Block 2

MPH-004
QUANTUM MECHANICS-I
Indira Gandhi National
Open University
School of Sciences
Block
2
APPLICATIONS OF QUANTUM MECHANICS
UNIT 5
Simple One-Dimensional Potentials-I 7
UNIT 6
Simple One-Dimensional Potentials-II 47
UNIT 7
Simple Harmonic Oscillator-I 87
UNIT 8
Spherically Symmetric Potentials 119
UNIT 9
The Hydrogen Atom 143
Programme Design Committee
Prof. V.B. Bhatia, Retd. Prof. Enakshi Sharma Prof. G. Pushpa Chakrapani Prof. Suresh Garg, Retd.
University of Delhi, Delhi University of Delhi, South BRAOU School of Sciences,
Campus, Delhi Prof. Y.K. Vijay IGNOU, New Delhi
Prof. Abhai Mansingh, Retd.
University of Delhi, Delhi Prof. H. S. Mani, Retd. University of Rajasthan, Prof. Vijayshri
IIT Kanpur Rajasthan School of Sciences,
Prof. Feroz Ahmed, Retd.
Prof. S. Annapoorni Prof. J. Nag, Retd. IGNOU, New Delhi
University of Delhi, Delhi
University of Delhi, Delhi Jadavpur University Prof. S. R. Jha
Prof Yashwant Singh, Retd.
Prof. D. Choudhury Prof. Zulfequar, School of Sciences,
Banaras Hindu University,
University of Delhi, Delhi Jamia Milia Islamia,New Delhi IGNOU, New Delhi
Varanasi
Prof. T.R. Seshadri Dr. Om Pal Singh Prof. Shubha Gokhale
Prof. Deepak Kumar
University of Delhi, Delhi IGCAR, Kalpakkam, School of Sciences,
J.N.U., New Delhi
Tamil Nadu IGNOU, New Delhi
Prof. Vipin Srivastava Prof. S. Ghosh
J.N.U., New Delhi Prof. Prabhat Munshi Prof. Sanjay Gupta
Central University of
IIT Kanpur School of Sciences,
Hyderabad, Hyderabad Prof. Neeraj Khare IGNOU, New Delhi
Prof. G.S. Singh IIT Delhi, Delhi Prof. R.M. Mehra, Retd.
Dept. of Electronics, South Dr. Subhalakshmi Lamba
IIT Roorkee Prof. V.K. Tripathi
Campus, University of Delhi, School of Sciences,
Prof. A. K. Rastogi. IIT Delhi, Delhi IGNOU, New Delhi
Delhi
J.N.U., New Delhi Prof. Pankaj Sharan, Retd.
Prof. S. K. Kulkarni Dr. M.B. Newmai
Prof. A. K. Ghatak Jamia Milia Islamia, School of Sciences,
New Delhi Pune University/
IIT Delhi, Delhi IGNOU, New Delhi
IISER Pune, Pune
Prof. Rupamajari Ghosh Prof. Kirti Ranjan
J.N.U., New Delhi University of Delhi, Delhi
Course Design Committee

Prof. A.K. Ghatak, Retd. Prof. Ashok Kumar Prof. Vijayshri Prof. Sanjay Gupta
IIT Delhi, Delhi Institute of Advanced School of Sciences, School of Sciences,
Prof. T. R. Seshadri Studies, CCS University, IGNOU, New Delhi IGNOU, New Delhi
University of Delhi, Delhi Meerut Prof. S.R. Jha Dr. Subhalakshmi Lamba
Prof. Patrick Dasgupta Prof. Pankaj Sharan, Retd. School of Sciences, School of Sciences,
University of Delhi, Delhi Jamia Milia Islamia, New Delhi IGNOU, New Delhi IGNOU, New Delhi
Prof. Subir Sarkar Prof. Suresh Garg, Retd. Prof. Shubha Gokhale Dr. M.B. Newmai
J.N.U., New Delhi School of Sciences, School of Sciences, School of Sciences,
IGNOU, New Delhi IGNOU, New Delhi IGNOU, New Delhi
Block Preparation Team

Dr. S. Lamba (Units 5-9) Dr. M.B. Newmai (Unit 5)
School of Sciences, IGNOU, New Delhi School of Sciences, IGNOU, New Delhi
Course Coordinators: Dr. Subhalakshmi Lamba, Dr. M. Boazbou Newmai

Block Production Team
Sh. Rajiv Girdhar
AR (P), IGNOU
Acknowledgement: Shri Gopal Krishan Arora, EDP, SOS for CRC preparation.
June, 2023
© Indira Gandhi National Open University, 2023
ISBN:
Disclaimer: Any materials adapted from web-based resources in this module are being used for educational
purposes only and not for commercial purposes.
All rights reserved. No part of this work may be reproduced in any form, by mimeograph or any other means,
without permission in writing from the Copyright holder.
Further information on the Indira Gandhi National Open University courses may be obtained from the
University’s office at Maidan Garhi, New Delhi-110 068 or the official website of IGNOU at www.ignou.ac.in.
Printed and published on behalf of Indira Gandhi National Open University, New Delhi by
Prof. Meenal Mishra, Director, SOS, IGNOU.
Printed at
CONTENTS
Block and Unit Titles 1
Credit page 2
Contents 3
BLOCK 2: APPLICATIONS OF QUANTUM MECHANICS 5
Unit 5 Simple One-Dimensional Potentials-I 7
5.1 Introduction 7
5.2 A Free Particle 8
5.3 Particle in a Box 13
5.3.1 Solving the Schrödinger Equation 14
5.3.2 Eigenfunctions and Eigenvalues 15
5.3.3 Time Dependent Wave Function 19
5.3.4 Probability Density 19
5.4 Quantum Dot 21
5.5 Step Potential and Schrödinger Equation 24
5.6 Energy (E) of the particle is less than V0 26
5.6.1 Reflection and Transmission Coefficients 29
5.7 Energy (E) of the particle is greater than V0 31
5.7.1 Reflection and Transmission Coefficients 33
5.8 Summary 34
5.9 Terminal Questions 37
5.10 Solutions and Answers 38
Unit 6 Simple One-Dimensional Potentials II 47
6.1 Introduction 47
6.2 One-Dimensional Barrier Potential 49
6.2.2 Reflection and Transmission Coefficient for E<V0 54
6.2.3 Reflection and Transmission Coefficient for E>V0 58
6.3 Applications of Quantum Tunnelling 59
6.3.1 Alpha Decay 59
6.3.2 Scanning Tunnelling Microscope 61
6.4 One-dimensional Finite Potential Well 62
6.4.2 Parity of the Eigenfunctions 65
6.4.3 Energy Eigenvalues 67
6.4.4 Symmetric Infinite Potential Well 70
6.5 Summary 73
Appendix 6A: Calculating the Reflection and Transmission
Coefficients 84
Unit 7 Simple Harmonic Oscillator-I 87
7.1 Introduction 87
7.2 Stationary State Schrödinger Equation 89
7.2.1 Power Series Solution 90
3
7.2.2 Boundary Conditions and Eigenfunctions 92
7.2.3 Energy Eigenvalues 96
7.2.4 Parity of the Eigenfunctions 97
7.2.5 Solution of the Time-dependent Schrödinger Equation 97
7.2.6 Comparison with the Classical Oscillator 98
7.2.7 Examples of Simple Harmonic Oscillator Systems 101
7.3 Solution with Ladder Operators 103
7.4 Summary 106
Unit 8 Spherically Symmetric Potentials 119
8.1 Introduction 119
8.2 Three Dimensional Schrödinger Equation for a Central Potential 120
8.2.1 Schrödinger Equation in Spherical Polar Coordinates 121
8.2.2 Eigenfunction and Eigenvalues of L̂z Spherical Harmonics 122
8.2.3 Special Harmonics 123
8.2.3 Parity of the Spherical Harmonics 129
8.3 Space Organization 130
8.4 Radial Eigenfunctions 132
8.5 Summary 133
Unit 9 The Hydrogen Atom 143
9.1 Introduction 143
9.2 Schrödinger Equation for the Hydrogen Atom 144
9.3 Solution of the Radial Equation 147
9.3.1 Asymptotic Properties of the Radial Wave Function 148
9.3.2 Power Series Function 148
9.3.3 Eigenfunctions 151
9.3.4 Bound States and Continuum States 154
9.3.5 Degeneracy of the Eigenfunctions 155
9.3.6 Radial Probability Density 156
9.4 Spectra of the Hydrogen Atom 157
9.4.1 Quantum Numbers and Constants of Motion 158
9.5 Summary 159
Further Readings 169

Table of Physical Constants 170
List of Blocks and Units: MPH-004 171
Syllabus: Quantum Mechanics (MPH-004) 172
4
_BLOCK 2 : APPLICATIONS OF QUANTUM MECHANICS
In Block 1 of this course we have introduced you to the basic concepts of quantum
mechanics. Quantum mechanics has forced us to reconsider our deepest convictions about
the reality of nature. For example, the idea that wave-particle duality is an inherent property of
all the elements that make up the physical universe does away with the distinction between
matter and radiation (which is at the root of classical physics). Then you have seen how
classical determinism and causality are called into question by quantum mechanics through
the uncertainty principle and the probabilistic description. These fundamental ideas were built
into a mathematical framework which you studied in Units 3 and 4.
The methods of quantum mechanics become important when we talk of microscopic systems,
far removed from our daily experiences. Yet it affects our lives in many different ways,
because in today’s world, a lot of the technology we take for granted owes its origins to
quantum mechanics: shrinking transistor sizes to power laptops and mobiles, medical
technology like MRI and PET and so on. This has become possible, in part, because of one
important aspect of quantum mechanics: its ability to predict. You will discover its power in
this block when you solve the Schrodinger equation for different potentials. You will learn how
to solve the one-dimensional and the three dimensional Schrödinger equation.
In Unit 5 of this block, entitled “Simple One-Dimensional Potentials-I” you will solve the
one-dimensional Schrödinger equation for the free particle and for a particle confined to a
rigid one-dimensional box. You will study an important distinction between classical and
quantum physics: that when a quantum particle is confined to a tiny region, its energies
cannot be continuous. Rather the particle’s energy can have only specified discrete values,
which depend on the parameters of the system. This idea has important applications in
semiconductor technology and in this context, you will study about the quantum dot. You will
also study the properties of a quantum particle moving in a one-dimensional step potential.
You will learn that it is possible for quantum particles to be found in regions that are
classically forbidden.
In Unit 6 entitled “Simple One-Dimensional Potentials-II”, we solve the Schrödinger

equation for the one-dimensional potential barrier. You will find that it is possible for a
particle to be transmitted across a potential barrier even when its kinetic energy is less that
the barrier potential. This phenomenon of barrier penetration is called “quantum tunnelling”
because it has no classical analogue. You will study applications of quantum tunnelling in
alpha decay and scanning tunnelling microscopy. You also study about the motion of a
particle in a finite potential well. This is important because an infinite well in which particles
can be strictly confined is actually an idealization, rarely found in nature. Once again, you will
see that a bound particle has discrete energy levels.
In Unit 7 entitled “Simple harmonic Oscillator-I”, you solve the Schrödinger equation for
the one-dimensional harmonic oscillator potential and obtain the eigenfunctions and
the discrete eigenvalues. This is an important problem in physics, because in the
neighbourhood of a local minimum, almost any potential function can be approximated by a
parabola. You will study some examples of simple harmonic oscillator systems like a diatomic
molecule, etc. You will also study a new method of arriving at these results using “ladder
operators” which you learn more about in Unit 12.
In Unit 8 we solve the three-dimensional Schrödinger equation in spherical polar coordinates

