the true picture
(experimental); that is, neither fitted the observed spectrum of radiation. The former agreed with
observation at high frequencies while the latter was fairly good for low frequencies.

By making the assumption that radiant energy could only be emitted or absorbed in quanta (or little packets
of energy), Planck arrived at the formula

8h v3
u
c 3 e hv / KT  1 5.3

This formula agrees with observation at all frequencies. Fig. 1 shows the spectrum for three temperatures:
2000 K, 3000 K and 4000 K. The figure also shows that the classical theory fits observation only at low
frequencies (long wavelength range).

7.E-16

6.E-16

5.E-16
Radiant energy (J)

2000 K
4.E-16
3000 K
4000 K
3.E-16
Classical
2.E-16

1.E-16

0.E+00
0.E+00 5.E-06 1.E-05 2.E-05
Wavelength (m)

Fig. 5.1: The spectrum of radiation for 2000, 3000 and 4000 K, and the classical result.

49
5.3 Photoelectric effect

Photoelectric effect is the phenomenon in which electrons are emitted from matter due to irradiation by
electromagnetic waves of suitable frequency. Classical mechanics predicts that the higher the intensity of
the incident wave, the higher the velocity of the emitted electron. Moreover, it expects any frequency of
light to emit electrons. This was, however, not observed in nature. To start with, the electrons were never
emitted unless the frequency of the incident wave was a certain value. In other words, there is a threshold
below which the incident radiation would not emit any electrons.

Einstein assumed that the incident radiation consisted of little packets of energy, quanta of light, called
photons, each of energy, 𝐸 = ℎ𝑓 where h is the Planck’s constant and v is the frequency of the incident
radiation. The assumption was that an electron would either absorb a photon or not. In the event that the
electron absorbs this energy, then, part of the energy would be used in overcoming the work function of the
metal, that is, the energy binding the electron to the metal. The excess energy would then be available for
the electron to move away from the metal. In other words, if the energy of the photon is just enough to
overcome the work function, the electron would be emitted, but would not have enough energy to leave the
surface of the metal. Thus, the formula guiding photoelectric effect is

1
E  hf    K max    2
mvmax
2 5.4

K max
is the maximum kinetic energy of an emitted electron. It follows that the maximum velocity an
emitted electron can have is

vmax  2(hf   ) 5.5

It is clear from equation 1.5, that for E  hf less than the work function, no electron is emitted, meaning
that, since h is a constant, if the frequency, f, is less than a threshold, no electron is emitted. Thus, the
maximum kinetic energy a photon can have, hf   , is independent of the intensity of the radiation, being
a function only of the frequency, provided the metal remains the same (i.e., the work function  remains
  hf
constant).We can also write the work function  as 0 , where f0
is the threshold frequency, below
which no electron will be emitted. We can therefore now writ

v  2h( f  f 0 )
5.6

f  f 0 , the threshold frequency. For a

From equation 1.6, it is clear that no electron will be emitted if
given metal and a given radiation, the intensity of the radiation is directly proportional to the number of
electrons emitted. This is because more electrons would be exposed to the radiation. The experimental set-
up for photoelectric effect consists of an emitter plate and a collector plate. What opposing potential would
stop the most energetic electrons from reaching the opposite plate of the collector.

50
 1 2
mvmax  eV0
2 5.7

Where 𝑒 is the charge on the electron and 𝑉0 is the stopping potential, since the potential energy just
balances the kinetic energy. Solving for, 𝑉0
2
mvmax
V0 
2e 5.8

5.4 Compton effect

Let a photon with wavelength  collide with an electron at rest. After the collision, the photon is scattered
in direction  relative to its initial direction. The electron is also scattered at an angle  with respect to the
reference direction as shown in Fig. … We shall apply relativistic conditions, meaning that the speeds
involved are large enough to be considered relativistic.

Fig. 5.2 Compton Scattering

Conservation of energy

E  Ee  E ' Ee'
5.9

hv  me c 2  hv' pe c 2  (me c 2 ) 2
2
5.10

since the relativistic energy-momentum relation

Ee'  pe c 2  (me c 2 ) 2
2 2
5.11
51
Conservation of linear momentum

p  p e  p'p e ' 5.12

pe  0
But . Hence,

p  p'p e '
5.13

From equation 1.10,

hv  m0 c 2  hv'  pe ' c 2  (me c 2 ) 2

2
5.14

Squaring both sides,

(hv  me c 2  hv' ) 2  pe' c 2  me c 4

2 2
5.15

Hence,

pe' c 2  (hv  me c 2  hv' ) 2  me c 4

2 2
5.16

From equation 1.13, we can write,

p e '  p  p'
5.17

Taking the dot product of both sides,

p e'  p e'  pe'  (p  p' )  (p  p' )  p 2  p' 2 2 pp' cos

2
5.18
2
Multiplying through with c ,

pe' c 2  p 2 c 2  p' 2 c 2  2 pp' c 2 cos

2
5.19

Setting pc  hv

pe' c 2  (hv) 2  (hv' ) 2  2(hv)(hv' ) cos

2
5.20

Equating the right sides of equations 1.16 and 1.20,

(hv  me c 2  hv' ) 2  me c 4
= (hv)  (hv' )  2(hv)( hv' ) cos 
2 2 2

(hv) 2  (me c 2 ) 2  (hv' ) 2  2(hv)(me c 2 )  2(hv)(hv' )  2(hv' )(me c 2 )  me c 4

2

 (hv) 2  (hv' ) 2  2(hv)(hv' ) cos 

52
Or

2(hv)( me c 2 )  2(hv' )( me c 2 )  2(hv)( hv' )  2(hv)( hv' ) cos 

5.21

from which we obtain,

hvme c 2  hv' me c 2  2h 2 vv'2h 2 vv' cos 

5.22

hvv' me c
Dividing through by ,

c c h
  (1  cos  )
v v ' me c 5.23

But c / v   and c / v'   ' . We can therefore write equation 1.23 as

h
   ' (1  cos  )
me c 5.24

5.5 Bohr’s Theory of the Hydrogen Atom

Rutherford tried to explain the structure of the atom by proposing a central nucleus, with the electrons
moving in orbits round the nucleus. However, quite a number of things were not right with his model. For
instance, you would have been taught before now, that an accelerating charge would radiate energy.
Remember that a body on curvilinear motion is accelerating even if the magnitude of the velocity remains
constant (centripetal acceleration). As such, you would expect an electron moving round the nucleus to
lose energy and subsequently fall into the massive nucleus. We therefore would have expected a
continuous spectrum of energy as against the line spectra obtained in reality. Bohr attempted to solve this
problem by making some postulates:

(i) An electron can only orbit the nucleus at certain radii, such that the condition mvr  n ,
where all symbols have their usual meanings. Thus, the radii are quantised; not just any radius
would do. While in such an orbit, the electron would not radiate energy.
(ii) Electrons can make a transition from one allowable orbit (or correspondingly, energy level) to
another by absorbing or emitting the difference in energy, as the case may be. That is, the
energy difference is
E f  E i  h

Where  is the frequency of the emitted photon,

Ei Ef
and being the initial and the final energy levels
respectively.

Taking into consideration the Coulombic force between the proton and the electron in the hydrogen atom,
Bohr arrived at the allowable radii of the form,

rn  Cn 2

53
Where C is a constant. Correspondingly, the energy levels are also of the form

D
En 
n2

Where D is a constant.

Clearly, you can see again the fact that quantisation enters into the picture.

Even though Bohr’s theory would also not meet the ultimate needs of physics, it afforded a bridge from
classical mechanics to quantum mechanics. Now you know that the electron in the atom is actually not to
be seen as a particle anymore, as it is as if a part of it resides in every region in the atom, aside the nucleus.
So, we see the electron as a cloud of sorts, and apply the laws of statistics in its distribution, especially in
view of the Heisenberg’s uncertainty principle which makes it impossible to simultaneously predict
accurately the position and the momentum of a partic5.6 Wave-particle duality

So far, we have considered radiant energy and photons as being quantised. On the other hand, according to
de-Broglie (1924), matter has a wave nature, with the wavelength given by

h  p 5.25

where h is the Planck’s constant, p is the linear momentum of the photon and  the associated
wavelength.

This formula was later confirmed experimentally via electron diffraction. Other phenomena that support
the wave nature of light include refraction, diffraction and interference.

From the foregoing, it is quite clear that matter behaves as if composed of particles or as a wave, hence the
term wave-particle duality.

5.6. Heisenberg’s Uncertainty Principle

Classical physics claims to know the future state of any particle, provided the position and the momentum
are known at any point. Heisenberg’s uncertainty principle excludes the possibility of simultaneously
measuring the position and the linear momentum of any object with infinite accuracy. Indeed, any attempt
to measure one of the quantities with infinite accuracy leads to infinite uncertainty in the other. Indeed, if
the uncertainty in position x and that of the linear momentum of the particle is  p , then,


xp 
2

Thus, our whole world of classical mechanics seems to have melted away. However, classical mechanics
has been tested and trusted for so long. Why does it now seem to fail? It is because we have been dealing
with matter on the large scale. Quantum mechanics makes the difference when we have to deal with small
objects, for example on the scale of atoms. Indeed, quantum mechanics has to agree with classical physics
in certain limits, in conformity with the correspondence principle.

54
In-text question

Q. Find the change in wavelength if a photon is scattered at an angle of 23 0 after its collision with
an electron initially at rest.

Solution
The change in wavelength is

h 6.626  10 34
   ' (1  cos  )  31
(1  cos 230 )  1.9294  10 13 m
me c 9.1  10  3  10 8

According to Classical electromagnetism, the wavelength of the scattered rays should remain the same, but
this did not agree with experiments. The latter found that the wavelengths of the scattered rays were greater
than the initial wavelength as borne out by equation 1.24.

Thus, for Compton Effect to be explained, we have to make recourse to radiation consisting of particles
that have energy and momentum.

Summary of Study Session 5

In this study session 5, you have learnt:
1 Classical mechanics is not adequate to describe motion at the atomic level.
2 Matter behaves like a wave.
3 Matter has the particle behaviour.
4 Indeed, matter exhibits the wave-particle duality.
5 A blackbody emits or absorbs energy in quanta.
6 Compton scattering is explained by the particle nature of light.
7 We need to assume that electromagnetic radiation consists of particles (photons) to explain
photoelectric effect.Self- Assessment Questions for Study Session 5

Self- Assessment Questions for Study Session 5

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

55
SAQ 5.1

Find the maximum kinetic energy with which an electron is emitted from a metal of work function
3.2  10 39 J when a radiation of energy E = 3.313  10 39 J falls on it, given that the work function is
3.2  10 39 J .

SAQ 5.2

What value does Rayleigh-Jeans formula predict for the radiation of frequency 6  10 Hz emitted by a
13

blackbody per unit time, per unit area at 2500 0K. Compare this value with that predicted by Planck.

REFERENCES/FURTHER READINGS

1. Introduction to Quantum Mechanics – David J. Griffiths.

2. Quantum Physics – Stephen Gasiorowicz.

56
Study Session 6 THE SCHROEDINGER EQUATION

Introduction
The fact that matter behaves like a wave implies that we would need an equation or a set of equations to
describe the wave behaviour of matter. As such, we would expect to have an equation describing the wave
nature of the electron. Likewise, the quantum-mechanical treatment of the harmonic oscillator would
involve an equation describing the wave behaviour of the oscillator. In this unit, we shall derive the
appropriate differential equations for the wave function, and consequently, describe the probability with
which the particle would be found at different points within the appropriate region of confinement, and
possibly beyond.

6.1 Objectives

At the end of this unit, you would be able to do the following:

1 Derive the time-dependent Schrödinger equation

2 Deduce the time-independent Schrödinger equation
3 Interpret the Schrödinger equation and its solutions

6.2 Main Content

We have seen in Unit 1 of this Module that matter behaves as a wave. Indeed, we noticed that the electron
in the hydrogen atom is best seen as a cloud. Moreover, you also saw that the wave-particle duality begs a
wave equation to describe the nature of particles, as it pushes our analysis into the statistical domain. The
Schrödinger equation does this job.

6.3 The Schrödinger Equation

We are quite aware that you know that the total mechanical energy of a body is the sum of the kinetic
energy T and the potential energy V. Of course, you also remember that the kinetic energy of a body of
mass m is

1 2 1 m2v 2 p2
mv  
2 2 m 2m 6.1

Then, the sum of its kinetic and potential energies, that is, the total mechanical energy is
p2
E  T V  V
2m 6.2

57
You surely recall that we can write p  k and E   . Substituting these into equation (6.2) give
2k 2
  V
2m 6.3

Let us consider the case of a free particle ( V  0 ). Then,

2k 2
 
2m 6.4

Let us try

( x, t )  Ae i ( kx t )

Then,

2
 ( x, t )  k 2 Ae i ( kx t )  k 2  ( x, t )
x 2


 ( x, t )  i Ae i ( kx t )  i   ( x, t )
t

2

Multiplying the first equation by 2m and the second by i gives

2  
 ( x, t )  i  ( x, t )
2m x 2
t

which is in line with equation (2.4).

For a particle that is not free, V ( x, t )  0 , and we arrive at

2  
 ( x, t )  V ( x, t )( x, t )  i ( x, t )
2m x 2
t 6.5

This is the time-dependent Schroedinger equation.

Now, suppose the potential is time-independent. Then,

2  
  ( x, t )  V ( x) ( x, t )  i  ( x, t )
2m x 2
t 6.6

Let

 ( x, t )   ( x) (t )

58
You would recognise this equation as showing that  ( x, t ) is variable separable, that is, we can write it as
a product of a function of x only, multiplied by a function of t only.
2  
 [ ( x) (t )]  V ( x)[ ( x) (t )]  i [ ( x) (t )]
2m x 2
t 6.7

2 d  d
  (t ) 2
 ( x)  V ( x) ( x) (t )  i ( x)  (t )
2m dx dt 6.8

Dividing through equation (2.8) by  ( x ) (t ) ,

1 2 d  1 d
  ( x)  V ( x)  i  (t )
 ( x) 2m dx 2
 (t ) dt 6.9

Notice that we could carry out this algebra only because the potential is a function of x alone. That is, the
potential is time-independent. You would also observe that the partial differentials have become ordinary
differentials since they now apply to a function of a single variable in each case. In equation 2.9, a function
only of x is equated to a function only of t. This means that they must be separately equal to a constant. We
know from the left part of equation 2.9, that the constant is the kinetic energy of the particle, E. Then,

1 d
i  (t )
 (t ) dt = E 6.10

We can rewrite this equation as

d E iE
 (t )  dt   dt
 (t ) i  6.11

Integrating,

iE
ln  (t )   tc
 6.12

where c is an arbitrary constant.

Taking the exponential of both sides,

 iE 
 (t )  exp  t  c
  

 De iEt /  6.13

59
Where D  e
c

But we can set D equal to unity as we shall be writing the wave function as  ( x ) (t ) , and in any case, the
wave function has not been normalised.

Hence, the time-dependent part of the wave function is

 (t )  e iEt /  6.14

The time-independent part of the equation is

2 d 
  ( x)  V ( x) ( x)  E
2m dx 2 6.15

Rearranging,

d 2m
2
 ( x)  2 [ E  V ( x)] ( x)  0
dx  6.16

This is the time-independent Schrödinger equation, applicable when the total (mechanical) energy of a
system is constant. We would not be able to solve equation 2.16 until we know the specific form of the
function 𝑉(𝑥).

We conclude that in the case where the potential is independent of the time,

( x, t )   ( x)e iEt /  6.17

6.4 Interpretation of the equation and its solutions

1. Generally, the solution  ( x, t ) of this equation is generally a complex function. You already know that
the magnitude of a complex function might not make any sense physically. But the square of the magnitude
(the intensity) certainly does, being always a positive real number. Indeed, it is a measure of the probability
 ( x, t )
2

of the particle being described arriving somewhere. is, therefore, a probability density. Of
 ( x, t ) dx
2

course, this implies that, is the probability of finding the particle between x and x  dx at
time t.  ( x, t ) is the probability amplitude and is the wave function. This is the function that describes the
behaviour of a particle as a function of x and t . This is the Born interpretation of  .

2. The solutions for equation 2.16 for a given functions 𝑉(𝑥) are possible for many solutions, each with a
corresponding energy E . This ensures that only certain values of E are allowed (energy quantisation).

3. for a system in which the particle is in one of these levels,

60
 ( x, t )   ( x)e iEt / 

2
 ( x, t )   ( x ) e  iEt / 
2

Then, since = 1 and the probability of finding the particle at any point is
independent of time. Such a state is called a stationary state.

4. Any function which satisfies the time-independent Schrödinger equation must be

(a) Single valued (b) continuous; (c) smoothly varying; and (d) tend to zero as

x   .

5. The probability of finding the particle along the chosen 1-dimension is 1, Hence,

  ( x) dx
2


= 1. 6.18

This is the normalisation condition, and it expresses the certainty (probability 1) of finding the particle
anywhere in the range    x  

In-text question
Q: Which of the following functions would you recommend as a possible eigenfunction in quantum
mechanics?

(i)  ( x)  e  ( x)  xe2 x
 x2
(ii)  ( x)  2 x (iii)
2

A:

(i) This is a bell-shaped function tends to zero as x tends to + infinity and – infinity. In addition, it
is twice differentiable, continuous, and normalisable.
(ii) The function is not normalisable even though it is continuous. This is because it continues to
grow as | x | becomes large.
(iii) This is also a valid function. As | x | grows, the exponential function tames the function,
preventing from tending to infinity. It is continuous and twice differentiable

Summary of Study Session 6

In this study session 6, you have learnt:
1. Schrödinger equation gives the dynamics of a quantum-mechanical particle.
2. The time-independent Schrödinger is applicable when the potential is time-independent.

61
3. The square of the amplitude of the solution of the Schrödinger equation gives the probability density.
4. The normalisation condition requires that the integral of the probability density over the whole of the
relevant range equals unity.

Self- Assessment Questions for Study Session 6

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 6.1

By solving the time-dependent Schrödinger equation for a free particle (V = 0), find the condition

imposed on the angular frequency and the wavenumber.

SAQ 6.2

By applying the method of separation of variables, solve the differential equation

 2 2 2 
 2  2  2 
 x y z 
=0

SAQ 6.3

𝑛
What would the potential function be if 𝜓(𝑥) = (𝑥𝑥 ) 𝑒−2𝑥⁄𝑥0 is an eigenfunction of the Schrödinger
0

equation ? Assume that when 𝑥 → ∞ , 𝑉(𝑥) → 0

References/Further Readings
1. Quantum Mechanics demystified - David McMahon.
2. Introduction to Quantum Mechanics – David J. Griffiths.
3. Quantum Physics – Stephen Gasiorowicz.

62
Study Session 7 Postulates of Quantum Mechanics
Introduction
So far, in the first units of this module, we have derived the Schrödinger equation, and have also
interpreted its solution. In this unit you shall learn the postulates of quantum mechanics. You shall also
learn how to expand a given wave function in terms of a complete set of orthonormal eigenstates, which
are the states in which the system can possibly exist, recover the coefficient of each eigenstate in the
expansion, and then determine the probability that it is indeed in any given state. It is necessary to know
what value you should expect when you carry out a measurement of a physical observable. This you shall
learn with appropriate examples in this Unit as well. You shall also get to know in what limit quantum
mechanics must agree with classical mechanics through the knowledge of the Correspondence Principle.

Learning Outcomes for Study Session 7

When you have studied this session, you should be able to:
By the time you are through with this Unit, you should be able to:

1. Expand a given wave function as a linear combination of the possible orthonormal eigenstates
2. Recover the coefficient of each eigenstate in the expansion
3. Calculate the probability that the system is in any given eigenstate
4. Find the (statistical) expectation value of a physical observable in any
given eigenstate of the system
5. Learn the limit in which quantum mechanics should conform to classical mechanics

7.1 Postulates of Quantum Mechanics

A postulate is a something that is assumed to be self-evident, requiring no proof, used as a basis for
reasoning. The postulates of Quantum Mechanics are the minimum conditions that must be satisfied for
Quantum Mechanics to hold. If Quantum mechanics works based on these postulates, it means that the
postulates are true.

(1) There exists a function, called state  ( x, t ) . This gives the state of the system. It contains all that are
relevant mechanically in the system. In other words, it gives the spatial and temporal (space and time)
evolution of the quantum-mechanical particle.
 is single-valued in x , differentiable in x and t ;  ( x, t ) is square integrable in x .

(2) To every measurement (of an attribute of a system) there is a linear self-adjoint (Hermitian)
operator, called an observable, A, operating on the wave function. E.g., A   . That is, the operator
acting on a state function  produces a state function  . If  is a multiple of  , say
    , then,
A    . Then  is an eigenvector and  an eigenvalue of A. The eigenvalues { } are the values

63
physically obtainable in measuring the attribute that has A as its observable. We therefore expect the  ’s to
be real. Hermitian operators are associated with real observables because they need to have real
eigenvalues (possible measurements).

For example, the operator corresponding to the energy of a system is the Hamiltonian, which is
2 d 2
  V ( x, t )
written as 2m dx 2 . There are two operators involved, and you would remember that
in Classical Mechanics, the two parts of the Hamiltonian are the kinetic energy and the potential
2 d 2

energy. Clearly, you can see that the kinetic energy operator in Quantum Mechanics is 2m dx 2 ,
and the potential energy operator is V ( x, t ) . But do you also remember that classically we write
p2
the kinetic energy as 2m ? Can you then see that in Quantum Mechanics, comparing the two
d
p  i 
expressions, we must write dx . Square this expression and confirm that indeed it gives
d2
 2
you dx 2 . Applying the energy operator to the state function, we get,

 2 d 2 
 2
 V ( x, t ) ( x, t )  E( x, t )
 2m dx  (7.1)

E
Depending on the potential V ( x, t ) , we could get the possible values of the energy, i , of the
particle. This is the idea behind quantisation.

{ j }
(3) Let form a basis of H . Then,
n
   c j j
j 1
c j  ( j ,  )
, with (7.2)

There are many possible states for a quantum-mechanical system, usually infinite in number with n in
equation 3.2 tending to infinity. Hence, the wave function  ( x, t ) is a linear combination of all these

possible states. This means the system described by  can be found in any of the possible states j . We
ci  (i , )
can recover the coefficient of each of the possible states using equation . Once we have done
ci
that, that is, we have found , we could find the probability that the system is in that particular state (i),
| ci | 2
{ j }nj1
, provided the wave function  is normalized, and the set is an orthonormal set. Refer to
Unit …

64
(4) The expectation value of any property w(x) can be found by obtaining the expectation value of the
corresponding operator with respect to the wave function.


 w    * ( x) wˆ  ( x) dx


Where ŵ is the operator associated with the physical quantity w.


(5) Reduction Doctrine: If we make a measurement in state  and obtain k out of all the possible
{ j }nj1 j
states , then if immediately after that measurement we make another, we obtain with
| ck | 2

probability = 1, i.e., the process of measurement reduces  to k .

(6) The time evolution of the wave function is given by the equation


i  ( x, t )  H  ( x, t )
t (7.3)

In-text question
2
 ( x, t  0) 
x
1. An electron whose normalized state function at time t = 0 is found to be L2 is
0, 0  x  L
V ( x)  
confined within the infinite time-independent potential well given by , elsewhere

2 n x
 n ( x)  sin
The allowed eigenstates are L L for 0  x  L ; n = 1, 2, 3, … The coefficient, c m , of
8 2
  2 2
 m in the expansion in the normalised  (x) is L m .

(i) Find the probability that the electron is in state m  2 .

(ii)What is the expectation value of the displacement and the linear momentum in state n = 3?

2 2  2
2
(iii)The energy is measured and found to be ma . What is the state of the system immediately after
measurement?

(vi)Time evolution

65
Solution
(i) The probability that the electron is in state m = 3:
8 2
cm    2 2
L m

8 2 32
c3    2 2 
L 3  81L 4

32
2
The probability that the electron is in state m = 3 is 3 = 81L
|c | 4

(ii) The expectation value of the displacement in state m = 3 is (since the operator for x is x̂ = x):

 x   3 * ( x) x3 ( x)dx


In this case,
 n is non-zero only between 0 and L . Hence, the integral becomes

*
L  2 3 x   2 3 x 
 x   
 L sin L 
 x sin dx
0
   L L 

2 L 3 x
= L
 0
x sin 2
L
dx

2 L 1  3 x 
L 0
x  1  cos 2
2   L
 dx
  (since
1
sin 2   [1  cos 2 ]
2
= )

1 L   6 x 
L 0 

x  x cos
 L
 dx

=

1 L
= L
0 xdx (the second integral is zero; can you show it?)
L
1 x2 L2 L
 
L 2 2L 2
= 0

d
pˆ  i
The expectation value of the linear momentum in state m = 3 is (since the operator for p is dx ):

  d 
 x    3 * ( x)  i  3 ( x)dx

 dx 

66
In this case,
 n is non-zero only between 0 and L . Hence, the integral becomes
*
L  2 3 x   d  2 3 x 
 p   
 L sin 
   i  sin dx
0
 L   dx  L L 

2 L  3 x   3  3 x 
*

L 0 
(i)  sin    cos dx
= L   L  L 

6 L 3 x 3 x
=
 i
L2  0
sin
L
cos
L
dx

6 L 1  3 x 

 i cos 2 1
dx sin  cos   cos 2
= L2 0 2  L  (since 2 )

3 L  6 x 
 i
L2  0
cos
 L 
dx
=

L
3 L  6 x 
 i 2  sin   0
L 6 x  L  0
=

(iii)The measured energy is equivalent to the energy state n = 2, since we can write it as
2 2  2  2 n 2 2  2

2ma 2 2ma 2 . Therefore, from the reduction principle, immediately after the measurement,
the electron will be found with probability 1 in the state n = 2, that is with eigenfunction
2 2 x
2  sin
L L .