for a spherically symmetric potential function which depends only upon the radial
5
coordinate r. You learn that the Hamiltonian in this case commutes with the angular
momentum operators L̂2 and L̂z . You derive the eigenfunctions of L̂2 and L̂z by solving the
angular part of the the three-dimensional Schrödinger equation. The eigenfunctions are the
spherical harmonics. You also determine the eigenvalues of L̂2 and L̂z .
The radial part of the three-dimensional Schrödinger equation depends on the potential
function and in Unit 9 we solve the equation for the Hydrogen atom potential, for an electron
moving in the Coulomb field of the proton. We derive the radial wave function and the energy
eigenvalues for the electron. You will learn about the degeneracy of the energy eigen states
and the significance of the quantum number for the hydrogen atom problem.
This Block is mathematically intensive and you should be familiar with the methods of solving
ordinary and partial differential equations, in particular the power series method and special
functions. Please work out all the mathematical steps and solve the SAQs and TQs.
We wish you all success!
6
Unit 5 Simple One-dimensional Potentials-I
UNIT 5
SIMPLE ONE-DIMENSIONAL
POTENTIALS-I
Structure
5.1 Introduction 5.6 Energy (E) of the Particle is less
Expected Learning Outcomes than V0
5.2 A Free Particle Reflection and Transmission
5.3 Particle in a Box Coefficients
Solving the Schrödinger Equation 5.7 Energy (E) of the Particle is greater
Eigenfunctions and Eigenvalues than V0
The Schrödinger equation
Time Dependent Wave Function Reflection and Transmission
cannot be derived, and like
Probability Density Coefficients the other postulates of
5.4 Quantum Dot 5.8 Summary quantum mechanics, its
5.5 Step Potential and Schrödinger 5.9 Terminal Questions validity lies in its agreement
5.10 Solutions and Answers with experimental results.
Equation
Schrödinger himself applied
his equation to the hydrogen
5.1 INTRODUCTION atom and in fact, it is in the
field of atomic physics that
In Unit 3 of Block 1 you have studied the time independent Schrödinger this equation has been most
equation. In this unit, you will learn how to solve Schrödinger equation in one successful. You should also
dimension. Although the real world is three-dimensional, these one- know that one cannot obtain
dimensional systems are of great interest. This is not only because several an exact solution for neutral
atoms more complex than
physical situations are effectively one-dimensional, but also because a
the hydrogen atom and one
number of more complicated physical problems can be reduced to the solution has to use approximation
of equations similar to the one-dimensional Schrödinger equation. methods, which you study
In Sec. 5.2, you will first study a quantum mechanical free particle, that is one about in Quantum
Mechanics –II in Semester
on which no force is being exerted. This is a particle that cannot be confined
2.
(bound) within a defined region of space by a potential, like a proton in a
nucleus. You will see that a quantum mechanical free particle has a Note also that the
Schrödinger equation is not
continuous spectrum of energy eigenvalues. In Sec. 5.3 we solve the quantum
a relativistic wave equation.
mechanical problem of a particle moving within a limited space like in a rigid So it is used only to study of
box. You will find that the energy eigenvalues of this system are different, a the non-relativistic behaviour
particle in a box has discrete energy levels . In Sec. 5.4 we discuss a quantum of atoms and molecules.
dot which is a practical realization of a particle in a box. A quantum dot is Just like Newton’s second
typically a nano sized semiconductor particle. The optical and electrical law which gives way to the
relativistic force law to study
properties of the quantum dot change with size and material, and they have
relativistic phenomena,
many applications in nano scale devices. there is also a relativistic
wave equation which was
Next in this unit you will learn how to solve the time independent Schrödinger
developed by Dirac in 1928.
equation for a step potential. You will obtain the eigenfunctions and energy
levels for such a potential. You are familiar with steps all around you. So, you
7
Block 2 Applications of Quantum Mechanics
can think of a step as a flat surface, which changes abruptly after a certain
distance so that another flat surface is obtained at a higher level.
Similarly, the step potential changes abruptly, much like a single physical step.
So, a step potential has an abrupt increase or decrease in its value. In
Sec. 5.6, we first define the step potential and write down the time
independent Schrödinger equation for a particle subjected to a step potential.
Then we take up two interesting cases: When the energy of the particle is less
than the height of the step potential (Sec. 5.6.1), and when it is greater
(Sec. 5.7). In both cases, we will get interesting results when we solve the time
independent Schrödinger equation for the step potential.
The step potential can be used to model a variety of physical systems and
some well known applications are seen in semiconductor devices, the physics
of normal-metal superconductor interfaces, and alpha decay in the nucleus.
Solving the step potential problem in this unit will also serve as a good
exercise for you to help you understand the next unit, where you will study
about the barrier potential.
Expected Learning Outcomes

After studying this unit, you should be able to:
 solve the time independent one-dimensional Schrödinger equation and
obtain the eigenfunctions and eigenvalues for a free particle;
 solve the time independent one-dimensional Schrödinger equation and
obtain the eigenfunctions and eigenvalues for a particle in a one
dimensional box;
 explain the properties of the quantum dot on the basis of the particle in a
box model;
 define a step potential;
 solve time independent Schrödinger equation for step potential under
given conditions;
 determine the reflection and transmission coefficients; and
 solve simple problems based on step potential.
5.2 A FREE PARTICLE

Recall the time independent Schrödinger equation for the stationary state
wave function (x ) for a quantum particle of mass m from Unit 3 (Eq.3.47):
 2 d 2 ( x )
  V ( x )( x )  E( x )  Hˆ ( x )  E( x ) (5.1a)
2m dx 2
You know that this is the eigenvalue equation for the Hamiltonian operatorc.
Any solution, (x ) , of this equation is the eigenfunction of Ĥ . And the energy,
E, is the eigenvalue of Ĥ corresponding to the eigenfunction (x ) .
The eigenvalues of Ĥ can form a discrete spectrum or have continuous

8
values. If 1( x ),  2 ( x ),  3 ( x )...... n ( x ) are the eigenfunctions of Ĥ with the
eigenvalues E1, E 2 , E 3 ......E n then the most general solution to the stationary
state Schrodinger equation is:
( x )  c11( x )  c 2 2 ( x )  c 3  3 ( x )  c n  n ( x )   c i  i ( x )
i
(5.1b)
And the general time dependent solution is:
iE i t

 ( x, t )   ci e  i (x)
(5.1c)
i
 2 ( x )
Note that in Eq. (5.1), the partial derivative has been replaced by
x 2
d 2 ( x )
because (x ) is only a function of x. Now, you know that a free
dx 2
particle is one on which no force is being exerted. You have studied that for a
conservative system, force is related to the potential as follows:
dV ( x )
F(x)   (5.2)
dx
So, for a system on which no force is exerted, we will have: The general solution of
dV ( x )
F(x)  0   0  V ( x )  V0 (5.3) d 2 ( x )
dx  k 2 ( x )  0
dx 2
where V0 is a constant potential.
As
Without loss of generality, we can take the constant V0 to be zero. So now we
have a particle of mass m moving freely in space. Putting V ( x )  0 in Eq. (5.1)  ( x )  Ae m1x  Be m2 x w
we can write the time independent Schrödinger equation for a free particle here m1 and m2 are the
as follows: solutions of the indicial
equation
 2 d 2 ( x )
  E( x ) (5.4) m2  k 2  0
2m dx 2
Here
In the classical picture, since the particle is not subjected to any force, its
m1  ik ; m 2   ik
linear momentum p does not change with time. E, the total energy of the
particle (which is equal to its kinetic energy) also does not change with time. E e m1x and e m2 x are the
and p are related by the equation: two linearly independent
solutions.
p2
E (5.5)
2m
 
From Eq. (1.14) you know that the linear momentum p  k . Since we are
considering only one-dimensional motion, we can assume that both the
momentum p and the wave vector k lie along the x-axis. So we write
p  p x  k . Then the expression for the energy of the particle is:
 2k 2
E (5.6)
2m
Let us now solve Eq. (5.1). Rearranging the terms we get:
d 2( x ) 2mE
  2 ( x ) (5.7)
dx 2 
2mE
Making the substitution k 2  2 we get the following second order ODE:

9
2
d ( x )
2
 k 2( x ) (5.8)
dx
For any given value of E and hence of k, there are two possible solutions to
this equation, which are also linearly independent (read the margin remark):
1 ( x )  e ikx ;  2 ( x )  e ikx (5.9)
Thus for one value of E, we have two possible eigenfunctions, 1( x ) and
2(x ) .
Recall from Sec. 8.3.4 that such eigenfunctions which have the same
eigenvalue for a given eigenvalue equation are called degenerate
eigenfunctions (otherwise they are non-degenerate). Thus, 1( x ) and
 2 ( x ) are degenerate eigenfunctions. The general solution of Eq. (5.4) is the
linear combination:
( x )  Ae ikx  Be  ikx (5.10)
The functions sin kx and
You can check for yourself by solving SAQ 1, that (x ) is also an
cos kx , as well as linear
combinations of these  2k 2
eigenfunction of Eq. (5.4) with the same eigenvalue E  .
functions are also 2m
solutions of the time
independent Schrödinger SAQ 1
equation for the free
particle. You will see this Show that ( x )  Ae ikx  Be  ikx is an eigenfunction of the time independent
for yourself when you work Schrödinger equation for the quantum mechanical free particle with an
out SAQ 2.
 2k 2
eigenvalue E  .
2m
For the solution (x ) to be physically acceptable, k cannot have an imaginary

part. Can you say why?
To understand, let us take k to be complex: k    i . In that case,
1 ( x )  e i (   i ) x  e ix e x (5.11)
Now, if  is positive, you can see that the term e x would increase
exponentially as x   . If  is negative, the term e x would increase
exponentially as x   . So this would not be an acceptable wave function, as
you have studied in Sec. 3.5. You can also check for yourself that  2 ( x )
would also would increase exponentially either as x   or as x   , if k
had an imaginary part.
So, k should be a real number and therefore k 2  0 . Therefore a quantum
mechanical free particle can have any non negative value of energy:
 2k 2
E 0 (5.12)
2m
Since any non-negative value of E is allowed, the energy spectrum of a free

particle is continuous, extending from E  0 to  . This, of course, is not
surprising, since E is the kinetic energy of a free particle. It also corresponds
to the classical result.
10
For a real value of k, 1( x ) and  2 ( x ) do not diverge at x   or as x   ,
but neither do they go to zero. For 1( x ) , you can see that :
  
 1 ( x )1 ( x )dx   e e dx   dx

 ikx ikx
(5.13)
  
which is clearly NOT finite. The same is true for  2 ( x ) . So, as you have
studied in Sec. 3.3.3, the eigenfunctions 1( x ) and  2 ( x ) are non-
normalizable.
We can also write down the time dependent wave function for the free particle.
In Sec. 3.4 you have studied that for a stationary state (x ) , the time
dependent wave function has the explicit form (Eq. 3.48):
( x, t )  ( x )e  iEt  (5.14)
So the time dependent wave function for the free particle with an energy E is
 