(iv)We can therefore write the wave function as


2 j x
 ( x,0)   c j sin
j 1 L L

At time t  0 ,


2 j x iEn t / 
 ( x, t )   c j (0) sin e
j 1 L L

7.2 The Correspondence Principle

The correspondence principle states that as the quantum number n becomes large, quantum mechanics
should approximate classical mechanics. For a new theory to be acceptable, it must conform to the well-
tested existing theories. In this vein, the special theory of relativity, which is important only when the

67
velocities involved are large, must conform to Newtonian mechanics when the velocities involved are
small day to day values. As an instance of this law, we consider the Hydrogen atom as treated by Bohr.
The frequency of the radiation emitted or absorbed in a transition between states n and n' is,

 1 1 
E nn'  RE  2  2 
 n' n  (7.4)

where R E is the Rydberg energy.

Now, classical mechanics predicts a continuous spectrum, while Bohr’s theory gives rise to discrete lines.
Let us take n' as being equal to n  1. Then,

 1 1 
Enn'  RE   2 
 (n  1) n  RE [(n  1) 2  n 2 ]
2
= (7.5)

Applying Binomial expansion,

RE [(n  1) 2  n 2 ] = RE [n 2  2n 3  (2)(3)n 4  ...  n 2 ] (7.6)

Hence,

2
 RE
E nn' n3 (7.7)

This tends to zero as n tends to infinity. We conclude therefore, that as n tends to infinity, the spectrum
becomes continuous, as predicted by classical mechanics.

Summary of Study Session 7

In this study session, you learnt the following:

1 The postulates of quantum mechanics

2 How to expand a given wave function as a linear combination of the possible orthonormal
eigenstates
3 How to recover the coefficient of each eigenstate in the expansion
4 How to calculate the probability that the system is in any given possible eigenstate
5 How to find the (statistical) expectation value of a physical observable in any given eigenstate of
the system
6 Learn the limit in which quantum mechanics should conform to classical mechanics

68
Self- Assessment Questions for Study Session 7
Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 7.1

A particle of mass m is confined within a one-dimensional box of length L / 2 , subject to a potential:

0, 0  x  L / 2
V ( x)  
, elsewhere

, the wave function is  ( x)  Ax( L  x / 2) , i.e.,  ( x,0)  Ax( L  x / 2) ,

t0
If at

(i) Normalise , and hence, determine the value of A.

(ii) Write  ( x, t ) as a series, where t  0 .

(iii) Write an expression for the coefficients in the series.

SAQ 7.2

 ( x)  A(ax  x 2 ) For | x | a . Normalise the wave function and find  x  ,  x 2  and x .

SAQ 7.3

A particle is confined within a one-dimensional region 0  x  L . At time t = 0, its wave function is given
  x  x
A1  cos  sin L
as  L  .

(i) Normalise the wave function.

(iii) Find the average energy of the system at time t = 0 and at an arbitrary time 𝑡0.
(iv) Find the average energy of the particle.
(v) Write the expression for the probability that the particle is found within 0  x  L / 2 ?

69
SAQ 7.4

A particle trapped in the well

0, 0  x  a
V 
, elsewhere

is found to have a wave function

i 2  x  2  3 x  2  3 x 
sin   sin   sin  
2 a  a  3a  a  16a  a 

(a) If the energy is measured, what are the possible results and what is the probability of obtaining
each result?
(b) What is the most probable energy for this particle?
(c) What is the average energy of the particle?

References/Further Readings
1. Quantum Mechanics demystified - David McMahon.

2. Introduction to Quantum Mechanics – David J. Griffiths.

3. Quantum Physics – Stephen Gasiorowicz.

70
Study Session 8 Time-Independent Schrödinger Equation in One Dimension I

Introduction
In Study session 2, we were able to review the inadequacies of classical mechanics, and found it necessary
to evolve a wave description of a particle. In Study session 2, we derived the differential equation that
describes the wave function associated with a particle. We have now been provided the tool needed to
describe the quantum-mechanical dynamics of a particle. We shall take a look at the particle in different
types of wells, and then compare our results with those of classical mechanics.

Learning Outcomes for Study Session 8

When you have studied this session, you should be able to:
1 Explain what is meant by bound states
2 Sketch the infinite potential well
3 Write and solve the Schrödinger equation for the infinite potential well
4 Apply the necessary boundary conditions to obtain the condition for bound states.

8.1 Bound States

You would recall that a body in simple harmonic motion bounces back and forth between the two points
where the total mechanical energy, E, of the body is equal to the potential energy. As the kinetic energy
becomes zero, the body must turn back. In just the same way, a quantum-mechanical oscillator is a particle
inside an infinite potential well. You shall see that indeed, the solutions will be sinusoidal, just the way it is
with a harmonic oscillator. Such a state is an example of a bound state. More specifically, we shall say a
system is in a bound state if, E  V () . As such the wave function involved must die at infinity, that is,
 ( x)  0 as x   . E  V () or/and E  V () is called a scattering state. Another way of
seeing a bound state is to say that the particle is subjected to an attracting potential.

For a bound state, the following conditions apply:

(i)  is continuous across the boundary

(ii) The first derivative, ' , is continuous across the boundary

71
8.2 Particle in an infinite potential well

This is also called a particle in a box

∞ ∞

x=0 x=L

Fig. 8.1: The infinite square well potential confining a particle to a region of width L

Inside the well, the potential is zero. Outside the well, the potential is infinite. We expect that the wave
function outside the well will be zero.

We recall Schrodinger equation

 2 2m
 ( E  V )  0
x 2  2 .

When V = 0,

d 2 2m
 E  0
dx 2  (8.1)

Which we can write as

d 2
2
 k 2  0
dx (8.2)

The general solution of this equation is,

 ( x)  c1e ikx  c2 e ikx (8.3)

Where 𝑐1 and 𝑐2 are constants to be determined, subject to the boundary conditions, and i   1 . (In the
case of the infinite potential well, the condition of continuity of the first differential is redundant, and you
should know why)

Since

(i)  (x ) = 0 for x  0

72
  (0)  c1e ik 0  c2 e ik 0
=0

 c1  c2  0  c1  c2  c

  ( x)  c(e ikx  e ikx )  A sin kx ,

Where A  2ic .

(ii)  (x ) = 0 for x  L

Asin kL  0  kL  n , where n = 0, 1, 2 …

n
kn 
It follows that the values of k are quantised, such that L .

2
 2kn n2h2
En  
Thus, the corresponding energy 2m 8mL2 , and the wave function corresponding to this energy
nx
 n ( x)  An sin
is L .

n=1

Fig. 8.2: The lowest three wave functions (for n = 1, 2 and 3) in the infinite square well potential.
73
Notice that the energy increases with n as n , i.e., for n = 1, 2, 3, …, E = E1 , 4E1 , 9E1 , …
2

E1 is the energy of the ground state ( n = 1) and E2  4E1 is the energy of the first excited state ( n  2 ).

Fig. 8.3: The probability of finding a particle at different values of x for the energy levels n
= 1, 2 and 3 in the infinite square well potential.

An
, called the normalisation constant is obtained by applying the normalisation condition:


  ( x) dx  1
2


(8.4)

i.e.

74
 nx

L
dx  1
2
An sin 2
0 L (8.5)

1
An  1
2

or 2 ,

from which it follows that,

2
An 
L.

We can therefore write

2 nx
 n ( x)  sin
L L for 0  x  L ; n = 1, 2, 3, … (8.6)

You can now see that the Eigen states we dealt with so extensively in study session 2 are the Eigen states
allowable inside an infinite potential well.

You would also notice that very much like harmonic oscillation in the classical case, the allowable
wavefunctions are sinusoidal.

8.3 The Finite Potential Well

In the last section, we dealt with the case of an infinite potential well, with the resulting wave functions
being zero at the boundaries. In the case of a finite potential well, the wave functions “spill over” into the
region outside the potential well. Let us see how.

Region I Region II Region III

x  L / 2 x  L/2

Fig. 8.4: The finite potential well of depth 𝑽𝟎 and width L .

When𝐸 < 𝑉0, a particle in the potential is said to be bound. Compared to the infinite potential well, it is
more convenient to take the centre of the well as 𝑥 = 0.
75
Solutions:

d 2 II 2mE
 2  II  0
, which can be written (where k  2mE /  ) as
2 2 2
Region II: dx 

d 2 II
2
 k 2 II  0
dx
(8.10)

The solution can be written in the form:

 II  A sin kx  B cos kx (8.11)

This consists of an odd and an even solution.

We first consider the even solution:  II  B cos kx

d 2 I 2m(V0  E )
2
 I  0  2  2m(V0  E ) /  2
Region I: dx 2 , which can be written (where

as

d 2 I
2
  2 I  0
dx (8.12)

The solution can be written in the form:

 I  Cex  De x (8.13)

However, we cannot allow an exponentially growing term; so we set D equal to zero.

 I  Cex (8.14)

d 2 III 2m(V0  E )
2
  III  0
Region III: dx 2 , which can be written as

d 2 III
  2 III  0
dx 2 (8.15)

The solution can be written in the form:

 III  Fex  Gex (8.16)

However, we cannot allow an exponentially growing term; so we set F equal to zero.

76
 III  Cex (why C again?) (8.17)

The solutions  and d / dx in the three regions can be written as:

Region I Region II Region III

 Cex Bcos kx Ce x

 ''  Cex  kBsin kx   Ce x

At x  L / 2 , both  and  ' ' are continuous. Thus, at x  L / 2 , we have

kL
B cos  Ce L / 2
2 (8.18)

and

kL
kB sin   Ce L / 2
2 (8.19)

Dividing one by the other, we get:

kL 
tan 
2 k (8.20)

This is the equation that determines the values of k and hence the energy E for even parity solutions.

For the odd parity solutions,

Region I Region II Region III

 Cex Asin kx  Ce x

 ''  Cex kAcos kx  Ce x

77
At x  L / 2 , both  and  ' ' are continuous. Thus, at x  L / 2 , we have

kL
A sin  Ce L / 2
2 (8.21)

and

kL
kAcos   Ce L / 2
2 (8.22)

Dividing one by the other, we get:

kL 
cot 
2 k (8.23)

This is the equation that determines the values of k and hence the energy E for odd parity solutions.

In Text Question

Let the total wave function of the particle in the potential well above be   Dx . Where D is a
normalization constant. Find the probability that the particle is in state 0, 1, 2 and 5.

Solutions

The first thing to do is to normalise the wave functions 𝜓𝑛(𝑥), which we have done. Next, we normalise
.

L
 x2  L2

L
Dxdx   D   D 1
0  2 0 2

2
D
Therefore, L2 , implying that

2
 ( x)  x
L2

Then, the total wave function can be expanded as a linear combination of the set of wave functions
 n (x) ,


 ( x)  c 
n 1
n n ( x)
as

78
  2 nx 2 
c n  ( n ,  )   cos , x 
 L L L2 
The coefficient

nx

L
2 2
 x cos dx
= L L2 0 L

Integrating by parts,

2 2   L nx  nx 
L


L
L
cn   x sin  sin dx
L L2   n L  0 0 n L 


nx
L
2 2 L2
  2  2 2   cos
L L n L 0
=


8
 2 2  cos n  cos 0
L n
1

8 1

 2 2 1  (1) n
L n

= =

8 2
  2 2
= L n  for n odd and 0 for n even

Could you comment on this result? The probability of finding the particle in a particular eigenstate reduces
as n increases. Thus, it is more likely that you find the particle in a lower eigenstate than in a higher one.
You may also compare this expression with that of the hydrogen atom according to Bohr. Can you write an
expression for the probability of finding the particle in state n?

Summary of Study Session 8

In this study session 8, you have learnt:
1 What is meant by a bound state, and what kind of potential can give rise to it.
2 How to solve the time-independent Schrödinger equation for both the finite and the infinite
potential well.
3 How to apply the appropriate boundary conditions in order to get the conditions for bound states.

79
Self- Assessment Questions for Study Session 8
Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 8.1

What are the allowable Eigen functions and energy eigenvalues of the infinite potential well?

0,  L  x  L
V ( x)  
, elsewhere

References/Further Readings

1. Quantum Mechanics demystified - David McMahon.

2. Introduction to Quantum Mechanics – David J. Griffiths.

3. Quantum Physics – Stephen Gasiorowi

80
Study Session 9 SCATTERING STATE

Introduction
In study session 8, we discussed the infinite and the finite potential wells. In this Unit, we shall consider
two cases: the potential step and the potential barrier for the case where the energy of the particle is less
than the potential and when it is higher. These are the scattering cases, as we have mentioned in Unit 1.
Rather than bound states, we are concerned in these cases with reflection and transmission coefficients. Yet
again, we shall compare our results with those of the equivalent classical case.

Learning Outcomes for Study Session 9

When you have studied this session, you should be able to:
1 Write the Schrödinger equations for the scattering states
2 Solve the appropriate Schrödinger equations
3 Apply the boundary conditions
4 Obtain the reflection and the transmission coefficients

9.1 The Potential Step

We have investigated the case of the infinite and the finite potential well. What happens when a particle
approaches a ‘bump’ of potential as indicated in Fig. 9.1 with energy E.