( x, t )  Ae ikx  Be  ikx e  iEt  (5.15)
If the quantum mechanical free particle is in a stationary state (x ) which

corresponds to the energy eigenvalues E (Eq. 5.1), while the wave function
will evolve with time as given by Eq. (5.15), the energy of the particle, E, will
not change with time. (x,t)
Ae i ( kx  t )
Using E   , we get: x
( x, t )  Ae i kx t   Be  i kx  t  (5.16)
In order to interpret Eq. (5.16) physically, let us consider some special cases.
If we set B = 0, the resulting wave function is the plane wave:
( x, t )  Ae i kx  t  (5.17)
This function represents a travelling plane wave as shown in Fig. 5.1. This
wave is associated with a free particle of mass m moving along the positive x- Fig. 5.1: A travelling wave.
 2k 2
axis with a definite momentum of magnitude ħk and an energy .
2m
The probability density corresponding to the wave function of Eq. (5.17) is:
2
P ( x, t )   * ( x, t )( x, t )  A (5.18)
which is independent of time t as well as position x. In other words, at any

instant of time t, the probability of finding the particle in an interval
2
between x and x  dx is the same ( A dx ) for any value of x. In other
words, the position of the particle is completely unknown.
This result can also be deduced from the uncertainty principle. Since the
momentum of the particle is precisely known, p  p x  0 . Then, from the
uncertainty principle, the uncertainty in the position of the particle x   .
Further, since the energy of the particle is also known precisely, we also have
E  0 . Hence, according to the energy-time uncertainty principle, we have
an infinite time ( t   ) to make a measurement of the energy of the particle.
11
You can carry out a similar analysis by setting A = 0 in Eq. (5.16). In this case
the plane wave will be travelling in the negative x-direction.
It is possible to verify that 1( x ) and  2 ( x ) are also the eigenfunctions of the
momentum operator. However the eigenvalues of the momentum operator are
different for the two eigenfunctions. We work that out in the following
example.
Example 5.1
Show that 1 ( x )  e ikx and  2 ( x )  e  ikx are eigenfunctions of the momentum

operator and determine the corresponding eigenvalues.
Solution : We have to show that the eigenvalue equation, Eq. (5.35), is valid
d
for the wave functions 1 ( x )  e ikx and  2 ( x )  e  ikx with Dˆ  pˆ x  i  .
dx
Using 1 ( x )  e ikx in Eq. (5.35) we get:
d d ikx
pˆ x  1  x   i   1  x   i  e   i ike ikx   ke ikx
dx dx
So pˆ x 1x   k 1x  . So 1 ( x ) is an eigenfunction of the momentum

operator p̂x with the eigenvalue k .
Using  2 ( x )  e ikx in Eq. (5.35) we get:
d d ikx
pˆ x  2  x   i   2  x   i  e 
dx dx
 i   ike ikx   ke ikx
Since pˆ x  2 x    k  2 x  , we can say that  2 ( x ) is also an eigenfunction of

the momentum operator p̂x with the eigenvalue  k .
The energy E [E(k)] and the momentum p [p(k)] of a quantum mechanical free
particle are both constants of motion and are defined in terms of k. k also
characterises the eigenfunction (x ) for the quantum mechanical free particle
and hence we call k a quantum number.
You may now like to answer an SAQ.
SAQ 2
Show that (i) ( x )  sin kx and (ii) ( x )  cos kx are solutions of the
Schrödinger equation for the free particle (Eq. 5.7).
In the classical picture, a free particle has “unbounded motion” because it is

not subjected to a force. In the corresponding quantum mechanical picture,
the free particle can have any energy as defined by Eq. (5.6) and the energy
eigenvalues E vary continuously with the value of k. So the energy spectrum is
continuous and ranging between zero to infinity. Each energy state has a two-
fold degeneracy corresponding to particles moving either along the positive x
12
direction or along the negative x direction, except for the state corresponding
to k = 0.
However, in the classical picture when the value of the momentum (p) is
known exactly, you can locate the position of the particle at any instant of time,
if the initial position is known. You have seen that this is not possible for a
quantum mechanical free particle because of the uncertainty principle.
Let us now see what happens when we confine the free particle to a box.
5.3 PARTICLE IN A BOX

Let us consider that a free particle is confined in a one-dimensional length V  V 
segment lying between x = 0 and x = L. You can imagine that there are rigid
barriers at x  0 and x  L that the particle cannot penetrate to move beyond
these points: so the particle cannot move to the left of x  0 or the right of
x = L. This is also called a rigid box. We describe this system by saying that
the potential V(x) is infinite for all values of x  0 and x  L . Between
0  x  L , the particle is free, and the potential is zero in this region. The
potential function is shown in Fig. 5.2 and is described as:
0 for 0  x  L
V (x)   (5.19)
 x  0 and x  L
V=0
The “particle in a box” problem is also called the “infinite potential well” or the
“infinite square well potential” problem. A simple classical analogue of this
system could be a ball bouncing elastically between two rigid walls (Fig. 5.3). x=0 x=L
Inside the box, the particle is “free” because the potential, and hence, the Fig. 5.2: Potential
force on the particle is zero. As the particle bounces back and forth, its speed function for a one-
dimensional box.
and kinetic energy remain constant. In the classical picture, the particle could
have any speed inside the box, and any kinetic energy. Let us now solve the
Schrödinger equation for this potential and determine the eigenfunctions and
eigenvalues of the Hamiltonian operator for this system.
5.3.1 Solving the Schrödinger Equation
v
For the (free) particle inside the box, V  0. The Schrödinger equation for the
stationary state wave function (x ) is:
 2 d 2( x ) v
  E( x ) for 0  x  L (5.20)
2m dx 2
Since the walls of the box cannot be penetrated, the particle cannot be found x
beyond the walls (that is in the regions x  0 and x  L ). So, the quantum Fig. 5.3: A ball going
mechanical probability of finding the particle in the region beyond the box, back and forth between
should be zero. Therefore ( x )  0 for x  0 and x  L . Further, the wave rigid walls.
functions for a physical system are always single-valued and continuous. To
satisfy this condition, (x ) should also be zero at the boundaries of the box,
i.e. at x  0 and x  L . So, we can write the boundary conditions for the
wave function (x ) as:
( x  0 )  ( x  L )  0 (5.21)
13
2mE
With these boundary conditions, we now solve Eq. (5.20). We put k 2  in
2
Eq. (5.20) and get the following second order ODE:
d 2( x )
 k 2 ( x )  0 (5.22)
dx 2
You have solved this equation in Sec. 5.2 (Eq. 5.8). From SAQ 2 you know
that ( x )  sin kx and ( x )  cos kx are both solutions of this equation. So we
In general, for can write the general solution for Eq. (5.22) as:
sin x  0  x  n ( x )  A sin kx  B cos kx (5.23)
where
To represent the particle in the box, this wave function must satisfy the
n  0,1,2,3...
boundary conditions given in Eq. (5.21). So, we now impose these conditions
However the wave on the wave function (x ) of Eq. (5.23).
function for a negative
value of n is not linearly ( x  0 )  0  A sin k (0 )  B cos k (0 )  0 or B  0 (5.24)
independent of the wave With B  0 , the expression for the wave function of Eq. (5.23) becomes
function for the positive
value of n. For example ( x )  A sin kx (5.25)
for n = 1: Applying the boundary condition on the wave function at x  L , we get:
 1 ( x )  A sin
 x 
 ( x  L )  0  A sin k (L )  0 (5.26)
 L 
And for n=1 Clearly, A  0 , because if it were, the wave function (x ) would be zero at all
 x  2
  1 ( x )  A sin    values of x and the probability density ( P ( x, t )  ( x ) ) for finding the particle
 L 
So, at any position x would also be zero. This is an unphysical solution and
  1( x )   1( x ) cannot be considered. In Eq. (5.26), since A  0 , we must have (read the
So we can consider only
margin remark):
one of these solutions, sin k (L )  0  kL  n , n  0,1,2,3.... (5.27)
either n=1 or n=1. And
For n  0 , k would be equal to zero and hence, (x ) would be zero at all
so on for all values of n.
values of x, once again an unphysical solution. So we discard the value
of n  0 in Eq. (5.27) and write the possible values of k, which satisfy the
boundary conditions as:
n
k , with n  1,2,3.... (5.28)
L
Notice from Eq. (5.28) that the all values of k are not permitted. k can have
only the following discrete or “quantized” values:
 2 3 
k , , ........
L L L
Therefore, we write k with a subscript n to denote the value of n:
n
kn  , with n  1,2,3.... (5.29)
L
We can now write down the possible stationary state solutions for this system
(which are the eigenfunctions) and determine the corresponding energy
eigenvalues.
5.3.2 Eigenfunctions and Eigenvalues
On substituting the value of k n in Eq. (5.25) we can write the stationary state
14 wave function as:
nx 
 n ( x )  A sin k n x  A sin   , 0  x  L and n  1,2,3....
 L 
(5.30)
n
The subscript n once again indicates the value of k  k n  in the wave
L
function. As you can see, for each value of n we get a different wave function.
Each of these functions 1( x ),  2 ( x ),... is an eigenfunction of the Hamiltonian
operator for the particle in a box. What are the corresponding eigen energies
En?
We determine the energy eigenvalues En from the Schrödinger equation for
the eigenfunction  n (x ) :
 2 d 2 n ( x )
  En  n ( x ) (5.31) Fig. 5.4: The energy levels
2m dx 2
for a particle in a one-
 nx  dimensional box. The
Since  n ( x )  A sin  , we get ground state energy is E1.
 L 
d n ( x )  n   nx 
 A  cos 
dx  L   L 
and
2 2
d 2 n ( x )  n   nx   n 
  A  sin     n (x )
dx 2  L   L   L 
On substituting the second order derivative in Eq. (5.31), we get:
 2   n  
2
n 2 2 2 In general, if we have a
      n ( x )  E n  n ( x )  E n  (5.32)
2m   L  2mL2 bounded system, or in

the terms of classical
So the energy eigenvalues for the quantum mechanical particle in the box mechanics, if the motion
are: is confined to a finite
region of space, solving
 2k n 2 n 2 2 2 the corresponding
En   with n  1,2,3.... (5.33)
2mL2 2mL2 Schrödinger equation
will give us a discrete
As you can see, the particle can have only certain discrete values of energy. energy spectrum.
The energy spectrum is no longer continuous.
From Eq. (5.33) we can write for n = 1:
 2k n 2 2 2
E1   (5.34)
2mL2 2mL2
And therefore
En  n 2E1 (5.35)
It is important to note the following properties of the energy eigenvalues of a
particle in a box:
1. Now that the particle has been confined to a box, you no longer have a
continuous energy spectrum. Its energy can take only the discrete values
given by Eq. (5.33) and shown in Fig. 5.4. This is in contrast to the
15
classical picture in which all finite values of the energy ( E  0 ) are
allowed.
2. The quantization of energy follows from the boundary conditions imposed
on the wave function due to the confinement of the particle.
3. The minimum energy of the particle in the box is not zero but has a finite
2  2
value: E1  . So in the quantum mechanical system, the particle
2mL2
can never be at rest. This is in keeping with the uncertainty principle and
the concept of zero point energy that you have studied in Unit 2. E1 is the
zero point energy for this system.
Notice that the wave function defined by Eq. (5.30) is not normalized. Let us
now determine the normalization constant A in the following example.
Example 5.2
Determine the normalization constant A for the following eigenfunction of a
particle confined in a box of length L:
 nx 
L
 n ( x )  A sin k n x  A sin , 0xL
 nx  dx  L 
 sin2  
0  L 
Solution : We use the normalization condition given by Eq. (3.50) with
1 L 2nx 
  1  cos dx  nx 
2 0  L   n * ( x )  A * sin  . We can write the normalization condition as:
 L 
L
1 L 2nx  L L
 x sin  nx   nx  2 2  nx 
2  2n L  0  A * sin
 L
 A sin
  L
 dx  1  A sin 
 
 L 
 dx  1
0 0
L
 On evaluating the integral we get (see the margin remark):
2
L 2
A 2    1  A 
 2 L
Substituting the value of A from Example 5.2 into Eq. (5.30), we can write the
eigenfunction  n (x ) for a particle confined in a line segment of length L as:
2  nx 
n (x)  sin  with n  1,2,3.... (5.36)
L  L 
The corresponding energy eigenvalues are given by Eq. (5.33). Let us write
down the eigenfunctions and energy eigenvalues for the first three states of
the system (n = 1, 2 and 3 respectively):
2  x  2 2
1( x )  sin ; E1  (5.37a)
L  L  2mL2
2  2x  2 2 2
2( x )  sin ; E2   4E1 (5.37b)
L  L  mL2
2  3x  9 2 2
3 ( x )  sin ; E3   9E1 (5.37c)
L  L  2mL2
These eigenfunctions are shown in Fig. 5.5.
16
3(x)
2(x)
1(x)
x
x=0 x=L
Fig: 5.5: The first three eigenfunctions for a particle in a box.
You can see that each of the eigenfunctions has a node (a point where the
wave function goes to zero) at x  0 and x  L. The number of nodes in the
eigenfunction increases with the value of n. The eigenfunctions corresponding
to different eigenvalues are orthogonal, as you can check for yourself in the
next SAQ.
SAQ 3
Show that the eigenfunctions for the particle in a box are orthogonal.
So how do we reconcile the quantum mechanical picture with our experiences

in the macroscopic world? By working through the following Example you will
get an idea of value of the zero point energy of a typical macroscopic object.
Example 5.3
A toy car of mass 0.2 kg is moving back and forth on a track 1.0 m long.
Treating the car as a particle in a box, what would be the minimum energy of
this car? Suppose it moves with a constant speed of 0.1 ms 1, what is the
value of n, corresponding to this speed?
Solution : Assuming the toy car to be a particle of mass m = 0.2 kg confined to
a region of length L = 1.0 m, can say that its energy is (Eq. 5.33):
n 2 2 2
En   2.5n 22 2 J (i)
20.2 kg 1.0 m 
2
The minimum energy is the zero point energy:

E1  2.5 2 2  2.5  ( ) 2  1.054  10 34 
2
J  2.7  10  67 J
The kinetic energy of the car is
1 1
 
K  mv 2   0.2 kg 0.1ms 1  1.0  10 3 J
2 2
2
(ii)
We calculate the value of n for which the energy of the bound state would be
equal to the kinetic energy of the car. So using :
K  E n  1.0  10 3 J  n 2 E1 and substituting for E1 we get:
17
3
1.0  10 J
n2   n  6  1031
 67
2.7  10 J
So the minimum energy of the car is too small to be measured, and we can
take it to be zero. Also the energy of the classical particle is characterized by
an extremely large quantum number. As you have studied in Sec. (4.6.3), the
Correspondence principle tells us that we need not use the quantum
formalism for such large quantum numbers. We can use Newtonian
mechanics. If n is small, on the other hand, we cannot use classical
mechanics.
Unlike for the free The integer n that is used to index the eigenfunctions  n (x ) and the energy
particle, for the particle eigenvalues E n is called the “energy quantum number” or the “principle
in a box, the
eigenfunctions of the
quantum number”. As you can see in Fig. 5.4, corresponding to any allowed
Hamiltonian operator are
value of n you have a discrete eigen energy (level). The state 1( x ) is the
not the simultaneous ground state eigenfunction and E1 , the ground state energy. The state
eigenfunctions of the  2 ( x ) is the first excited state and E 2 , the energy of the first excited state.
momentum operator, The state  3 ( x ) is the second excited state and E 3 , the energy of the second
which is why we call n
the energy quantum
excited state. You can think that these energy states are like the energy
number. You will see this levels in an atom. Suppose a particle “occupies” the energy state (or energy
for yourself in Terminal level) E1 . On being given an energy ( E 2  E1 ), it can move into the first
Question 8. excited state E 2 . On the other hand, if the particle occupies the first excited
state E 2 , it can emit an energy ( E 2  E1 ) and move to the ground state E1 .
Note from Eq. (5.33) that the eigen energy values depend upon the size of
the box, L and the mass of the particle m. The energy difference between the
corresponding energy levels also depends on L and m. The energy difference
between the nth and (n+1)th energy levels is:
2 2 2 2
E  E n 1  E n  (n  1) 2  n 2   ( 2n  1) (5.38)
2mL2 2mL2
What does Eq. (5.38) tell us? Smaller the region in which a particle is
confined (L), the larger is the difference in energy (E) between the
consecutive energy levels. Conversely, as L increases, the energy separation
decreases. When L is much larger than atomic dimensions, the energy
separation is so small that we approach the classical correspondence limit.
Note that for large L, the zero point energy also tends towards zero, as you
have seen in Example 5.3.
In fact in the limit when mL2 becomes much larger than h2 ( mL2   2 ), which
is typically true for everyday macroscopic objects, the spacing between the
energy levels keeps getting smaller. So the energy of a classical particle in a
box will vary continuously and quantum effects will not be visible.
Let us now try to model the electron in an atom like a particle in a box.
Example 5.4
Calculate the energy levels of an electron confined to an atomic box of size
1.0 Å.
Solution : The energy levels for the electron are calculated using Eq. (5.33)
18 with L  1.0 Å  1.0  10 10 m . Now,
n 2 2 2 n 2h 2
En   , n = 1, 2, 3,…
2 me L2 8m e L2
For n = 1:
E1 
12 h 2

h2

6.626  10  34 Js 2
8me L2 8me L2 89.109  10  31 kg 1.0  10 10 m2
 0.602  10 17 J  38 eV
So the remaining energy levels are:
E n  n 2 E1  38n 2 eV
5.3.3 Time Dependent Wave Function

The complete time dependent wave function  n ( x, t ) corresponding to  n (x )
is
2 nx   iE t 
 n ( x, t )  sin e n (5.39)
L  L 
Using Euler’s Formula, we can write:
1 ik x
sink n x   e n  e  ikn x  (5.40)
2i
n
where k n  , and E n  n . Then we get:
L
1 2 i kn x nt 
 n ( x, t )  e  e i kn x nt   (5.41)
2i L
This is the superposition of two travelling waves having the same frequency
and amplitude. The first term in Eq. (5.40) represents the wave travelling along
the positive x-direction. The second term is for the wave travelling in the
negative x-direction. This is similar to the standing wave solutions of a
vibrating string fixed at both ends.
We next determine the probability density corresponding to these
eigenfunctions.
5.3.4 Probability Density
The probability density for a stationary state is time independent and we can
write for the nth eigen state:
2 nx 
Pn ( x, t )  Pn ( x )   n ( x ) 2  sin2   (5.42)
L  L 
From Eq. (5.42), you can see that Pn (x ) depends on the position of the
particle in the box. In Fig. 5.6 below, we have plotted the values of Pn (x )
corresponding to first three eigenfunctions of the system (for n = 1, 2, 3).
The probability of finding the particle is maximum around the maxima (humps)
of the probability distribution and is zero at the nodes. So the probability of
finding the particle in any finite region of the box also changes with the
eigenfunction.
19
P3 ( x )
P2 ( x )
P1( x )
x
x=0 x=L
2
Fig. 5.6: The probability density (  n (x ) ) as a function of position for the first
three eigenfunctions of a particle in a box.
To understand, let us work through the following example.

Example 5.5
Calculate the probability of finding the particle between
i) x  L / 4 and x  3L / 4 , and
ii) x  0 and x  L / 4
when the particle is in the ground state.
Solution : We use Eq. (3.15) to calculate the probability of finding the particle
between two points x1 and x2 .
2 x
i) With x    * x   sin for the ground state, x1  L 4 and x2  3L 4 in
L L
Eq. (3.15) we calculate the probability as:
3L / 4 2
3L / 4
 2 x   1  2x 
P 

 L
sin
L  
dx   
L  L/4 
1  cos 
L 
 dx
L/4 
3L / 4
1 L 2x 
  x sin
L 2 L L / 4
1  3L L  L 2   3L  L 2  L  
    sin    sin   
L  4 4  2 L  4  2 L  4 
1 L L 3 L  L 1 2  1 1
  sin  sin        

L  2 2 2 2 2  L  2 2   2  
2 x
ii) For x    * x   sin , x1  0 and x 2  L 4 the probability is:
L L
L/4 2 L/4
 2 x  1 L 2x 
P 
 L 
sin
L 
dx   x 
L 2
sin
L  0
0 
20
1 L L 2  L   1  L L  1 1 
   sin       
L  4 2 L  4   L  4 2  
 4 2 
You may now like to work out an SAQ on calculating the probability.
SAQ 4
Calculate the probability of finding the particle between x  L / 4 and

x  3L / 4 , when the particle is in the first excited state.
Let us compare this with the classical picture. Classically, the particle bounces
back and forth between the walls. Since the kinetic energy of the particle is
constant, it moves at a constant speed between collisions with the walls.
Therefore, it spends the same amount of time in all intervals which have an
equal length. So the normalized probability per unit distance, of the particle
being found anywhere in the box is just 1/L which is independent of the
position of the particle. This is clearly not the case for the quantum mechanical
particle. However, when n is very large the wave function begins to have so
many nodes and humps that in the limit as n   each position of the particle
is equally probable. Then the quantum mechanical probability distribution
begins to look like the classical distribution.
For a large value of n, you would observe only the average value of the
probability density which is
2  nx  2  1 1
Pn ( x, t )  sin2      (5.43)
L  L  L 2 L
(This is because the average of the square of the sine function over one or
more cycles is always half.)
Our understanding of the way charge carriers behave in materials is also an
application of quantum mechanics. Solving the Schrödinger equation using the
exact potentials experienced by the charge carriers is very complicated.
However, it is often possible to understand the observed phenomena using
simple models of the potential which are easier to solve.
Some of you may have read (or you can read now) popular science articles
about “nano” sized materials like quantum wells, quantum wires and quantum
dots and how they have exciting applications, mainly because they exhibit
quantum effects. In the next section we study about quantum dots. We can
understand the special properties of a quantum dot by modelling a charge
carrier inside the quantum dot as a particle in a box.
5.4 QUANTUM DOT

Quantum dots (sometimes called artificial atoms) are typically semiconductor
(or metal) particles with sizes ranging from 1 to 100 nm. Their properties are
very different from the bulk properties of the materials of which they are made
and depend on the size of the particle.
You have all studied about conductors, insulators and semiconductors in your
school physics courses. As you have studied in Bohr’s atomic model, in an
21
isolated atom, the electrons occupy discrete energy levels. Whereas in bulk
materials made up of several atoms, we have continuous energy bands which
are collections of available energy states and energy gaps between these
bands.
The quantum dot is a very tiny object. Therefore, the charge carriers inside the
dot are confined within a very small region, very much like a particle in a rigid
box. The simplest approximation for a quantum dot is an infinite potential well
where all the charge carriers are confined. So like a particle in a box (or an
electron in an atom), the charge carriers in the quantum dot occupy discrete
energy levels.
While in a bulk semiconductor material you would have energy bands, in a
quantum dot made out of the same material, the energy bands are replaced
by discrete energy levels, as shown in Fig. 5.7.
Ideally you should think
of an electron in a Conduction
quantum dot as an Band
electron confined in a
three dimensional box. Energy
Band Gap
To understand the Levels
physical properties, here
we assume it to be a Valence
one-dimensional box. Band
(a) (b)
Fig. 5.7: Energy levels of a) a bulk semiconductor; and b) a quantum dot of the
same semiconductor.
Notice that for the
electron in the Absorption and emission of light by these quantum dots (as in any atom or
semiconductor quantum molecule) takes place by the transition of charge carriers between these
dot we are writing the energy levels.
mass of the electron as
m e * and not me. This is Suppose an electron occupying the energy state j in a quantum dot of size D
because in a
makes a transition to state i. Let us assume that the quantum dot is a one-
semiconductor, the
electron has an effective dimensional rigid box of size D (read the margin remark). Using Eq. (5.33) for
mass m e * (which is the energy of the electron, we can write the expression for the energy emitted
much less than the free in this transition as:
electron mass m e ) which
decides to a large extent j 2 22 i 2 2 2
how the electron
E ji  E j  E i  
2me D 2 2m e D 2
behaves inside a
semiconductor! It is
because of this that the  22
 ( j2  i2) (5.44)
de Broglie wavelength in 2me * D 2
a semiconductor is
typically of the order of This energy difference depends on the size of the dot D. If D increases, other
nanometers, and not
parameters remaining the same, the value of E ji decreases. This is depicted
angstroms as for a
in Fig. 5.8 below.
conduction electron in a
metal.
22
Eji
j
i Eji
i
L1 L2
(a) (b)
Fig. 5.8: Energy level variation with size. For a box of size a) L1; and b) L 2 with
L1  L2 , the difference between the energy levels E ji  E ji .
This energy difference is emitted from the dot as a photon of light. What is the
wavelength of this photon? If the frequency of the emitted photon is  and  its
wavelength (   c  ), then:
hc  2 2
E ji  h   ( j2  i2) 
(5.45)
 2me D 2
So the wavelength of the emitted photon is directly proportional to the size of
the dot D. As the size of the dot increases, the wavelength of the emitted
photon also increases. As D increases, even though the material of the dot
may remain the same, the wavelength of the emitted lights changes from the
blue towards the red end of the visible spectrum! So by tuning the size of the
quantum dot in the process of preparation, you can obtain a specific
wavelength of the emitted light.
A real quantum dot is definitely more complicated than what we have

described here. However as you can see it is possible to understand certain
salient features of light emission from QDs using the particle in a box model. A
more realistic estimate of the numbers is obtained by considering the
confinement of an “electron-hole” pair also called an “exciton” inside a
spherical quantum dot.
Several other applications of the quantum dot are also related to the fact that
its energy levels can be tuned by its size. Quantum dots are used in single-
electron transistors, solar cells, LEDs, lasers, single-photon sources, cell
biology research, microscopy, and medical imaging.
 Quantum dots (sometimes called artificial atoms) are typically semiconductor