We can write this potential as

………………………..

E V0

x=0

Fig 9.1 Potential Step

0, x  0
V 
V0 , x  0

Let us consider the two possibilities: 𝐸 > 𝑉0 and 𝐸 < 𝑉0. The former is such that the energy of the
incident particle is higher than the potential step. Classically, the particle will pass the potential step and
move on to the right, past x = 0. In the case 𝐸 < 𝑉0, classical mechanics predicts that the particle will be
reflected to the left at 𝑥 = 0. We now take these cases one at a time.

81
Case 1: 𝐸 > 𝑉0

Then, Schrödinger equation for 𝑥 < 0 is

d 2 L 2mE
 2 L  0
dx 2  , the subscript L denoting the region to the left of x = 0.

The solution to this equation is of the form:

 L  Aeikx  Be ikx 9.1

Where

2mE
k
2 9.2

For the region to the right of x = 0,

d 2 R 2m( E  V0 )
 R  0
dx 2 2
9.3

and we can write the solution as

x  Ce i x  De i x 9.4

Where

2m( E  V0 )

2 9.5

From our knowledge of classical mechanics, we would expect perfect reflection at the barrier. We can take
 i x
the incident wave as being Ae for x  0 from the left and De for x  0 from the right, since the
ikx

latter originated from the potential step and is traveling to the left. Experimentally, we should have a
source to only one side of the step. If we assume that the incident wave is from the left, then we have to set
 ikx i x
D = 0. In that case, Be is the wave reflected at x = 0 and Ce is the wave transmitted at x = 0. We can
now write the following:

82
 inc  Aeikx 9.6

 ref  Be ikx
9.7

 trans  Cei x 9.8

Now we match these solutions at x = 0, that is, both  and  ' are continuous at this point.

: Ae ik ( x 0)  Be ik ( x 0)  Ce i ( x 0)

 A B  C 9.9
i ( x  0 )
 ': ikAe ik ( x  0 )
 ikBe  ik ( x  0 )
 i Ce

 ikA  ikB  i C 9.10

From equation 2.10,

k ( A  B)   C
9.11

Putting equation 2.9 in equation 2.11,

k ( A  B)   ( A  B) 9.12

Hence,

kA  kB   A   B 9.13

or

(k   ) A  (k   ) B 9.14

 k  
B A
We conclude, therefore, that  k   9.15

83
 and

k  kA   A  kA   A  2k 
C  A A  A
k  k   k   9.16

Since the flux intensity is proportional to the product of the velocity and the probability, we make the
following identification:

v1Pinc
Incident flux is proportional to

v1 Pref
Reflected flux is proportional to

v2 Ptrans
Transmitted flux is proportional to

Pinc   inc
*
 inc  ( Aeikx ) * ( Aeikx )  A * e ikx Aeikx | A |2
9.17

Similarly,

Pref | B |2
9.18

and

Ptrans | C |2
9.19

The transmission coefficient is

2
 2k 
v2   | A|
2
2
k  
 
transmittd flux v 2 Ptrans v |C |
T   2 2
incident flux v1 Pinc v1 | A | v1 | A | 2 9.20

v2 4k 2
T
v1 (k   ) 2
Hence, 9.21

The reflection coefficient is,

 k  
2

  | A|
2
2
 
 
reflected flux v1Pref | B | k
R  
incident flux v1Pinc | A |2 | A |2
9.22

or
84
 (k   ) 2
R
(k   ) 2 9.23

1 2
2m mv1
2mE 2 mv
k 2
 2
 1
  
9.24

1 2
2m v2
2m( E  V0 ) mv
  2  2
2  2
 9.25

k v1

 v2 9.26

Adding,

(k   )2 v2 4k 2 (k   )2  4k 2 (k   ) 2 4k
R T      
(k   ) 2
v1 (k   ) 2
(k   ) 2
k (k   ) 2
(k   ) (k   ) 2
2

k 2   2  2k  4k k 2   2  2k

 2 1
= (k   ) 2 k   2  2k 9.27

implying the conservation of particles across x = 0.

E  V0
Note: Since k   , R  0 . This implies that even particles with could be reflected. Classical
mechanics predicts that as the particles with energy 𝐸 > 𝑉0cross the point x  0 into a region of finite
V
potential 0 , no particle should be reflected. They should only be slowed down, as part of their energy has
been used to overcome the potential. This indeed is why your simple pendulum slows down as it
approaches a region where 𝐸 = 𝑉0 ., and then changes direction as soon as it gets to the point 𝐸 = 𝑉0..
That is why it oscillates between the two positions where 𝐸 = 𝑉0. Thus in classical mechanics no particle

with energy greater than

V0 will be reflected, and R would have been zero. In the limiting case where E
V0 k
is just slightly greater than , is almost the same as and R will be close to zero, meaning that only a
very small number of particles will be reflected. The fact that some particles are reflected even when
E  V0
is due to the fact that particles have a wave nature. You would expect part of a wave to be reflected
at such a point, wouldn’t you?

85
Case 2: 𝐸 < 𝑉0.

Equations 2.1 and 2.2 remain unchanged.

Then, Schrödinger equation for x  0 is

d 2 L 2mE
 2 L  0
dx 2  9.28

The subscript L denoting the region to the left of x = 0.

The solution to this equation is of the form:

x  Aeikx  Be ikx 9.29

Where

2mE
k
2 9.30

For the region to the right of x = 0,

d 2 R 2m(V0  E )
 R  0
dx 2 2
9.31

and we can write the solution as

x  Ce x  De   x 9.32

Where

2m(V0  E )

2 9.33

In this case, we have to set C  0 as we do not want solutions that grow exponentially as they cannot be
normalized.

In this case,

 inc  Aeikx
9.34
86
 ref  Be ikx
9.35

 trans  De   x 9.36

Now we match these solutions at x = 0, that is, both  and  ' are continuous at this point.

 : Aeik ( x  0)  Be ik ( x  0)  De  ( x  0)  A B  D 9.37

 ' : ikAeik ( x  0)  ikBe ik ( x  0)   De   ( x  0)  ikA  ikB   D 9.38

From equation 2.38,

k ( A  B)  i D (as 1/ i  i ) 9.39

Putting equation 2.37 in equation 2.39,

k ( A  B)  i ( A  B) 9.40

Hence,

kA  kB  i A  i B 9.41

or

(k  i ) A  (k  i ) B 9.42

We conclude, therefore, that

 k  i 
B    A
 k i  9.43

and

k i kA  i A  kA  i A  2k 
D  A A    A
k i k  i  k  i  9.44

Since the flux intensity is proportional to the product of the velocity and the probability, we make the
following identification:

Pinc | A |2
9.45

87
Pref | B |2
9.46

and

Ptrans | D |2 9.47

The reflection coefficient is,

2
k  i
| A |2
reflected flux v1Pref | B | 2
k i
R   
incident flux v1Pinc | A |2 | A |2 9.48

or

| k  i |2 (k  i ) * (k  i ) (k  i )( k  i )
R  1
| k  i |2 =
(k  i ) * (k  i ) (k  i )( k  i )
9.49

Thus, all the incident particles are reflected. However, the probability of finding the particle at a point a
𝑥 > 0 is

| De   x |2 | D |2 e2  x 9.50

This is finite for finite x. In the classical situation, the particles are definitely reflected at 𝑥 = 0. That is
also what ‘common sense’ would predict. But the fact that particles behave also like waves necessitates
that ‘some of the wave’ would be found beyond 𝑥 = 0 .However, note that the probability decays
exponentially, so that the probability is finite only if the distance 𝑥 > 0. If 𝑉0 ≫ 𝐸, then 𝛽 is very large,
meaning that the decay of the probability will be much faster. In the extreme, limiting, case, where 𝑉0 tends
to infinity, there is perfect reflection at the barrier, just as you had it for the infinite potential well.

9.2 The Potential Barrier

We have seen the finite and the infinite potential well, as well as the potential step. Now, what happens if
the potential is actually a rectangular barrier of height 𝑉0 of width a as shown in Fig. 9.2.

88
 …………………………….......

V0

𝑥=0 𝑥=𝑎

Fig. 9.2: The potential barrier

We identify three regions, one to the left of 𝑥 = 0 (region I), between 𝑥 = 0 and 𝑥 = 𝑎 (region II), and
beyond 𝑥 = 𝑎 (region III).

More compactly, we would write this as

0, x  0

V  V0 , 0  x  a
0, x  a


Region I:

d 2 2m
 E  0
dx 2  2 9.51
d 2 2m
 ( E  V0 )  0
dx 2  2 9.52

The solutions are,

 I  Aeikx  Be ikx 9.53

 II  Cei x  De i x

 III  Eeikx  Feikx 9.54

89
 2mE 2m( E  V0 )
k 2

where  , and 2

2m(V0  E )

E  V0
Bear in mind that if , we can write   i , where 2 9.55

Matching the solutions  I and  II at x  0 ,

 I (0)   II (0)  A B  C  D 9.56

d d
I  I
dx x 0 dx x 0  ik ( A  B)  i  (C  D) 9.57

 II (a)   III (a)  Cei a  De i a  Eeika 9.58

d d
 II   III i a i  a
dx xa dx xa  i  Ce  i De  i k Eeika 9.59

Multiplying equation 3.6 by ik , we obtain

ikA  ikB  ikC  ikD 9.60

Adding equations 3.7 and 3.10,

2ikA  ikC  ikD  i  C  i D  i(k   )C  i (k   ) D 9.61

Hence,

k  k 
A C D
2k 2k 9.62

Multiply equation 3.8 by i and add to equation 3.9.

2i Cei a  i Eeika  ikEeika  i (  k ) Eeika

or

  k i ( k  ) a
C Ee
2 9.63

90
We can also get D as follows:

Multiplying equation 3.8 by  i and adding to equation 3.9.

 2i De i a  i(k   ) Eeika 9.64

  k i ( k  ) a
D Ee
2 9.65

Putting equations 3.13 and 3.14 into equation 3.12, we conclude that,

k     k i ( k  ) a k     k i ( k   ) a
A Ee  Ee
2k 2 2k 2 9.66

Therefore,

A (k   ) 2 i ( k  ) a (k   ) 2 i ( k  ) a
 e  e
E 4k 4k
9.67

Or

A (k   ) 2 i ( k  ) a (k   ) 2 i ( k  ) a
 e  e
E 4k 4k 9.68

V0  E
We have assumed that , and this resulted in some wave being transmitted beyond the point x = 0.

V0  E   i
For , , meaning that, from equation 3.18,

A (k  i )2 i ( k  (i )) a (k  i )2 i ( k  (i )) a
 e  e
E 4ik 4ik 9.69

(k  i )2 (ik   ) a (k  i )2 (ik   ) a
e  e
= 4ik 4ik 9.70

(k  i )2 ika  a (k  i )2 ika   a
e e  e e
= 4ik  4ik  9.71

eika

(k  i )2 e  a  (k  i )2 e  a 
= 4ik 9.72

But (k  i )  k    2ik , and (k  i )  k    2ik

2 2 2 2 2 2
9.73

91
Hence, if   k    2ik , then  *  k    2ik
2 2 2 2

A

4
eika
ik 

(k 2   2 2ik )e a  (k 2   2 2ik )e  a 
E 9.74

eika
 
*
 A
   (k 2   2 2ik )e a  (k 2   2 2ik )e  a
E  4ik
9.75

Hence, if   k    2ik , then  *  k    2ik

2 2 2 2
9.76

A

eika

 e a   *e  a 
E 4ik 9.77

eika
 
*
 A
    * e  a   e  a
E  4ik
9.78

Multiplying equations 9.77 and 9.78,

 e   
*
 A  A  1 a
  * e  a   *e  a   e  a
   
 E  E  16k 
2 2
9.79

16k 
1
2 2
|  | 2
e2  a  ( )2  ( *)2  |  |2 e 2  a 
= 9.80


1
16k 2  2
|  | 2
(e2  a  e 2  a )  [( )2  ( *)2 ] 
9.81
  (k    2ik )  k    4k   2k   4ik  4ik 
2 2 2 2 4 4 2 2 2 2 3 2
9.82

( *) 2 (k 2   2 2ik )2  k 4   4 4k 2  2 2k 2  2 4ik 3 4ik 2  9.83

 2 ( *)2 

= 2k  12k   2 = 2(k  2k   4k    )
4 2 2 4 4 2 2 2 2 4

= 2(k  2k    )  8k 
4 2 2 4 2 2

92
 = 2(k   )  8k 
2 2 2 2 2
9.84

|  |2    * = (k 2   2 2ik )(k 2   2 2ik ) = (k 2   2)2 9.85

since | z |  (Re z ) for any complex number.