(or metal) particles with sizes ranging from 1 to 100 nm with properties that
are vastly different from the bulk material of which they are made and depend
on the size of the particle.
 The simplest approximation for a quantum dot is an infinite potential well
where all the charge carriers are confined to a region and occupy discrete
energy levels, like electrons in an atom.
23
 A change in the size of the dot changes the spacing between the energy
levels. This changes the energy released in the transition of the electron
between the energy levels. So the wavelength of the light emitted by the
quantum dot changes with the size of the dot.
5.5 STEP POTENTIAL AND SCHRÖDINGER
EQUATION
A step potential is shown in Fig. 5.9. You can see that we can define it
mathematically as:
 0, for x  0
V (x)   (5.46)
 V0  0, for x  0
V(x)
V  V0
I II
V 0
0 x
Fig. 5.9: A step potential. Note that the potential is zero in Region I and it has a
non-zero value (V0 ) in Region II.
In Fig. 5.9, note that the step potential V (x ) is zero in the region x  0 and
non-zero (equal to V0 ) in the region x  0. Note also from Fig. 5.10 that the
step potential is discontinuous at x  0.
You may like to ask: What does such a potential mean physically? To
understand this, and to get a physical picture of the potential, let us see what
happens to a particle subjected to the step potential.
Let us consider a particle of mass m having kinetic energy, E, and moving
from left to right along the x-axis as shown in Fig. 5.10.
(a) (b)
Fig. 5.10: Classical picture of a particle with energy (a) lower than the step
potential V0 and (b) higher than the step potential V0 .
Let us first ask: What would happen if the particle was a classical particle?
From classical mechanics, you know that if the energy (E) of the particle were
24 less than the step potential V0 , that is, if E < V0 , then the particle would not be
able to enter Region II from Region I (see Fig. 5.10a). A classical particle for
which E < V0 , would be reflected at the step. However, if the energy of the
classical particle was greater than V0 , it would be able to cross the step
potential and move into Region II (see Fig.5.10b).
Let us now ask: What would happen to a quantum particle subjected to this
potential? Refer to Fig. 5.11. Note that the particle is incident from Region I of
the step potential (V0 ). Note that in the Region I, the energy (E) of the particle
is purely kinetic energy, while in Region II, the energy is partly kinetic and
partly potential. Now recall from Unit 3, how quantum mechanics differs
fundamentally from classical physics. According to quantum mechanics, the
wave function of the particle has finite value in Region II, which means that
there is a finite probability of finding the particle in Region II even when the
particle has less energy than the potential! This is the result we get when we
solve the Schrödinger equation.
V (x )
V  V0
I II
(a)
E V 0
0 x
V (x )
V  V0
I II (b)
E
V 0
0 x
Fig. 5.11: A particle with energy E is incident from Region I and subjected to a
step potential V0 at x  0. We consider two cases, a) when E < V0 ;
b) when E > V0 .
We now write the time independent Schrödinger equation for a particle in a

step potential defined by Eq. (5.46). Note that there are two regions, Region I
( x  0) in which the potential is zero and Region II ( x  0) in which it is V0 .
So, we write the Schrödinger equation for a particle of mass m in the two
regions (see Fig. 5.11) as:
 2 d 21
  E 1 x 0 (5.47a)
2m dx 2
 2 d 2 2
  V0  2  E  2 x 0 (5.47b)
2m dx 2
25
We solve the Schrödinger equation for the two regions [Eqs. (5.47a and b)].
We also have to apply the boundary conditions to get acceptable wave
functions. Note that we have denoted the solutions of the Schrödinger
equation in Regions I and II as 1 and  2 in Eqs. (5.47a and b). The
boundary conditions for this problem are: At x  0
1 ( x  0)   2 ( x  0) (5.48a)
 d1   d 
and     2 (5.48b)
 dx  x 0  dx  x 0
Now, there are two interesting possibilities regarding the energy of the particle
with respect to the step potential. The first case is when the energy of the
particles is less than the step potential V0 and the second, when it is greater.
Therefore, in Secs. 5.6 and 5.7, we will solve Eqs. (5.47a and b) for the
following two cases:
Case 1: The energy E of the particle is less that V0 (Fig. 5.11a)
Case 2: The energy E of the particle is greater than V0 (Fig. 5.11b)
5.6 ENERGY (E) OF THE PARTICLE IS LESS

THAN V0
Let us solve Eqs. (5.47a and b) for Regions I and II.
Region I:    x  0 for which V  0
Let 1( x ) be the solution of the Schrödinger equation for the particle in
Region I. Since the potential is zero in this region, the particle is free and the
corresponding time independent Schrödinger equation is:
  2 d 21
 E1 (5.49a)
2m dx 2
d 21 2mE
  1  0 (5.49b)
dx 2 2
d 21
  k121  0 (5.49c)
2
dx
2mE
where k1  (5.49d)
2
You know [Eq. (5.10)], that the solution of Eq. (5.49c) is:
1 ( x )  A e ik1x  B e  ik1x (5.50)
You know that, the term A e ik1x in Eq. (5.50) represents a wave of amplitude
A travelling in the positive x-direction while the term B e  ik1x represents a
26
wave with amplitude B travelling in the negative x-direction. The second term
in Eq. (5.50) is a wave reflected at the step potential (see Fig. 5.12).
Ae ik1x Be ik1x
I Ce x II V  V0
E
V 0
0
x
Fig. 5.12: The solutions [Eqs. (5.47a and b)] of the Schrödinger equation for a
free particle in Regions I and II of the step potential.
So, the solution of Eq. (5.47a) is a linear combination of wave functions

representing an incident wave and a reflected wave.
Let us now solve Eq. (5.47b).
Region II:  0  x    for which V  V0

Let 2 ( x ) be the solution of the Schrödinger equation for the particle in
Region II. The time independent Schrödinger equation for Region II is given
as:
 2 d 2 2
  V0  2  E 2 (5.51a)
2m dx 2
d 2 2 2m
  (V0  E ) 2  0 (5.51b)
2
dx 2
2m(V0  E )
We put  (5.51c)
2
so that we can write Eq. (5.51b) as:
d 2 2
  2 2  0 (5.51d)
dx 2
The solution of Eq. (5.51d) is:
 2 ( x )  Ce x (5.52)
In Eq. (5.52), the term Ce x represents an exponentially decreasing wave

function for positive values of x (Fig. 5.12). This means that the wave function
is finite in Region II of the step potential. Therefore, the probability of finding
the particle in Region II is finite. The general solution of Eq. (5.51d) is of the
form  2 ( x )  Ce  x  De  x .
Can you say: Why have we not included the term Dex in Eq. (5.52)? Recall
what you have learnt in Unit 3 about physically acceptable wave functions.
27
x
You know that it is because the term De represents an exponentially
increasing wave function in the Region II. It is not physically meaningful since
an acceptable wave function must remain finite as x  . Thus, we must have
D  0.
Let us now put the solutions of the time independent Schrödinger equation for
the step potential defined by Eqs. (5.49c and Eq. 5.51d), at one place. The
solutions are:
1 ( x )  A e ik1x  B e  ik1x for Region I: x < 0 (5.50)
 2 ( x )  Ce  x for Region II: x > 0 (5.52)
The next step, as you know, is to apply the boundary conditions at x  0 to

determine the constants A, B and C. Applying the first boundary condition
[Eq. (5.48a)], we get:
AB C (5.53a)
Remember that A and B are the amplitudes of the incident wave and the
reflected wave in Region I and C represents the amplitude of a wave that
decreases exponentially in Region II. Applying the second boundary condition
[Eq. (5.48b)], we get:
 d1   d 2 
  
 dx   dx 
 Aik1  Bik1  C

i
 AB  C (5.53b)
k1
Let us now solve Eqs. (5.53a and b) to determine the constants A, B and C.
Adding Eqs. (5.53a and b), we get:
 i 
2 A  1  C
 k1 
 2k 1 
 C A (5.54a)
 k1  i 
Subtracting Eq. (5.53b) from Eq. (5.53a), we get:
 i   2 k1 
2B   1  C  C   B (5.54b)
 k1   k 1  i 
Equating the LHS of Eqs. (5.54a and 5.54b) we can write:
 k  i 
B   1  A (5.54c)
 k1  i 
Thus, we have obtained the constants B and C in terms of A. Let us now

28 interpret these results.
5.6.1 Reflection and Transmission Coefficients

Let us first ask: What is the incident probability density, i.e., the probability of
finding the particle in Region I? This is known as the incident probability
density.
From Unit 3, you know that the probability density in any region is defined as:
P ( x )    ( x ) ( x ) (5.55)
Note that from Eq. (5.50) that the wave function for the incident wave is
1 ( x )  A e ik1x . Therefore, the incident probability density is given as:
2
P1 ( x )  1 ( x ) 1 ( x )  ( Ae  ik1x )( Ae ik1x )  A
We would now like you to calculate the reflected and transmitted probability
densities corresponding to the reflected and transmitted wave functions using
Eqs. (5.50 and 5.52). Try SAQ 5.
SAQ 5
Determine the reflected and the transmitted probability densities for a step
potential.
Recall Eq. (3.31a) of Sec. 3.3.2 from which the probability current density is
given by:
   d d  
J   
2mi  dx dx 
i   d d  
 J     (5.56)
2m  dx dx 
Substituting 1 ( x ) and  2 ( x ) in the Regions I and II, respectively, we obtain
the reflected and transmitted probability current densities, respectively. Let
Jr be the probability current density of the reflected wave. Let us take up an
example to calculate it.
Example 5.6
What is the probability current density ( J r ) of the reflected wave?
Solution : We use Eq. (5.56) to determine Jr .

i   d d  
Since J    
2m  dx dx 
 d Be  ik1x  d Be  ik1x 


i 
Jr   Be  ik1x   Be  ik1x  
2m  dx dx 

 Jr  
i
B B(ik1 )  B B(ik1 )   k1 B B
2m m
29
k k
Hence, J r   1 B B   1 B 2
m m
You can now calculate the incident probability current density corresponding
to the incident wave function using Eqs. (5.50 and 5.46). Try SAQ 6.
SAQ 6
Determine the probability current density (Ji ) of the incident wave.
The reflection coefficient (R) is the probability that an incident particle is

reflected from the step potential at x = 0. It is defined as the ratio of the
magnitudes of the probability current densities of the reflected wave and the
incident wave:
Jr
R (5.57)
Ji
The transmission coefficient (T) is the probability that the incident particle is
transmitted and is defined as the ratio of the magnitudes of the probability
current densities of the transmitted wave and the incident wave:
Jt
T  (5.58)
Ji
Using Eq. (5.57), we can determine the reflection coefficient for a step
potential for E  V0 :
2
B
R (5.59)
2
A
 k  i   k 1  i  2
From Eq. (5.54c), we can see that B 2   1   A which means
 k 1  i   k 1  i 
2 2
that B  A . Substituting this result in Eq. (5.59), we get R  1.
Since the incident wave is either reflected or transmitted without any

absorption, the sum of the reflection coefficient (R) and the transmission
coefficient (T) should be unity:
R T  1
Therefore, for R  1, we can see that the transmission coefficient is zero:
T  0
This result seems to agree with the predictions of classical physics that when
a particle possesses lower energy than the potential step, the particle cannot
cross it! However, from Eq. (5.52) you have learnt that there is a finite
probability of finding the particle in Region II since the wave function ( 2 ) in
the region is non-zero.
You have already encountered a similar problem when you were dealing with
bound state (particle in a box) where the wave function is non-zero in the
classically forbidden region. So, how do we reconcile the two seemingly
contradictory results?
30
In order to understand the result, it is important to remember how we have
defined the step potential. You are aware that the step potential is an idealized
potential and beyond x > 0, the potential (V0 ) may extend to infinite distance.
Hence, whatever is incident towards the barrier will eventually be reflected and
it is for this very reason that we got R  1. However, since the wave function is
non-zero in Region II [Eq. (5.52)], we can conclude that the particle turns back
eventually but before turning, it penetrates into classically forbidden region.
Thus, from the above result we can conclude that:
1. The reflection of incident wave at the step potential is total when E  V0 ;
2. In the Region II, the wave function is non-zero. Therefore, there is a finite
probability (  2 ( x )  2 ( x )  C *Ce 2 x ) of finding the particle in Region II;
3. The wave function penetrates into classically forbidden region, x > 0.

However, eventually it is completely reflected.
Let us explain the point 3 further.
Penetration in the Classically Forbidden Region
One of the most interesting outcomes of solving the Schrödinger equation for
the step potential is that the wave function in the Region II (x > 0) is non-zero.
This means that there is a finite probability of finding the particle in Region II
even if E  V0 .
Note that according to classical mechanics, particles with energy
E  V0 cannot penetrate the step potential since it would lead to negative
kinetic energy.
According to classical physics, when the initial energy of the particle is less
than the height of the potential, the particle is reflected back with its original
speed. The quantum particle’s behaviour is in sharp contrast to the classical
result. There is a finite probability of finding the particle in Region II given by
Eq. (5.52). Let us now take up the case when the energy of the incident
particle is greater than the height of the step potential.
5.7 ENERGY (E) OF THE PARTICLE IS GREATER

THAN V0
The case when the energy of the particle is greater than the potential step
(E  V0 ) is shown in Fig. 5.13. Again, we solve Eqs. (5.47a and b) in Regions
I and II.
Region I:    x  0 for which V  0, the particle is free and the solutions

remain the same as in Sec. 5.2 and are given by Eqs. (5.49c and 5.50),
respectively.
Region II: For 0  x   , the time independent Schrödinger equation is given

as:
d 2 2 2m
 (E  V0 ) 2  0
2
dx 2
31
2
d 2
 k 2 2  0 (5.60)
dx 2
2m(E  V0 )
where k 2 
2
The solution of Eq. (5.15) is:
 2 ( x )  F e ik 2 x (5.61)
Ae ik1x V (x ) F eik2 x
B e  ik1x
K  E  V0
E K V 0 V0
0 x
Fig. 5.13: The solutions of Schrödinger equation given by Eqs. (5.50 and 5.61).
The solution of Eq. (5.60) exhibits an oscillatory nature of the wave function
(Fig. 5.13). The general solution of Eq.(5.60) is  2 ( x )  F e ik 2 x  G e  ik 2 x . The
term F e ik2 x represents a wave travelling along the positive x-axis in the
Region II, while a term like G e  ik 2 x would mean a wave travelling along the
negative x-axis in Region II. However, since the particle is incident only from
the left we can assume that G  0.
Let us now apply the first boundary condition: At x  0, [Eq. (5.48a)], we have
1( x )   2 ( x ). Hence, from Eqs. (5.50 and 5.61), we get:
AB F (5.62a)
Here, F represents the amplitude of the wave in Region II. Similarly, applying
the second boundary condition [Eq. (5.48b)], we get:
 d 1   d  2 
  
 dx   dx 
 Aik1  Bik1  iFk2
k
 AB  2 F (5.62b)
k1
Adding Eqs. (5.62a and b), we get:
32
k k   2k1 
2 A   1 2 F  F    A (5.63)
 k1   k1  k 2 
Subtracting Eq. (5.62b) from Eq. (5.62a), we get:

The wave function in
k k  k k 
2B   1 2 F  B   1 2  A (5.64) region II is:
 k1   k1  k 2 
   2 ( x )  F e ik2 x
Once again, let us understand the results we have obtained for Case II.
 *   *2 ( x )  F * e 2
 ik x
5.7.1 Reflection and Transmission Coefficients And

d
Let us now determine the reflection and the transmission coefficients for the

d
F e ik x 
2
case when the energy of the particle is greater than the potential: E  V0 . dx dx
From Eq. (5.47), the reflection coefficient (R) is given by:  ik 2 e ik2 x F
B 2  k1  k 2  2 d *
R
Jr
   (5.65) 
d
F * e  ik x 
2
dx dx
Ji A 2  k1  k 2 
 ik 2 e ik2 x F *
Similarly, we determine the transmission coefficient (T) using Eq. (5.58) (read
Substituting in
the margin remark): Eq. (5.46) we get the
k 2 2 probability current
Jt F density of the
T   m (5.66a) transmitted wave as:
Ji k1 2
A
m i 
Jt   ik 2 FF *
2m 
Substituting Eq. (5.58) in Eq. (5.66a), we get:
2  ik 2 FF * 
k  2k1  4k1k 2
T  2    (5.66b)
k1  k1  k 2  k1  k 2 2
You may like to attempt an SAQ on reflection and transmission coefficients.
SAQ 7
Show that R  T  1 for E  V0 .
Let us now write the wave function for Region I in terms of the incident
amplitude (A).
Substituting Eq. (5.64) in Eq. (5.50), we get the wave function for Region I
when E  V0 as:
 k  k 2  ik1x 
1( x )  A e ik1x  1 e  for x < 0 (5.67)
 k1  k 2 
Attempt SAQ 8 to write the wave function for Region II.
SAQ 8
Write the wave function for Region II in terms of the incident amplitude when
E  V0 .
Let us consider an example to apply the results obtained so far. 33

Example 5.7
A particle encounters a step potential of height V0. What is the reflection and
transmission coefficient if E  2V0 ?
Solution : We use Eqs. (5.65 and 5.66b).

2 2
 k  k2   E  E  V0 
Since R   1    
 k1  k 2   E  E V 
 0 
2
 2V0  2V0  V0 
 R 
 2V  2V  V 
 0 0 0 
 R  0.029
This means that about 3% of the wave is reflected even when the energy of the
particle is greater than the height of the potential.
Similarly, from Eq. (5.66b), the transmission coefficient is given as:
4k1k 2
T
k1  k2 2
4 E E  V0
 T
 E  E  V0 2
4 2V0 V0
 T 
 2V0  2V0  V0 2
T  0.971
Thus, you can see that R  T  1.
Now we will summarize the concepts you have studied in this Unit.
5.8 SUMMARY
 The time independent Schrödinger equation for a quantum mechanical
free particle of mass m is:
 2 d 2( x )
  E( x )
2m dx 2
The general solution of Eq. (5.4) is:
2mE
( x )  Ae ikx  Be ikx ; with k 2 
2
The quantum number k is real and the free particle can have any non-
negative value of energy:
 2k 2
E 0
2m
The energy of the particle does not change with time.
The time dependent wave function for the free particle with an energy
E   is:
( x, t )  Ae i kx t   Be i kx  t 
34
 A particle is confined within a length segment lying between x = 0 and
x = L by the following potential function(infinite potential well)
0 for 0  x  L
V (x)  
 x  0 and x  L
The time independent Schrödinger equation for the particle is:

 2 d 2( x )
  E( x ) for 0  x  L
2m dx 2
The boundary conditions to be satisfied by the wave function are:
( x  0 )  ( x  L )  0
The normalized eigenfunctions of the Hamiltonian are

2  nx 
n (x )  sin  , 0  x  L for n  1,2,3....
L  L 
The corresponding energy eigenvalues are:
n 2 2 2
En  with n  1,2,3....
2mL2
n is called the energy quantum number.
The energy of a particle bound inside an infinite potential well can have
only discrete values.
2  2
The minimum energy of the particle is not zero, it is E1  .
2mL2
The complete time dependent wave function  n ( x, t ) corresponding to
 n (x ) is
2  nx  iE nt
 n ( x, t )  sin e

L  L 
The eigenfunctions of the system are orthogonal.
 Quantum dots (sometimes called artificial atoms) are typically
semiconductor (or metal) particles with sizes ranging from 1 to 100 nm
with properties that are vastly different from the bulk material of which
they are made and depend on the size of the particle.
The simplest approximation for a quantum dot is an infinite potential

well where all the charge carriers are confined to a region and occupy
discrete energy levels, like electrons in an atom.
A change in the size of the dot changes the spacing between the
energy levels. This changes the energy released in the transition of the
electron between the energy levels. So the wavelength of the light
emitted by the quantum dot changes with the size of the dot.
 A one-dimensional step potential is an idealized potential function

which is zero for negative values of the independent variable and
constant for positive values of the independent variable. The step
potential is defined mathematically as:
35
0 for x  0
V (x)  
V0  0 for x  0
It is discontinuous at the point x  0 and constant on each side of the

discontinuity.
 The time independent Schrödinger equation for the step potential

is:
 2 d 21
  E 1 x 0
2m dx 2
 2 d 2 2
  V0  2  E  2 x 0
2m dx 2
with the boundary conditions that at x  0,
 d 1   d  2 
1 ( x )  2 ( x ) and   
 dx   dx 
The Schrödinger equation is solved for two cases: 1) When the energy,
E, of the particle is less than V0 ; 2) when the energy, E, of the particle
is greater than V0 .
 When E  V0 , the solutions of the Schrödinger equation are:
1 ( x )  A e ik1x  B e  ik1x for    x  0
 2 ( x )  Ce  x for 0  x  
2mE 2m(V0  E )
where k1  2
and  
 2
 k  i   2k1 
and B   1  A, C    A
 k1  i   k1  i 
The reflection coefficient (R) is the ratio of Jr and J i , where Jr and

Ji are the probability current densities of the reflected and incident
waves, respectively. The transmission coefficient (T) is the ratio of
Jt and J i , where Jt and Ji are the probability current densities of the
transmitted and incident waves, respectively. The reflection and
transmission coefficients are 1 and 0, in this case. Thus,
R  T  1. The particles with energy lower than the step potential have
finite probability of penetrating through the classically forbidden region.
 When E  V0 , the solutions of the Schrödinger equation are:
1 ( x )  A e ik1x  B e  ik1x for    x  0

36
and  2 ( x )  F e ik 2 x for 0  x  
2m(E  V0 )
where k2  ,
2
k k   2k1 
and B   1 2  A, F    A
 k1  k2   k1  k 2 
In this case, the reflection and transmission coefficients are:
2
k k  4k1k2
R   1 2  and T
 k1  k2  k1  k2 2
such that R  T  1. A particle with energy higher than the height of
the step potential can get reflected back at the point of discontinuity.
5.9 TERMINAL QUESTIONS

1. A free quantum mechanical particle has a wave function given by
( x, t )  Ae i ( k0 x 0t ) , k 0  4.0  1011 m-1 and 0  5.0  1015 s 1.
Determine its momentum and energy.

2. Determine the eigenvalues of the operator Hˆ  i for the free particle
t
wave function ( x, t )  Ae i kx  t  .
3. Let us model the proton in an atomic nucleus as a particle in a box.