2 2

Hence,

(k 
*
 A  A  1
  2 )2 (e2  a  e 2  a )  [2(k 2   2)2  8k 2  2]
   
2

 E  E  16k 
2 2

2 a
1  2 2 2 (e  e 2  a ) 
 2( k   )  2(k 2   2) 2  8k 2  2 
16k 2  2
 2 
=

(k 2   2 ) 2  (e 2  a  e 2  a ) 4k 2  2 
 1  2
16k 2  2  2 (k   2 ) 2 

(k 2   2 ) 2  4k 2  2 
   
8k 2  2  (k 2   2 ) 2 
cosh( 2 a ) 1

9.86

But
2 2
e 2  e 2  e  e    e  e  
   
2  2   2 

or

cosh( 2 )  cosh 2   sinh 2 

But, we also know that

2 2
 e  e    e  e   e 2  e 2  2 e 2  e 2  2
 2   2    1
    4 4

or

93
cosh 2   sinh 2   1

cosh 2  cosh 2   sinh 2   1  sinh   sinh   1  sinh 2  or cosh 2  1  sinh 2  .

Therefore,

 A  A 
*
(k 2   2 ) 2  4k 2  2 
    8k 2  2 
2
 
(k 2   2 ) 2 
sinh ( a )
 E  E  

(k 2   2 ) 2 1
 sinh 2 (  a) 
8k 2 2
2
9.87

The transmission coefficient is,

1
(k   )
2 2 2 2
E 1
T  sinh 2 (  a) 
A 8k 
2 2
2 9.88

As you can see, the transmission coefficient is not zero as predicted by classical physics for a particle with
energy less than the potential barrier. This is what is called tunnelling, as the particle has effectively
tunnelled through the barrier.

Summary of study Session 9

For : 𝐸 > 𝑉0

1 In both kinds of potential, a part of the particle is reflected, quite unlike the classical prediction that
the particle should only slow down.

For a finite potential, if: 𝐸 < 𝑉0

2 Due to the wave nature of a particle, provided the potential is not infinite, there is a finite
probability of finding it beyond the point x = 0.
3 In the case of potential barrier, there is a possibility of finding the particle beyond x = 0, if the
potential is finite.
4 If the barrier is thin enough, the particle can tunnel through it.

Self- Assessment Questions for Study Session 9

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

94
SAQ 9.1

A particle of mass m is incident from the left on the potential step shown in Fig. Find the probability that
it will be scattered backward by the potential if

(a) 𝐸 > 𝑉0 , and

(b) 𝐸 < 𝑉0

𝑉0

0 𝑎 𝑥

95
Study Session 10: Time-Independent Schrödinger Equation in One Dimension II

Introduction
Perhaps one of the most important concepts in Physics is that of the harmonic oscillator. This is because a
lot of physical phenomena are fashioned after this kind of idealized motion. You must be quite familiar
with the motion of a harmonic oscillator: it is confined within the two positions where the total mechanical
energy, E, is equal to the potential energy. At these positions, the kinetic energy, or equivalently, the
velocity is zero, and the particle has to change directions. In this Unit, you will learn about the quantum-
mechanical equivalent. You will then see that the energy can only take specific values, and that the ground
state energy is not zero.

Learning Outcomes for Study Session 10

When you have studied this session, you should be able to:
1 How to derive the Schrödinger equation for the harmonic oscillator
2 How to derive the dimensionless form of the equation of the quantum oscillator
3 How to solve the Schrödinger equation for the harmonic oscillator
4 The allowed energy levels (energy eigenvalues) and the corresponding eigenstates

10.1 The simple harmonic oscillator

You would recall that the potential for the simple harmonic oscillator is

1
m 02 x 2
V (x) = 2 10.1

m 0 is the force constant.

2
where

The time-independent Schrödinger equation then becomes:

 2 d 2 1
  m 02 x 2  E
2m dx 2 2 10.2

As we do at times, we consider an extreme value of 𝑥, in this case large values of 𝑥 where 𝑉 ≫ 𝐸. We can
then modify the function that satisfies the large values of x by multiplying by a polynomial when the
extreme condition is no longer valid.

The equation approximates in the limit of large x to,

 2 d 2 1
 2
 m 02 x 2  0
2m dx 2 10.3

Let

96
 ( x)  A exp(  x 2 / 2a 2 ) 10.4

Then,

d x
  2 A exp(  x 2 / 2a 2 )
dx a 10.5

d 2 1 x2
2
  2
A exp(  x 2
/ 2 a 2
)  4
A exp(  x 2 / 2a 2 )
dx a a

d 2 1 x2
    
dx 2 a2 a4 10.6

or

d 2 x 2
 4
dx 2 a 10.7

since x is large.

Substituting in equation (1.3) and rearranging,

2 x2 1
   m 0 x 2
2
4
2m a 2 ,

i.e.,

2
a 
4

m 2 0
2
10.8

We conclude that  ( x)  A exp(  x / 2a ) is a solution of the ordinary differential equation provided

2 2

2
a4 
m 2 0 .
2

For small x , the approximation V  E is no longer valid, and the wave function must be modified
appropriately. As is often the practice, we shall multiply  for the extreme value of x by a polynomial,
x
H 
which in this case is  a  . To take into consideration the various possible eigenfunctions, we write the
trial function in the form,

 x
 n ( x)  An H n   exp(  x 2 / 2a 2 )
a 10.9

97
 An Hn
where is the normalisation constant and is a polynomial in x / a .

H 0 ( x / a)  1
For the ground state, let us choose the simplest polynomial, . So, we try the solution
 0 ( x)  A0 exp(  x / 2a ) . 2 2

1
m 0
2

Dividing through equation (1.2) by 2 , we get

2 d 2 2 E
  x 2 
m 2 0 m0
2 2 2
dx
10.10

2
 a4
m 0 2 2
Putting , we obtain

d 2 2 E
 a4  x 2 
m 0
2 2
dx
10.11

Substituting for d  / dx from equation (1.6),

2 2

 1 x2  2 E0
 a   2  4   x 2 
4

m 0
2
 a a 
10.12

giving, by comparing terms in powers of x,

2 E0
a2 
m 0
2
x0 : 10.13

1 1
E0  m 0 a 2   0
2
2
x : 2 2 10.14

The normalisation is determined from

 

  dx  A0 2
 exp(  x 2 / a 2 )dx  A0 a 
2 2

 
10.15

Therefore,

1
 0 ( x)  exp(  x 2 / 2a 2 )
a  10.16
98
and

1
E0   0
2 10.17

Before we take on the excited states, we shall derive the dimensionless form of the time-independent
Schrödinger equation for the harmonic oscillator.

We recall that,

d 2 2m
2
  2 [ E  V ( x)]
dx 

d d dx d d 2 2 d 
2
 a  a
Let us set R  x / a . Then, dR dx dR dx and dR 2 dx 2 , implying that,

1 d 2 2m
2 2
  2 [ E  V ( x)]
a dR  10.18

or

d 2 2ma 2
  [ E  V ( x)]
dR 2 2
10.19

2ma 2 2ma 2
 E W ( R )  V ( R)
Let 2 and 2

Then, we get the time-independent Schrödinger equation in a dimensionless form:

d 2
 [  W ( R)]
dR 2 10.20

or

d 2
  W ( R)  
dR 2 10.21

We recall that we made the substitutions

99
 2ma 2 2ma 2
 E W ( R )  V ( R)
R  x / a, 2 and 2

2 2
These are all dimensionless quantities. We can then see that the energy scale is  /( 2ma ) , i.e.,
E V ( R)
 2 2
W ( R)  2
 / 2ma and  / 2ma 2 .

Let us write
  E / E0 . Then, equation 1.21 becomes

d 2 x 2
a 2
   
dx 2 a 2 10.22

Further substituting R  x / a we get another dimensionless form,

d 2
 2  R 2  
dR 10.23

Let
 n  An H n ( R) exp(  R 2 / 2)

Then,

d n dH n
 An exp(  R 2 / 2)  RAn H n ( R) exp(  R 2 / 2)
dR dR
10.24

d 2 n d 2Hn dH n
2
 An exp(  R 2
/ 2) 2
 RAn exp(  R 2 / 2)
dR dR dR

dH n
 An H n ( R) exp(  R 2 / 2)  RAn exp(  R 2 / 2)  R 2 An H n ( R) exp(  R 2 / 2)
dR
10.25

n
Substituting equation 1.25 for the second differential of with respect to R in equation 10.23 with
 n  An H n ( R) exp(  R / 2)
2
,

 d 2Hn dH n
  An exp(  R / 2)
2
2
 RAn exp(  R 2 / 2)
 dR dR

100
 dH n 
 An H n ( R) exp(  R 2 / 2)  RAn exp(  R 2 / 2)  R 2 An H n ( R) exp(  R 2 / 2)
dR 

 R 2 An H n ( R) exp(  R 2 / 2)   n An H n ( R) exp( R 2 / 2)
10.26

 d 2 H n ( R) dH n ( R) dH 
 R  H n ( R)  R n  R 2 H n ( R)
  R H n ( R)   n H n ( R)
2 2
 dR dR dR

or

d 2Hn dH n
2
 2R  ( n  1) H n  0
dR dR
10.27

an ordinary differential equation known as the Hermite’s equation.

H n  an R n  an2 R n2  ...

Recall that we are proposing solutions of the form

H n  an R n  an2 R n2  ...

Let us now substitute into equation 1.27:

dH n
 nan R n1  (n  2)an2 R n3  ...
dR

d 2Hn
 n(n  1)a n R n2  (n  2)( n  3)a n2 R n4  ...
dR 2 10.28

dH n
 2R  2na n R n  2(n  2)a n  2 R n  2  ...
dR 10.29

( n  1) H n  ( n  1)( a n R n  a n  2 R n  2  ...)
10.30

Adding equations 1.28, 1.29 and 1.30,

d 2Hn dH n
2
 2R  ( n  1) H n
dR dR

 n(n  1)a n R n  2  (n  2)( n  3)a n  2 R n  4  ...

 2nan R n  2(n  2)a n2 R n2  ...

 ( n  1)(a n R n  a n2 R n2  ...)

101
  a n [n(n  1) R n  2  2nR n  ( n  1) R n ] n2 n4
+ terms in R , R , … = 0 10.31
n
But this would be true for all values of R and can therefore only happen if the coefficient of R vanishes,
i.e., if
 2n  ( n  1)  0 , i.e., if  n  2n  1 . Thus,

 2  2 m 0  2 m 0  (2n  1)
2 2
 1
En        0   0   n   0
2ma 2
2m 2 E 0 2m 2 0 4 4  2 where we made use
2 E0 2ma 2
a2   E
m 0
2
of: 2 .

5ħw0 /2

3ħw0 /2

ħw0 /2

Fig. 10.1: The energy levels of the quantum-mechanical harmonic oscillator

Let us see if we can find the wave functions. We try by substituting a trial wave function into Hermite’s

equation. We choose the second excited state (n = 2) H 2 ( R)  1  bR . Then,

2

dH 2 d 2H2
 2bR 2
 2b
dR and dR

Substituting, we get 2b  2 R  2bR  4(1  bR )  0 and hence b  2 . Thus,

2

 x2 
 2 ( x)  A2 1  2 2  exp(  x 2 / 2a 2 )
 a  10.32

The constant A2 can be found by normalisation.

102
We list the first three energy eigenvalues and wavefunctions for the simple harmonic oscillator. For the
graph in Fig. 1.2, we have chosen a = 1.

n Wavefunction
0 1/ 2
 1 
  exp(  x 2 / 2a 2 )
a  
1 1/ 2
 1   x
  2  exp(  x 2 / 2a 2 )
 2a   a
2
 1 
1/ 2
 2

   2  4 x   exp(  x 2 / 2a 2 )
 8a     a  

3
 1 
1/ 2
  x   x 3 
  12   8   exp(  x 2 / 2a 2 )
 48a    a a 
 

 1 1
E n   n   0  0
Notice that  2  ; and thus, the ground state energy is not zero, but 2 .

But, you may ask, why do we not observe the ground state in everyday Physics. It is because the number 
is so small that only when the system is very small (typically atomic and subatomic) that it becomes
significant. This is because some other parameters, such as the mass and distances are also small.

n=0
n=1
n=2

Fig. 10.2: Eigenfunctions of the quantum-mechanical harmonic oscillator

103
Summary of Study Session 10
In this study session 10, you have learnt:
1 That the quantum-mechanical harmonic oscillator can only attain a set of quantised energies.
2 That the ground state energy is ℏ𝜔0, as is the case with the classical oscillator.
3 The energy levels are equispaced.
4 The ground state is not observed in day to day Physics because  is such a small number.

Self- Assessment Questions for Study Session 10

Now that you have completed this study session, you can assess how well you have achieved its
learning outcomes by answering the following questions. Write your answers in your Study Diary
and discuss them with your Tutor at the next Study Support Meeting.

SAQ 10.1

 m 2 
 ( x)  A exp   x 
Normalise the eigenfunction  2  . Hence, find the probability that the particle

subjected to harmonic oscillation lies in the range 0  x   .

SAQ 10.2

A quantum-mechanical oscillator of mass m moves in one dimension such that its energy

eigenstate  ( x)  ( /  ) exp(  x / 2) with energy E    / 2m .

2 1/ 4 2 2 2 2

a. Find the mean position of the particle.

b. Find the mean momentum of the particle.