Suppose the proton is confined to a box of size 2.0  10 14 m (size of a
nucleus). Calculate the frequency of the emitted photon if the proton
makes a transition from the first excited state to the ground state.
4. Suppose an electron confined to a box emits photons. The longest

wavelength that is registered is 400.0 nm. What is the width of the box?
5. The wave function of a particle of mass m inside an infinite square well of

width 2a (  a  x  a ) is given by:
3x   3x 
( x )  A cos   B sin 
 2a   2a 
Obtain the values of A and B and the eigen energy corresponding to the
above eigenfunction.
6. A particle in a one-dimensional box has an energy of 10 eV in its first

excited state. Calculate its ground state energy and the energy in the
second excited state.
7. Calculate the expectation value of p 2 for the wave function

( x, t )  Ae i kx  t  . 37
8. Show that the eigenfunctions  n (x ) for a particle in a box are not the
eigenfunctions of the momentum operator.
2  nx 
9. For the stationary state  n ( x )  sin  show that
L  L 
i)  pˆ x  0
n 2 2 2
ii)  pˆ x 2 
L2
10. A 1.1 mA beam of electrons enters a sharply defined boundary with a

velocity 2.5 10 6 ms 1, and then its velocity reduces to 1.0 10 6 ms 1, due
to the difference in potential. Determine the transmitted and reflected
currents.
11. Determine the expression for the reflection coefficient when E  V0 .
12. A beam of particles with energy 7eV is sent towards a step potential of
height 5eV. What fraction of the beam is reflected back?
13. A particle with energy 5 eV is incident on a potential with energy 7 eV.

Calculate the distance in which the probability of finding the particle goes
to 0.05 as it goes into the classically forbidden region.
14. A beam of free electrons are moving from left to right and suddenly
encounters a step potential of height 8 eV. What should be the
approximate energy (E) of the electrons such that 20% get reflected?
5.10 SOLUTIONS AND ANSWERS
Self-Assessment Questions
d 2
1. We use Eq. (5.4) with   Ae ikx  Be ikx . We first evaluate :
dx 2
d d
 Ae ikx  Be ikx   ik Ae ikx  Be  ikx 
dx dx
d 2 d
 ik Ae ikx  Be  ikx   ik 2 Ae ikx  Be  ikx   k 2 ( x )
dx 2 dx
d 2
Substituting   k 2( x ) in Eq. (5.4) we get,
2
dx
 2 d 2  2k 2  2k 2
    x   E  x   E 
2m dx 2m 2m
38 2. i) To show that x   sin kx is solution of Eq. (5.4), we have to show:

2 d 2 2 2
 sin kx   E sin kx with E   k (i)
2m dx 2m
We have
d2
2
sin kx   d  d sin kx   k d  cos kx   k 2 sin kx (ii)
dx dx  dx  dx
d2
Substituting for sin kx  from Eq. (ii) into the LHS of Eq. (i) we see
dx
that:
 
  2k 2  2 2
2  sin kx  E   k
  k 2 sin kx  
2m  2m  2m
 
So Eq. (i) is true for  x   sin kx .
ii) To show  x   cos kx is a solution of Eq. (5.4), we have to show:
2 d 2 2 2
 cos kx   E cos kx where E   k (iii)
2m dx 2m
d2
Since cos kx   k 2 cos kx , we get:
dx 2
 
  2k 2  2 2
2  cos kx  E   k
 .  k 2 cos kx  
2m  2m  2m
 
So Eq. (iii) is true for  x   cos kx .
3. For any two eigenfunctions  n x  and  m (x ) to be orthogonal we must

show that
L
  n x   m ( x ) dx  0 when m  n
0
Substituting for  n x  and  m (x ) in this integral we get:
L
 2 nx   2 mx 
  sin   sin  dx
L  L   L  L 
0
2 sin A sin B
L  cos( A  B )  cos( A  B )
2 nx   mx  dx
 sin 
  sin  
L  L   L 
0
1 
L
n  m x  cos n  m x  dx (see the margin remark)

L  
cos
L L 

0
L

1 L
sin
n  m  x  L sin n  m  x 
 
L  n  m   L n  m   L 0
39
 0 (because m and n are integers
sinn  m   sinn  m    0 )
4. To calculate the probability for the first excited state we use Eq. (5.42) with
2 2x
 x    * ( x )  sin , x1  L 4 and x2  3L 4 . So the probability is:
L L
3L / 4 2
 2 2x  1 3L / 4  4x 
 P   
 L sin L  dx  L  1  cos L  dx
L/4  
L/4  
1  3L L  L 4   3L  L 4  L 
    sin   sin  
L  4 4  4  L  4  4 L  4  
1 L L L  1
  sin 3  sin  
L  2 4 4  2
5. The reflected (Pr) and the transmitted (Pt) probability density is given as:
2 2 2 2
Pr ( x )  Be ik1x B and Pt ( x )  Cex  C e  2x
6. The probability current density of the incident wave can be obtain from
Eq. (5.56):
i  * d d * 
J    
2m  dx dx 
The incident wave function is ( x )  Ae ik1x , then the conjugate wave

function is ( x )  Ae  ik1x .
Ji  

i 
Ae ik1x 
 d Ae ik1x
 
 Ae ik1x
 
d Ae ik1x
  

2m  dx dx 
 
Ji  
i 
2m
 k k

A Aik1   A A( ik1)  1 A A  1 A
m m
2
7. From Eqs. (5.65 and 5.66b) we have:

2 2 2
Jr B k k  k  2k1  4k1k2
R    1 2  and T  2   
Ji A
2 k
 1  k 2 K1 1 k  k 2 k1  k2 2
2
 k  k2  4k1k 2 k  k 2  2  4k1k 2
or R T   1    1
 k1  k 2  k 1  k 2  2 k 1  k 2 2
k1  k 2 2
 R T  1
k1  k 2 2
8. Substituting Eq. (5.63 in Eq. (5.61), the wave function for Region II for
E  V0 :
 2k1  ik 2 x
 2 ( x )  A   e for x > 0
 k1  k 2 
40
Terminal Questions
1. Using Eq. (5.3) we can write
p  k  1.054  10 34 Js  4.0  1011 m 1   4.2  10 23 kg ms 1
E    1.054  10 34 Js  5.0  10 5 s 1   5.3  10 19 J

2. The eigenvalue of Hˆ  i  is determined using Eq. (4.35). With Dˆ  Hˆ
t
we can write:
Ĥ  E (i)
Calculating the derivative, we get:

i Ae i kx  t    i  i Ae i kx  t     (ii)
t
Substituting for Ĥ in Eq. (i) we get
    E (iii)
So the eigenvalue of Ĥ is  .
3. The ground state energy of the proton in the nucleus is calculated with
n = 1, mass of the proton is m p  1.673  10 27 kg and L  2.0  10 14 m :
E1 
h2

6.626  10  34 Js 2
8 m p L2 8  1.673  10  27 kg 2.0  10 14 m2
 0.82  10 13 J  0.51MeV
The first excited state energy: E 2  n 2E1  4   0.51MeV  2.1MeV
When there is a transition from E2  E1, the emitted photon has an

energy: E ph  E 2  E1  2.1  0.51MeV  1.6 MeV
The frequency of the photon is,
E ph 1.6  1.6  10 J  0.39  1021 Hz

13

h

6.626  10 Js
 34
4. The longest wavelength will correspond to the minimum energy difference,

hc
because E  h  . The energy difference, E , defined in Eq. (5.38)

will be minimum for the lowest value of n which is n=1. Substituting n  1
in Eq. (5.38) we can write:
3 2 2 3h 2
Emin   (i)
2me L2 8me L2
41
hc
Using E  in Eq. (i) we get

hc 3h 2  3 h 
 or L2    (ii)
 2
8me L  8mec 
So the size of the box, L, with   400 nm  4.0  10 7 m is:
 
 3  6.626  10  34 Js  4.0  10  7 m   1/ 2
L
 
 8  9.109  10  31 kg  3.0  10 8 m s 1 
 0.60  10 9 m  0.60 nm .
5. We have the following boundary conditions to be satisfied at the

boundaries of the potential well: ( x  a )  ( x  a )  0 . Using
( x  a )  0 we get:
a
3a   3a   0
 cos
2  3 x


 dx (a)  A cos   B sin 
a
 2a   2a   2a 
1a  3x 
  1  cos dx  3   3 
2 a  a   A cos   B sin   B  0
 2   2 
a
1 a 3x 
 x sin
2  3 a   a So the wave function is:
a
 3x 
( x )  A cos 
 2a 
Using the normalization condition of Eq. (5.2) we can get:

a a
 3x   3x   3x 
  cos 2  2a  dx  1
2
A * cos  A cos  dx  1  A
 2a   2a 
a a
On evaluating the integral we get (see the margin remark):
1
A a   1  A 
2
a
Since V(x) is zero inside the well, the Schrödinger equation is:
 2 d 2 ( x )
  E ( x ) (i)
2m dx 2
 3x 
With ( x )  A cos  we get:
 2a 
2 d 2  3x  2 d  3 3x 
  A cos      A  sin 
2
2m dx   2a  2m dx   2a   2a 
2
 2   3  2 2
 3x   2  3    3x 
  A
   cos      A cos 
2m   2a   2a  2 m   
2a  2a 
42
9 2 2
 ( x )
8ma 2
On comparing with Eq. (i) we get the energy eigenvalue as:
9 2  2
E 
8ma 2
6. From Eq. (5.35) we know that En  n 2 E1. n = 2 corresponds to the first

excited state and E1 is the ground state. So,
 
E 2  10 eV  2 2 E1  E1  2.5 eV
 The second excited state is E3, with n = 3 in Eq. (5.35). So:
E3  3 3 E1  9  2.5 eV  22.5 eV
7. Using Eq.(4.26), we can write for an unnormalized wave function  x  :
  * pˆ x2 dx
pˆ x2    (i)

  *  dx

d d 2
` pˆ x  i   pˆ x2    2 (ii)
dx dx 2
And
d 2 d d
  Ae i kx  t   
d
ik Ae i kx  t    k 2 Ae i kx t 
dx 2 dx  dx  dx
 Ae i kx  t     2 (k 2 )Ae i kx  t  dx

*
 p x2   

 Ae i kx  t   Ae i kx  t  dx

*

 A * e i kx t  Ae i kx t  dx

  2 k 2    2k 2

 A * e i kx t  Ae i kx t dx


8. For  n x  to be an eigenfunction of pˆ x , we have to show (Eq. 4.35) :
pˆ x  n x    n x  (i)
43
2 nx
Using  n ( x )  sin in Eq. (i), we get :
L L
d d  2 nx 
pˆ x  n x   i   n x    i   sin 
dx dx  L L 
2  n nx  2 nx
 i   cos   in 3 cos (ii)
L  L L  L L
Since we cannot write pˆ x  n x  as  n x  ,  n x  is not an eigenfunction

of the momentum operator.
9. Since  n x  is the normalized wave function we use Eq. (5.36) and

2 nx
 n * x    n x   sin to calculate the expectation values.
L L
L
 2 nx   2 nx 
i) px  
 L 
sin pˆ x
L   L
sin
L 
dx
0
As calculated in Eq. (ii) of Terminal Question 8,
2 nx
pˆ x  n x   in cos
3 L
L
L L
2in nx nx in 2nx
 px  
L 2
sin
L
cos 
L
dx  
L2
sin
L
dx 
0 0
i n  L   2nx  L i
   cos  cos 2n  cos 0  0
L2  2n   L  0 2L
 px  0
L
ii) p x2 
  *n  x  pˆ x2  n  x  dx
0
 2 n nx 
L
2  nx  pˆ x2  n  x   pˆ x pˆ x  n  x   pˆ x  i . cos 
 sin   dx
 L   L L L 
0
1 L 2nx 
  1  cos dx (using the results of Terminal Question 8)
2 0  L 
2
1 L nx 
L in 2  d nx  2  n  2 nx
 x sin    i cos    i     sin
2  2n L  0 L L dx L   L  L L
L
 L
2 n 2 2 nx 
So p x2   2     sin 2  
 dx (see the margin remark)
 L  L  L 
0
44
2 2 n 2  2  L    2n 22
  