References/Further Readings
1. Quantum Mechanics demystified - David McMahon.

2. Introduction to Quantum Mechanics – David J. Griffiths.

3. Quantum Physics – Stephen Gasiorowicz.

104
APPENDIX

105
 SOLUTIONS TO SELF-ASSESSMENT QUESTIONS

STUDY SESSION 1

SAQ 1.1

Show that the following are vector spaces over the indicated field:

(i) The set of real numbers over the field of real numbers.

Let the set be R be the set of real numbers, then,

a b R  a,bR

and  aR  a R,  R

(ii) The set of complex numbers over the field of real numbers.
Let the set be C be the set of complex numbers, then,

c1  c2  C  a , b C

and  cR  c C ,  C

(iii) The set of quadratic polynomials over the complex field.

Let this set be P. Then

P1  a1 x 2  b1 x 2  c1 and P2  a2 x 2  b2 x 2  c2 are in P, where
a1 , a2 , b1 , b2 , c1 , c2 are constants.

a1 x 2  b1 x 2  c1  a2 x 2  b2 x 2  c2

 (a1  a2 ) x 2  (b1  b2 ) x  (c1  c2 )  P  P1 , P2  P

 (a1 x 2  b1 x  c1 )  P  P1  P,   the complex field.

106
SAQ 1.2

Check whether the following vectors are linearly independent.

(i) 2i  3 j  k ,  i  j  3k and  3i  2 j  k

 2    1   3   0 
       
a 3   b 1   c 2    0 
  1  3   1   0 
       

2a  b  3c  0

3a  b  2c  0

 a  3b  c  0

The solution set is (0, 0, 0), i.e., a = b = c = 0.

The vectors are linearly independent.

Alternatively,

2 1  3
3 1 2  2(1  6)  1(3  2)  3(9  1)
1 3 1

 2(5)  5  30  35  0

 i 1  2 1   1 2   i 2i 
 2 2i   i 2i   3  i   
(ii)  ,  ,   and  i  2

107
SAQ 1.3

 i 1  2 1   1 2   i 2i  0 0
a   b   c   d  
 2 2i   i 2i   3  i   i  2 0 0

Expanding,

ai  2b  c  id  0
(i)

a  b  2c  2id  0
(ii)

 2a  ib  3c  id  0
(iii)

2ia  2ib  ic  2d  0
(iv)

Multiplying (i) by 2 and adding to (ii),

a(1  2i)  5b  0 (v)

Multiplying (iii) by i and adding to (iv) gives

(1  2i)b  2ic  3d  0
(vi)

Multiplying (ii) by 2 and adding to (iii),

(2  i )b  7c  5id  0 (vii)

Multiplying (vi) by 5i and (vii) by 3 and adding,

5i(1  2i)b  10c  15id  0

(vi)

3(2  i)b  21c  15id  0 (vii)

(5i  10)b  10c  15id  0

(6  3i)b  21c  15id  0

(4  2i)b  11c  0 (viii)

108
 a(1  2i) 11
b  c
From (v) and (viii), 5 2i  4

Hence,

a(1  2i)(2i  4) 8  6i
c  a
55 55

Substituting for b and c in equation (vi),

 (1  2i)   8  6i 
(1  2i)  a   2i  a   3d  0
 5   55 

(3  4i) 16i  12
a a  3d
5 55

33  44i  16i  12 45  60i 9  12i

a a ad
165 165 55 (ix)

Putting b, c, d in (i),

 a(1  2i )  8  6i 12i  9
ai  2   ai a0
 5  55 55

2a 4ai 8a 6ai 12a 9ia

ai       0
5 5 55 55 55 55

 4 6 9   8 2 12 
ai1      a     0
 5 55 55   55 5 55 

55  44  6  9 8  22  12
ai  a 0
55 55

 18 26 
a i    0
 55 55  . Hence, a = 0, meaning that b, c, and d are also zero.

ai  2b  c  id  0
(i)

a  b  2c  2id  0
(ii)

 2a  ib  3c  id  0
(iii)

109
 2ia  2ib  ic  2d  0
(iv)

i 2 1  i
1 1 2 2i
0
2 i 3 i
2i 2i  i  2
Check if

SAQ 1.4

1   1
 ,  
1 1
Show whether or not the set     is a basis for the two-dimensional Euclidean space.

For the set to be a basis, the vectors must be linearly independent.

1   1  0 
a   b    
1   1  0 

a b  0 , a b  0

ab

a and b do not have to be zero. Hence, the vectors are not linearly independent. Sketch the vectors
and satisfy yourself that they are indeed linearly dependent: one can be got from the other because
they are indeed on a line.

Alternately,

1 1
0
1 1

SAQ 1.5

 1 2
 
Find the coordinates of the vector  2 i  with respect to the basis

110
 1 0 0 1 0  i  1 0  
 ,  ,  ,  
0 1 1 0  i 0  0  1  .

 1 2 1 0 0 1 0  i  1 0 
 2 i   a 0 1  b1 0  c  i 0   d 0  1
         

1 a  d (i)

2  b  ic (ii)

 2  b  ci (iii)

i  ad (iv)

Adding (i) and (iv):

1 i
a
2

(i) – (iv):

1 i
d
2

(ii) + (iii):

0b

(iii) – (ii):

2
  2i  c
i

Hence,

 1 2 1  i 1 0 0 1 0  i  1  i 1 0 
 2 i   2 0 1  01 0  2i  i 0   2 0  1
         

111
STUDY SESSION 2

SAQ 2.1

Which of the following functions are even and which ones are odd?

2
(i) x sin x cosh x (ii)
| e x | cosh 2 x (iii) sec x

Solutions

(i) is odd, being the product of two even functions and an odd function.
(ii) is an even function, a product of two even function.
(iii) is an even function:
1 1
sec( x)    sec x
cos( x) cos x

SAQ 2.2

Show that

(i)sin 𝑚𝑥 and cos 𝑛𝑥 are orthogonal,   x   .

(ii) sin 𝑚𝑥 and sin 𝑛𝑥 are orthogonal, 𝑚 ≠ 𝑛,

  x   .

Solutions


(i)
 sin mx cos nxdx  0 , the integrand is an odd


function

(ii)
 
 sin mx sin nxdx    2 cos(m  n) x  cos(m  n) xdx
1
 

112
  
1 1 1 
  sin( m  n) x  sin( m  n) x   0
2  m  n  mn  


113
STUDY SESSION 3

SAQ 3.1

Write the following as a sum of odd and even functions.

x
(i) e cosh x (ii) x ln x

Solutions

x
(i) h( x)  e cosh x , h( x)  e x cosh(  x)  e x cosh x

1 e  x cosh x  e x cosh x e x  ex 

f ( x)  [h( x)  h( x)]    cosh x  
2 2  2 

  cosh x sinh x

1 e  x cosh x  e x cosh x e x  ex 

g ( x)  [h( x)  h( x)]   cosh x  
2 2  2 

 cosh 2 x

SAQ 3.2

Evaluate the following integrals

a

(i)

a
x 2 n 1 dx n  0,1, 2, ....
,
a

(ii)
 a
x 2 n dx n  0,1, 2, ....
,

114
 Solutions

(i)
 a
x 2 n1dx
= 0, the integrand being an odd function.
a
x 2 n1 a 2 n1
a  2 2
a x dx  20 x dx 2n  1
a
2n 2n

(ii) 0 2n  1

SAQ 3.3

3 x 
 
If the matrix 1 2 is a proper orthogonal matrix, find x.

3 x 
det    6 x 1
1 2 ,

And

𝑥=5

115
STUDY SESSION 4

SAQ 4.1

 y i
 
If the matrix  i 2 is Hermitian, find the value of y.

The matrix is Hermitian if it is equal to its Hermitian adjoint, i.e.,

*
  y i T 
   y i
  i 2   
is   equal to  i 2
*
  y i T 
   y  i   y i 
*

  i 2  i 2 
 
  i 2
 
 

The matrix is Hermitian.

SAQ 4.2

 2   1   3 
     
  1,  0 ,  7 
 0    1   1
Show that the set       is a basis for the 3-dimensional Euclidean space. Show
that the set is not orthogonal. Hence, construct an orthogonal set from the vectors.

Solution
The set must be linearly independent:

2 1 3
0 7 1 7 1 0
1 0 7  2 1 3  2(7)  1(1)  3(1)  16  0
1 1 0 1 0 1
0 1 1

116
 Since we have 3 linearly independent vectors in the 3-dimensional Euclidean
space, the vectors form a basis for the space.

The set is not orthogonal:

1 3 1

     
2  1 0 0   2 2  1 0 7   1 3 7  1 0   4
  1   1   1
     

 2   1   3 
     
  1,  0 ,  7 
 0    1   1
To construct an orthogonal set:      

2
 
u1  v1    1
0
 

( v 2 , u1 )
u2  v2  2
u1
u1

In this case,

2
 
( v 2 , u1 )  1 0  1  1  2
0
 
2
 
 2  1 0   1  5
2
u1
0
 

Hence,

1  2   1   4 / 5   1/ 5 
( v 2 , u1 )   2       
u2  v2  2
u1   0     1   0     2 / 5    2 / 5 
u1   1 5  0    1  0    1 
         

In this case,
117
 2  1/ 5 
   
u1    1 u 2   2 / 5
0  1 
   

2
 
( v 3 , u1 )  3 7  1  1  1
0
 

5
2
u1

 1/ 5 
  3 14
( v 3 , u 2 )  3 7  1 2 / 5     1 
22
 1  5 5 5
 

 1/ 5 
  1
 1 / 5 2 / 5  1 2 / 5  
4 30 6
 1  
2
u2
 1  25 25 25 5
 

Hence,

3 2  1/ 5 
( v 3 , u1 ) (v 3 , u 2 )    1   22 / 5  
u3  v3  u1  u 2   7       1   2 / 5
  1  5  0  6 / 5   1 
2 2
u1 u2
     

3 2  1 / 5   3   2 / 5   11 / 15   40 / 15 
   1   11          
  7       1   2 / 5    7     1 / 5    22 / 15    80 / 15 
  1  5  0  3   1    1  0    11 / 3   8 / 3 
             

 8/3 
 
 16 / 3 
 8/3 
 

The orthogonal set is,

118
  2   1 / 5   8 / 3 
     
  1,  2 / 5 , 16 / 3 
 0    1   8 / 3 
     

SAQ 4.3

Show that the set you obtained in SAQ 6.2 is not an orthonormal set, even though it is an orthogonal set.

Normalise each vector in the set. Hence, expand the normalised vector  1 2 1 / 6 in terms
of this basis. Calculate the probability that the system represented by the vector is in each of these
states.

Solution

The norms of the vectors in the set are, respectively, 5, 6 / 5 and 128 / 3 . Hence, the
normalised set is,

  2  1/ 5   8 / 3 
 1   1   1  
   1,  2 / 5 , 16 / 3 
 5  0  6 / 5   1  128 / 3  8 / 3 
      

2  1/ 5   8/3 
c1   5   3  
  1  c2  2 / 5   c3 16 / 3 
5  6   128  
0  1   8/3 

  1
 
c1  (1 , )  2  1 0 2    4
1
6 5 1 30
 

  1
 
c2  (2 , ) 
5 1
1 / 5 2 / 5  1 2    5   1  4  1
6 6 1 6  5 5 
 

  1
 
c3  (3 , ) 
3
1 2 1 2   3 8 4  8  8  2
8
128 3 6 1 4 83 6 3 6 3 2 3
 

119
 16 8 1 4
| c1 | 2   | c2 |2  | c3 | 2 
30 15 45 9

Adding to check if we are right,

16 1 4 48  2  40 90
| c1 | 2  | c2 | 2  | c3 | 2      1
30 45 9 90 90

You know we could have expanded the vector in terms of the basis vectors, but we chose to make
use of the fact that,

ci  (i , )

{ k }nk 1
Provided is an orthonormal set, and
n

| c
i 1
i |2  1

Provided, in addition, the vector we expanded in terms of the orthonormal basis is itself
normalised, as it is the case in this SAQ.

SAQ 4.4

A particle trapped in the well

0, 0  x  a
V 
, elsewhere

is found to have a wave function

i 2  x  2  2 x  1  3 x 
sin   sin   sin  
2 a  a  3a  a  2a  a 

2 n x
sin
Given that the allowable wave functions are of the form a a , what is the probability of
obtaining each result?

120
 Solution
3
 2 n x 
 sin 
 a a  n1
The allowable wave functions are of the form . Hence the expansion in terms of
these eigenfunctions is,

i 2 x 1 2 2 x 1 2 3 x
sin sin  sin sin  sin
2 a a 3 a a 2 a a

c  i / 2 c 2  1 / 2 c3  1 / 2
Since 1 , , , the probability, respectively, that the particle will be found
in these states (n = 1, 2, 3) is:

i  i 1 | c |2  1  1  1 1 1 1
| c1 | 2    2 | c3 | 2   
2 2 4, 2 2 2 , 2 2 4

SAQ 4.5

A particle in a one-dimensional box 0  x  a is in state:

1 x A 2 x 3 3 x
 ( x)  sin  sin  sin
5a a a a 6a a

(c) Find A so that  (x) is normalized.

(d) What are the possible results of measurements of the energy, and what are the respective
probabilities of obtaining each result?

Solution

We first put  in the form of the allowable eigenfunctions:

1 x A 2 x 3 3 x
 ( x)  sin  sin  sin
5a a a a 6a a

1 2 x A 2 2 x 3 2 3 x
 sin  sin  sin
5a 2 a a 2 a 6a 2 a

121
 1 2 x A 2 2 x 3 2 3 x
 sin  sin  sin
10 a a 2 a a 2 6 a a

1 A 3
 1  2  3
10 2 2 6

For  to be normalized,

( ( x), ( x))  1

 1 A 3 1 A 3 
 1  2  3 , 1  2  3   1
or  10 2 2 6 10 2 2 6 

1 A2 9
  1
10 2 24

 9 1  120  45  12  63
A 2  21     2 
or  24 10   120  60 .

63
A
Hence, 60

The normalised  is therefore,

1 63 3
  1  2  3
10 120 2 6

The possible values of the energy are:

 2 2 2 2 2 9 2 2
E1  E 2  E 3 
2ma 2 , ma 2 , 2ma 2 ,

1 63 9
2 2 2
|c |
and the probabilities, respectively, are | c1 | = 10 , | c2 | = 120 , 3 = 24

122
STUDY SESSION 5

SAQ 5.1

Find the maximum kinetic energy with which an electron is emitted from a photometal of work
39 39
function 3.2  10 J when a radiation of energy E = 3.313  10 J falls on it, given
39
that the work function is 3.2  10 J .

Solution

The maximum kinetic energy is given by

1 2
mv  E  W  (3.313  10 39  3.2  10 39 ) J
2  1.13  10 40 J

SAQ 5.2

6
What is the wavelength of the wave associated with an electron moving at 10 m / s .

Solution

h  p

h 6.626  10 34
  31
 7.28  10 10 m
p 9.1 10  10 6

123
STUDY SESSION 6

SAQ 6.1

1. By solving the time-dependent Schrödinger equation for a free particle (V = 0), find the condition
imposed on the angular frequency and the wavenumber.

2  
 ( x, t )  i  ( x, t )
2m x 2
t (i)

Solution

Let

( x, t )  Ae i ( kx t )

Then,

2
 ( x, t )  k 2 Ae i ( kx t )  k 2  ( x, t )
x 2


 ( x, t )  i Ae i ( kx t )  i   ( x, t )
t

Putting these into the equation (i), we see that

2k 2
 
2m

SAQ 6.2

By applying the method of separation of variables, solve the differential equation

 2 2 2 
 2  2  2 
 x y z 
=0

124
 Solution

We shall assume that

( x, y, z )  X ( x)Y ( y ) Z ( z )

Putting this in equation (133),

1 d2X 1 d 2Y 1 d 2Z
  0
X ( x) dx 2 Y ( y ) dy 2 Z ( z ) dz 2 (i)

For i.e., equation (i) to be valid, each term on the left must be separately equal to a constant.

1 d2X 1 d 2Y 1 d 2Z
   k3
2 2 2
2
k1 ; 2
k 2 ; 2
X ( x) dx Y ( y ) dy Z ( z ) dz

subject to the condition:

k1  k 2  k3  0
2 2 2

We can see that not all these constants can be real nor all imaginary. At least one of them must be
real and one imaginary. The third may be real or imaginary. Which constants are real and which
are imaginary will be decided by the physical situation of the problem we are dealing with.

Let

k1   2 k 2    2 k  2
2 2 2
, and 3

Then,

1 d2X 1 d 2Y 1 d 2Z
  2
;    2
; 2
X ( x) dx 2 Y ( y ) dy 2 Z ( z ) dz 2

The solutions are:

125
 X ( x)  A1e ix  A2 e ix (i)
iy  iy
Y ( y)  B1e  B2e (ii)
z z
Z ( z )  C1e  C2e (iii)

These are just a particular solution of the partial differential equation. As there may be other values
of
 ,  and  which could give valid solutions, the general solution is:

   

( x, y, z )   Ar1ei r x  Ar 2ei r x B s1ei s y  B s 2ei s y C t1e t z  C t 2e t z
r , s ,t

SAQ 6.3

n
 x
 ( x)    e 2 x / x0
What would the potential function be if  x0  is an eigenfunction of the

Schrödinger equation? Assume that when x   ,

V ( x)  0 .

Solution
n
dF x n1 2 x
 n n e 2 x / x0    e 2 x / x0
dx x0 x0  x0 

2 n
d 2F x n2 2 x / x0 x n1   2  2 x / x0 2 x n 1 2 x / x0  2   x  2 x / x0
 n ( n  1) e  n  e  n n e      e
dx 2 x0  x0 
n n
x0 x0 x0  x0   x0 
2 n
x n2  2  x n 1  2  x n 1  2  x   2 x / x0
 n(n  1) n e  2 x / x0  n  n e  2 x / x0  n  n e  2 x / x0      e
x0  x0  x0  x0  x0  x0   x0 
2 n
x n2  4  x n 1  2 x / x0  2   x   2 x / x0
 n(n  1) n e  2 x / x0  n  n e      e
x0  x0  x0  x0   x0 

2 n
n(n  1) x n  2 x / x0  4  x n  2 x / x0  2   x   2 x / x0
 e  n  n e      e
x 2 x0 n  0 0
xx x  x0   x0 
126
  n(n  1)  4   2 
2
 xn
  n      n e  2 x / x0
 x  x0
2
 xx0   x0 

 n(n  1)  4   2 
2

  n      ( x)
 x  xx0   x0  
2

Putting this into the Schrödinger equation,

 2 d 2 ( x)  2  n(n  1)  4   2 
2

   n      ( x)  (V  E ) ( x)
2m dx 2 2m  x 2  xx0   x0  
 

When x   ,
V ( x)  0 , meaning that,

2 2
E 2
mx0

as the two terms that have x in their denominator tend to zero.

Therefore,

 2  n(n  1)  n 
V ( x)   2
 4 
2m  x  xx0 

127
STUDY SESSION 7

SAQ 7.1

A particle of mass m is confined within a one-dimensional box of length L / 2 , subject to a

Potential:

0, 0  x  L / 2
V ( x)  
, elsewhere

If at
t0
, the wave function is
 ( x)  Ax( L  x / 2) , i.e.,  ( x,0)  Ax( L  x / 2) ,

(i) normalise , and hence, determine the value of A.

(ii) Write
 ( x, t ) as a series, where t  0 .

(iii) Write an expression for the coefficients in the series.

Solution
L/2
2 2 x x4 x5 
2 3
2 2 x   
L/2
 0
A x  L   dx
 2 
A 

L
3
L
4 5 0
(i) =

1 1 1  31 2 5
 A 2 L5     A L 1
 24 64 160  960

960 15 15
A  8 8 x( L  x / 2)
5 5 5
Hence, 31L 31L . The normalised wave function is 31L .

(ii) The allowed eigenfunctions for the well can be deduced from those of the well

0, | x | L
V ( x)  
, elsewhere

by making a transformation L  L / 2 .

128
 2 n x
 n ( x)  sin
Recall that for the latter well, L L . For the well at hand, then,
2 n x 4 2n x 2 2n x
 n ( x)  sin  sin  sin
L/2 L/2 L L L L , and the energy eigenvalues are
n 2 2  2 n 2 2  2
En  E 
2m( L / 2) 2 as against n 2mL2 for the well of length L.

We can therefore write the wave function as


2 2 j x
 ( x,0)   c j sin
j 1 L L

At time t  0 ,

2 2 j x iEn t / 
 ( x,0)   c j (0) sin e
j 1 L L

L/2  2 2 j x 8 15 
 
 L sin  x ( L  x / 2) dx

c j (0)  ( j , ( x,0)) 0
 L 31L5

where

16 15 L/2  2 j x 

L3
31
 0
 sin
 L
 x( L  x / 2) dx


16 15 3L3 48 15
  
L3 31 j j 31

Details of the integration

We first carry out the integration

L/2 L/2 L/2 x2
0
x( L  x / 2) sin wxdx  L
0
x sin wxdx  
0 2
sin wxdx
. We shall later make the
2 j
w
identification L . We shall evaluate these integrals separately so that you can make use of
them at some other time, if need be.

129
 L/2

I1   0
x sin wxdx

1
v cos wx
Let x  u . Then, dx  du . Let sin wxdx  dv . Then, w

Applying the formula for integration by parts,

 udv  uv   vdu ,
L/2
x L L/2
w 0
I 1   cos wx  cos wxdx
w 0

L/2 L/2
x 1
I 1   cos wx  2 sin wx
w 0 w 0

L/2 x2
I2   sin wxdx
0 2
L/2
2I 2   x 2 sin wxdx
0

We have integrated
2I 2 , so you would know how to integrate
 x 2 sin wxdx
.

1
v cos wx
Let x  u . Then, 2 xdx  du . Let sin wxdx  dv . Then,
2
w

Applying the formula for integration by parts,

 udv  uv   vdu ,
L/2
x2 L/2
2I 2   cos wx  2 x cos wxdx
w 0
0

In the integral, let x  u . Then, dx  du . Let dv  cos wxdx . Then,

1
v sin wx
w

Applying the formula for integration by parts,

 udv  uv   vdu ,

2 x 
L/2 L/2
x2 1 L/2
w 0
2 I 2   cos wx   sin wx  sin wxdx
w 0
w  w 0 

130
 2 x 
L/2 L/2 L/2
x2 1
2 I 2   cos wx   sin wx  2 cos wx 
w 0
w  w 0 w 0 

1
I  LI 1  I 2  LI 1  2 I 2
The integral we are interested in is, 2 .

Therefore,

  1
 2 x 
L/2 L/2 L/2 L/2 L/2
 xL L  x2 1 
I   cos wx  2 sin wx    cos wx   sin wx  2 cos wx 

 w 0 w 0 
 2
w 0
w  w 0 w 0  

 
L/2 
  1 1 
L/2 L/2 L/2 L/2
 L L  x2 1
I   x cos wx  2 sin wx    cos wx   x sin wx  cos wx 

 w 0 w 0 
  
2w 0
w  w 0 2w 2 0  


Putting in the limits,

 L L L  L  L 
I    cos w  0 cos 0  2 sin w  sin 0 
 w 2 2  w 2 


 ( L / 2) 2 L  (0) 2  1  L L  1  L 
  cos w   cos 0  2  sin w  0 sin 0   cos w  cos 0  
  w  2  2w 
3
 2 w 2  2w 2 2 

 L L L L  L 
I    cos w   2 sin w  
 w 2 2 w  2 

 ( L / 2) 2 L 1 L L 1  L 
  cos w   2  sin w    cos w  1  

 2 w 2 w 2 2  2w3 2 

2 j
w
But L , or wL  2 j

 L L  L   ( L / 2) 2  1 L  
I    cos 2 j   2 sin 2 j     
1
cos 2 j   2  sin 2 j   cos 2 j  1 
 w 2  w  w 2  2w
3
  2 w 
 L L    ( L / 2) 2  1 
cos 2 j   2 0  cos 2 j  1 
1
I    cos 2 j   0  
 w 2 
3
  2 w  w 2w 

131
  L L   ( L / 2) 2  
  0  2 1  1 
1 1
I      0    
 w 2   2w  w w 
 L L2   ( L / 2) 2 L    L3 L3  3L3
I           
 2 j 2   2(2 j )    4 j 16 j  j

SAQ 7.2

 ( x)  A(ax  x 2 ) for | x | a . Normalise the wave function and find  x  ,  x 2  and x .

Solution

a a
 a
 * ( x) ( x)dx   A 2 (ax  x 2 ) 2 dx
a

a
 2 A 2 (ax  x 2 ) 2 dx
0
, since the integrand is an even function, and the limit of
integration is from – a to a.
a a
2 A 2 (ax  x 2 ) 2 dx  2 A 2  (a 2 x 2  x 4  2ax 3 )dx
0 0

a3 a5 a4 1 1 1
 2 A (a 
2
 2a )  2a 5 A 2    
2

3 5 4 3 5 2

10  6  15 1
 2a 5 A 2  2a 5 A 2 1
30 30

A  30 / 2a 5

30 a 30 a

 x  2a 5  a
(ax  x 2 ) x(ax  x 2 )dx 
2a 5  a
x(ax  x 2 ) 2 dx  0
, since the integrand is an
odd function.

30 a 30 a

 x  2a 5
2  a
(ax  x 2 ) x 2 (ax  x 2 )dx  2 
2a 5  0
x 2 (ax  x 2 ) 2 dx  0
, since the integrand
is an even function.

132
 30 a 60 a
2
2a 5  0
x 2 (ax  x 2 ) 2 dx 
2a 5  0
x 2 (a 2 x 2  x 4  2ax 3 )dx

a
 30a 5  (a 2 x 4  x 6  2ax 5 )dx
0

30 7  1 1 1  2 21  15  35 30a 2 2a 2
 5 a      30a  
a 5 7 3 105 105 7

2a 2 2a 2 2
x   x 2    x  2   02  a
7 7 7

SAQ 7.3

A particle is confined within a one-dimensional region 0  x  L . At time t = 0, its

  x  x
A1  cos sin
Wave function is given as  L  L .

(i) Normalise the wave function.

(vi) Find the average energy of the system at time t = 0 and at an arbitrary time𝑡0 .
(vii) Find the average energy of the particle.
(viii) Write the expression for the probability that the particle is found within 0  x  L / 2 ?

Solution
 x x
2

 A 1  cos dx  1
2
 sin 2
 L  L

L x x x  x 2  x
A 2   sin 2  sin 2 cos 2  2 cos sin dx  1
or
0
 L L L L L 

x x 1 2 x
sin 2  1  cos 2  [1  cos ]
But L L 2 L

133
 L 1 2 x 1 2 x 1 2 x x 1 2 x 
A 2   [1  cos ]  [1  cos ]  [1  cos ]  2 cos  [1  cos ] dx  1
0
2 L 2 L 2 L L 2 L 

=
L 1 2 x 1 2 x 1 2 x x 1 2 x 
A 2   [1  cos ]  [1  cos ]  [1  cos ]  2 cos  [1  cos ] dx  1
0
2 L 2 L 2 L L 2 L 

L 1 1 2 x 1 2 x x x 2 x 
A 2    cos  [1  cos 2 ]  cos  cos cos dx
=
0 2
 2 L 4 L L L L 

L 1 1 2 x 1 1 4 x x x 2 x 
 A 2    cos  [1  {1  cos }]  cos  cos cos dx
0 2
 2 L 4 2 L L L L 

L 1 1 2 x 1 1 4 x x x 2 x 
 A 2    cos  [  cos }]  cos  cos cos dx
0 2
 2 L 8 8 L L L L 

cos a cos b 
1
cos(a  b)  cos(a  b)
Now, 2 . Hence,

x 2 x 1  3 x  x
cos cos  cos  cos
L L 2 L L  , since cos(  x)  cos x

We can now integrate

L 1 1 2 x 1 1 4 x  x 1  3 x  x 
A2
 0  2  2 cos L  [ 8  8 cos L }]  cos L  2 cos L  cos L   dx
  

L
x L 2 x x L 4 x L x L 3 x L  x
A  
2
   sin  
L  0
sin sin sin sin
=  2 4 L 8 32 L  L 6 L 2

L
x x L L 5
A2    A2     A2 L  1
= 2 8 0 = 2 8 8

Hence,

8
A
5L

134
 8   x  x
1  cos  sin
The normalized wave function is 5L  L  L

2  n x 
n  sin  
(ii) We know that the allowable solutions are of the form L  L  and the energy
eigenvalues are (from equation 3.2, Module 3, Unit 1)

n 2 2  2
En 
2mL2 , n = 1, 2, 3, …

Any wave function

 ( x, t ) can be expressed in  n as

 ( x, t )   An (t ) n ( x,0)
n

where

 iE t 
An (t )  An (0) exp   n 
  

In this case,

8  x  x
 ( x,0)  1  cos  sin
5L  L L

8 x 1 2 2 x
sin  sin
= 5L L 2 5L L

we have,

2 1
A1 (0)  A2 (0)  
5, 2 5 , An (0)  0 for n  1, 2.

8  i 2 t 0  x 1 2  i 2 t 0  2x
 ( x, t 0 )  
exp   2 
 sin  exp   2 
 sin
5L  2mL  L 2 5 L  mL  L

(iii) The average energy of the system is

135
  E     n | E |  n 
n

 A (0)
n
n
2
En
=

4 1
E1  E2
= 5 20

4  2 2 1 4 2  2 17 2 2  2 17 2  2
2
  
= 5 2ma 20 2ma 2 20 ma 2 10ma 2

(iv) The probability of finding the particle within 0  x  L / 2 is

 x 2  x
2
L/2 L 8 
0
|  | 2 dx  
0
1  cos
5L 
 sin
L  L
dx

SAQ 7.4

A particle trapped in the well

0, 0  x  a
V 
, elsewhere

is found to have a wave function

i 2  x  2  3 x  2  3 x 
sin   sin   sin  
2 a  a  3a  a  16a  a 

a. If the energy is measured, what are the possible results and what is the probability of
obtaining each result?

b. What is the most probable energy for this particle?

c. What is the average energy of the particle?

136
Solution

2 n x
sin
(a) The allowable wave functions are of the form a a . Hence the expansion in
terms of these eigenfunctions is,

i 1 1 2 3 x
sin 1  sin  2  sin
2 3 2 a a

The possible results of the measurement of the energy are:

n 2  2 2
En 
2ma 2 , with n = 1, 2, 3

Respectively, the possible results are:

 2 2 2 2  2 2 2 2 2 32  2 2 9 2 2
E1  E 2   E 3  
2ma 2 , 2ma 2 ma 2 , 2ma 2 2ma 2

c  i / 2 , c 2  1 / 2 , c3  1 / 4 , the probability, respectively, that the particle will

Since 1
be found in these states (n = 1, 2, 3):

i i 1 1 1 1 1 1 1
| c1 | 2    | c2 | 2    | c3 | 2   
2 2 4, 2 2 2, 2 2 4

2 2 2 1
E2  2
(b) The most probable energy of the particle is 2ma , with probability 2 .

PE i i 3
i 1
3
  Pi Ei
P
i 1
i
i 1
 is
(c) The average energy of the particle is , since the wave function
3

P
i 1
i 1
normalized, so that .

137
Hence, the average energy of the particle is
1  2 2 1 2 2 2 1 9 2 2 1  2 2 8 2 2 9  2 2 9 2 2
     
4 2ma 2 2 ma 2 4 2ma 2 8 ma 2 8ma 2 8 ma 2 4ma 2

138
STUDY SESSION 8

SAQ 8.1

What are the allowable eigenfunctions and energy eigenvalues of the infinite potential
well?

0,  L  x  L
V ( x)  
, elsewhere

Solution
Potential well.

−∞ ∞

𝑣=0

𝑥 = −𝐿 𝑥=𝐿

Fig. 3.1: The infinite square well potential confining a particle to a region of width 𝑳

Inside the well, the potential is zero. Outside the well, the potential is infinite. We expect
that the wave function outside the well will be zero.

 2 2m
 2 ( E  V )  0
We recall Schrödinger equation x
2
 .

When 𝑉 = 0,
139
 d 2 2m
 E  0
dx 2 
(1.1)

Which we can write as

d 2
 k 2  0
dx 2
(1.2)

The general solution of this equation is

 ( x)  A cos kx  B sin kx
(1.3)

Where 𝐴 and 𝐵 are constants to be determined, subject to the boundary conditions.

Since

 (x ) = 0 for x  L

 ( L)  A cos( kL)  B sin( kL)

A cos kL  B sin kL  0
(i)

 (x ) = 0 for x  L

A cos kL  B sin kL  0
(ii)

Adding (i) and (ii)

A cos kL  0

140
For this to be satisfied,

n
kL 
2
, for n = 1, 2 …
(iii)

Therefore,

n
  A cos x
2L
, n = 1, 2, …
(iv)

These solutions are the even-parity solutions, as.

 ( x)   (  x)

The allowable energies are then given by

2mE
k2 
2

OR

 n  2
2

  
 2L 
2 2
 k
En  
2m 2m

n 2 2  2
En 
8mL2

(ii) – (i):

B sin kL  0

In this case,

kL  n
, for n = 1, 2, …

n
k
L

Hence, for the odd-parity solution,

141
 n x
 ( x)  B sin
L

The allowable energies are then given by

2mE
k2 
2

OR

 n  2
2

  
 L 
2 2
 k
En  
2m 2m

n 2 2  2
En 
2mL2

We normalise the odd-parity solution.

L n x
( , )  A 2  sin 2 dx
L L

L 1 2n x 
 A2  1  cos dx
L 2 L 

2n x 
L
1  L
 A2  x  sin
2  2n L   L

1 2
 A [ L  ( L)]  A 2 L  1
2

Hence, the normalisation constant is

1
A
L

The odd parity solution is

1 n x
 ( x)  sin
L L , n = 1, 2, …

You can also show that the even parity solution is

142
 1 n x
 ( x)  cos
L 2 L , n = 1, 2, …

143
STUDY SESSION 9

SAQ 9.1

1. A particle of mass m is incident from the left on the potential step shown in Fig. Find
the probability that it will be scattered backward by the potential if

(a) 𝐸 > 𝑉0 , and

(b) 𝐸 < 𝑉0

𝑉0

𝑎 𝑥

Solution

(a) 𝐸 > 𝑉0 :

Region I:  I  e
ik ( x a )
 Re ik ( xa )

 Ii  Teik '( xa )

Region II: (wave travelling to the right only)

2mE 2m( E  V0 )
k k'
Where  2
and 2 .

144
Matching the wave functions at 𝑥 = 𝑎,

1 R  T

and the first derivative at 𝑥 = 𝑎,

ik  ikR  ik 'T

from which we deduce that

k  kR  k ' (1  R)  k ' R  k '

Therefore,

k  k'
R
k  k'

(k  k ' ) 2
| R |2 
The probability of reflection is
(k  k ' ) 2 .

(b) 𝐸 < 𝑉0

We identify two regions:

Region I: 𝑉 = 0,0 < 𝑥 < 𝑎

Region II: 𝑉 = 𝑉0 , 𝑥 > 𝑎

The solutions are:

Region I: 𝜓𝐼 = 𝑒𝑖𝑘(𝑥−𝑎) + 𝑅𝑒−𝑖𝑘(𝑥−𝑎)

145
 Region II: 𝜓𝐼𝐼 = 𝑇𝑒−𝑘(𝑥−𝑎)

2mE 2m(V0  E )
k k'
Where  2
and 2 .

Matching the wave functions at 𝑥 = 𝑎,

1 R  T

and the first derivative at 𝑥 = 𝑎,,

ik  ikR  k 'T

from which we deduce that

ik  ikR  k ' (1  R)  k ' R  k '

Therefore,

ik  k '
R
ik  k '

The probability of reflection is

k 2  k '2
|R|  2
2
1
k  k '2 ,

since the amplitude of incidence is unity.

146
 STUDY SESSION 10

 m 2 
 ( x)  A exp   x 
10.1 Normalise the eigenfunction  2  . Hence, find the probability that the
𝟏
particle subjected to harmonic oscillation lies in the range 𝟎 ≤ 𝒙 ≤ .
𝟐

Solution
 
 A 2 (e mx ) dx  A 2  e mx
2 2
/ 2 2 /
dx
 

m

where 

 1 
A2  e  x dx  A2 
2

 2 

   
A 2  e  x dx  A 2  A2  A2
2

  m /  m

Equating this expression to unity,


A2 1
m

m
A2 


147
  m 
1/ 4

A 
  

 m   m 2 
1/ 4

 ( x)    exp   x 
    2 

1
The probability of finding the oscillator between x = 0 and is
2

2
   m 1 / 4  m 2 
 0
 ( x) dx  
0
  exp  
    2
x 


 m   m 2 
1/ 2

 
  
 0
exp  
 
x dx


m 

 0

exp   x 2 dx

m 1  m 1   1
   
 2   2 m
= 2

This makes sense, right? If the probability of finding the oscillator between   and  is unity,
since the wave function is normalised, then the probability of finding it between 0 and  must be
half.

148
SAQ 10.2

A quantum-mechanical oscillator of mass m moves in one dimension such that its energy

ℏ𝟐 𝜸𝟐⁄
eigenstate 𝝍(𝒙) = (𝜸𝟐⁄𝝅)𝟏⁄𝟒 𝐞𝐱𝐩(−𝜸𝟐𝒙𝟐 ⁄𝟐)with energy.𝑬 = 𝟐𝒎
(a) Find the mean position of the particle.
(b) Find the mean momentum of the particle.

Solution
 
 x    * ( x) x ( x) dx  ( 2 /  )1 / 2  x exp(  2 x 2 ) dx
 
(a)

 ( 2 /  )1 / 2  x exp(  2 x 2 ) dx


Let

y  exp(  2 x 2 ) .

Then,

dy
 2 x exp(  2 x 2 )
dx .

Hence,

dy  2 x exp(  2 x 2 ) dx

 dy  2 x exp(  2 x 2 ) dx  y  c

But

y  exp(  2 x 2 ) .

Therefore,

149
 1
 x exp(  2 x 2 ) dx 
2
exp(  2 x 2 )  c

Hence,


1
 x  exp(  2 x 2 ) 0
2 

(b)

  d    d 
 p    * ( x)  i  ( x) dx  ( 2 /  )1 / 2  exp(  2 x 2 )  i  exp(  2 x 2 )dx

 dx   
 dx 


 (i)( 2 /  )1 / 2  (2 2 x) exp(  2 x 2 ) dx



 (i) 2 ( 2 /  )1 / 2  2 x exp(  2 x 2 ) dx



 (i) 2 ( 2 /  )1 / 2 exp(  2 x 2 ) 0


150