L3  2 L2
10. From Eq. (5.66b)

4k1k 2 4k1 / k 2
T  
k1  k2  k1 / k 2  12
2
k1 v1 10
For the given condition,   2.5. Hence, T 
k2 v 2 12.25
The transmitted current is T  1.1 mA  0.816  1.1 mA  0.898 mA
and the reflected current is 0.102 mA.
11. From Eq. (5.65),

2 2
 k  k2   E  E  V0 
R   1    
 k1  k 2   E  E V 
 0 
Thus, R  1 as E  V0 .
12. From Eq. (5.65), Using
E  E  V0
2 2
 k  k2   E  E  V0  E  E  V0
R   1    
 k1  k 2   E  E V  E  V0
 0  1
 E
2 E  V0
 7 2 1
or R   0.092. E
 7 2 
  And
Thus, 9.2% is reflected back. E  V0
x
13. Here, E < V0 . From Eq. (5.52),  2 ( x )  C e  x E
We can write :
1 x
2m(V0  E )  0.447  x  0.38
where  . 1 x
2 E  V0
  0.38
E
(7  5) eV E  V0
So,   7.25 nm 1   0.146
38  103 eV nm  2 E
 E  9.35
If the probability of finding the particle at x  0 is P, then the probability of
finding the particle at x  d is 0.05P. Thus,
2 2
PC or C e  2d  0.05P
Dividing the two relations, we get
e 2 d  20 or 2d  ln 20 and d  0.21 nm
2
 E  E  V0   E  E  V0 
14. R     0. 2  
 
  0.447

 E  E  V0   E  E  V0 
45
For (see the margin remark) V0  8 eV , E  9.35 eV  9 eV
46
Unit 6 Simple One-dimensional Potentials-II
UNIT 6
SIMPLE ONE-DIMENSIONAL
POTENTIALS-II
Structure
6.1 Introduction 6.4 One-Dimensional Finite Potential
Expected Learning Outcomes Well
6.2 One-Dimensional Barrier Potential Solving the Schrödinger Equation
Solving the Schrödinger Equation Parity of the Eigen Functions
Reflection and Transmission Energy Eigen Values
Coefficient for E  V0 Symmetric Infinite Potential Well
Reflection and Transmission 6.5 Summary
Coefficient for E  V0 6.6 Terminal Questions
6.3 Applications of Quantum Tunnelling 6.7 Solutions and Answers The idea of quantum
Alpha Decay tunnelling was used by
Scanning Tunnelling Microscope George Gamow in 1928 to
explain -decay and by
Cockcroft and Watson in
6.1 INTRODUCTION 1932, in their experiment
In Unit 5 you have solved the time independent Schrödinger equation for a to split an atom at
energies forbidden by
free particle, a particle confined to a certain region of space and a step
classical physics. Over the
potential. You have seen that for a free particle you can have a continuous years several Nobel Prizes
spectrum of energies, whereas when the particle is localized to a small region, have been awarded for
you have a discrete energy spectrum. In general, whenever you solve the time research based on the
independent Schrödinger equation for a potential V(x), you will have either (i) idea of quantum
“bound” states with discrete energy values, or (ii) “unbound” states with tunnelling: to Cockcroft
and Watson in 1954, to
continuous energy values. In this unit we first solve the time independent
Esaki in 1973 for
Schrödinger equation for a particle incident on a potential barrier and then for
discovering tunnelling in
a particle in a finite potential well. semiconductors, also in
Just like a potential well tends to attract and localize a particle, a barrier, as the same year to Giaever
the name suggests, has a tendency to deflect particles. When a classical and Josephson for
tunnelling in
particle is incident on a barrier, it will cross the barrier if its energy is greater
superconductors and to
than the barrier potential. It will be reflected at the barrier if its energy is less Binnig and Rohrer in 1986
than the barrier potential. What about a quantum particle? Will it be reflected for the scanning tunnelling
or transmitted through the barrier? No matter what the energy of a quantum microscope. Since
particle is, because the matter wave associated with it is finite everywhere – tunnelling across potential
there is no region of space that is forbidden to a quantum particle. In Unit 5, barriers is a purely
quantum mechanical
while solving the step-potential problem also you have seen that there is a
phenomenon, it is also
finite probability of finding a quantum particle in a classically forbidden region. seen as an important
The phenomenon in which a quantum particle is transmitted through a success of quantum
potential barrier and penetrates into regions which are classically forbidden to theory.
it is called “quantum tunnelling”. Quantum tunnelling was first noticed in 47
1927 by Friedrich Hund and since then it has been used to explain many
physical phenomena and has several important applications (read the margin
remark).
In Unit 5 you have seen how a particle confined to a one-dimensional box.
was modelled by a one-dimensional infinite potential well, in which the
probability of finding the box outside the well was exactly zero. A particle
rigidly confined by an infinite potential well is an idealization which does not
correspond to any real physical situation. Given sufficient energy, it is always
possible for a particle to escape the confines of a well. Hence, it is important
that we study the behaviour of a quantum particle in a finite potential well. The
important difference between the two systems is that in this case, even if the
kinetic energy of the particle in the well is less that the potential energy, there
is a finite probability of finding the quantum particle outside the well.
In Sec. 6.2 we solve the Schrödinger equation for the barrier potential and
calculate the reflection and transmission probabilities at the barrier. In
Sec. 6.3 we study some important applications of the concept of quantum
tunnelling in explaining -decay and in scanning tunnelling microscopy.
In Sec. 6.4 we solve the Schrödinger equation for the quantum particle in the
symmetric finite potential well. As you have studied, whenever a particle is
bound to a certain region of space, the energy spectrum is discrete and not
continuous. We learn how to calculate the energy eigenvalues and the
corresponding eigenfunctions. You will see that the eigenfunctions have a
definite parity which is a consequence of the symmetry of the potential
function. We also study the symmetric infinite potential well.
With this unit we complete our study of quantum mechanics. In the next block
you will study the fundamental concepts of nuclear physics.
 solve the time independent one-dimensional Schrödinger equation for a
quantum particle incident on a potential barrier;
 carry out calculations using the reflection and transmission coefficients;
 identify the factors on which the tunnelling probability depends;
 explain how tunnelling enables alpha decay;
 describe the working of a scanning tunnelling microscope;
 solve the time independent one-dimensional Schrödinger equation for the
finite potential well;
 determine the eigenfunctions and energy eigenvalues for a particle in a
one-dimensional finite potential well;
 determine the zero-point energy;
 determine the eigenfunctions and energy eigenvalues for a particle in a
one-dimensional symmetric infinite potential well; and
 state the parity of the stationary state wave functions.
48
6.2 ONE-DIMENSIONAL BARRIER POTENTIAL

Let us first look at the classical version of a potential barrier. Recall the
principle of conservation of energy that you study in undergraduate
mechanics. Consider a ball of mass 200 g (a cricket ball, for example) rolling
along a (frictionless surface) with a constant speed of 20 ms 1. So its kinetic
energy is 40 J. Now suppose it comes to a hump in the surface which has a
height of 15 m as shown in Fig. 6.1a below. If the ball were to reach the top of
the hump, its potential energy at the top of the hump would be ~30 J. So you
expect that the ball will easily reach the top of the hump and roll over it to
come down the other side. What if the height of the hump is 30 m (Fig. 6.1b)?
Then the potential energy of the ball at the top of the hump would be ~ 60 J. In
this case, the ball would not be able to cross over to the other side of the
hump. It would roll back at some point as is shown in the figure. Therefore, if
the potential energy is greater than the kinetic energy, the ball is not able to
cross the hump. You can imagine that the hump is the potential barrier which
prevents the ball from rolling over to the other side. This is the classical
picture.
V (x )
30 m
40 J 40 J V0
15 m
(a) (b)
I II III
Fig. 6.1: A ball having a kinetic energy of 40 J trying to cross a hump of height
a) 15.0 m; b) 30.0 m.
Let us now look at the quantum mechanical analogue of this. In quantum

mechanics the hump can be represented by a potential barrier. We define the
x
potential energy function for a one-dimensional potential barrier as follows: x =0 x=L
0 for x  0 Fig. 6.2: A potential

V ( x )  V0 for 0  x  L (6.1) barrier.

 0 for x  L
V(x) is shown in Fig. 6.2.

We divide the entire one-dimensional space into three regions as shown in the
figure. Region I extends from   to 0 ; region II from 0 to L and Region III from
L to  . The central region is known as the potential barrier. You can see that
the shape of the barrier is rectangular. In text books you may come across the
term “rectangular potential barrier” for this or “square potential barrier”
when the potential barrier has a square shape. They refer to the same
problem.
A classical particle incident on the barrier from the left cannot cross the barrier
if the energy (E) of the particle is less than the barrier potential V0 i.e. if
E  V0. If E  V0 , the particle continues its motion over the barrier with a 49
kinetic energy E  V0 and crosses it. So Region II and Region III are
inaccessible to a particle which has an energy E  V0 .
We now solve the time independent Schrödinger equation for a quantum

mechanical particle incident on the barrier with (i) an energy E  V0 and
(ii) E  V0 .

Let us consider the motion of a particle of mass m and total constant energy E
in the above mentioned one-dimensional space. Let us first write down the
time independent Schrödinger equations for the particle in regions I and III.
Since the potential energy in these two regions is zero, we have (putting
V(x) = 0 in Eq. 3.47):
Notice that we use either:  2 d 2 ( x )
  E ( x ) (for Regions I and III) (6.2)
( x )  Ae ikx  Be  ikx 2m dx 2
or Recall that you have already solved this equation in Sec. 5.2. We therefore
( x )  A sin kx  B cos kx write down the solutions (as in Eq. 5.10) for Regions I and III, which are  I (x )
as the general solution of and  III (x ) , respectively(see also the margin remark):
the Schrödinger equation
for V(x) = 0. Both solutions  I ( x )  Ae ikx  Be  ikx (for Region I) (6.3)
are equivalent. The choice and
of the form of the solution
is made for convenience.  III ( x )  Fe ikx  Ge  ikx (for Region III) (6.4)
where
 2k 2
E (6.5)
2m
and
2mE
k (6.6)
2
You have studied in Unit 5 that the function e ikx represents a particle moving
along the positive x-direction. And the function e  ikx represents a free particle
moving along the negative x-direction.
In the expression for  I (x ) , Ae ikx represents the plane wave that is incident
on the barrier at x  0 , which has an amplitude A and is travelling to the right.
Be  ikx represents the plane wave reflected at the barrier, which has an
amplitude B and is moving towards the left from x  0 . So in Eq. (6.4) both A
and B are non-zero. In Region III, since there is only the wave that is
transmitted through the barrier and moving towards the right, and no wave
travelling to the left, we can set G  0 . So, the wave function for Region III is:
 III ( x )  Fe ikx (for Region III) (6.7)
We next write the time independent Schrödinger equation for the particle in
Region II ( 0  x  L ), where the wave function is denoted by II (x ) :
 2 d 2 II ( x )
  V0II ( x )  EII ( x ) (for Region II) (6.8)
2m dx 2
50
Let us now define a parameter k  similar to k in Eq. (6.6):
2m(V0  E )
k  (6.9)
2
Note that unlike in
Then we rewrite Eq. (6.8) as:
Eq. (5.7), in Eq. (6.11) we
d 2 II ( x ) 2m retain both the terms Ce k x
 V0  E  II ( x )  0
dx 2 2 and De k x in the solution
or for region II. That is
d 2 II ( x ) because, in this case this
 k 2  II ( x )  0 (6.10) solution is valid only over
dx 2
the range 0  x  L and
From your knowledge of homogeneous second order ODEs you know that the both terms in Eq. (6.11)
solution of Eq. (6.10) is (see the margin remark): would remain finite in this
range.
 II ( x )  Ce k x  De  k x (for Region II) (6.11)
Now we consider the following two cases:

i) The energy of the particle: E  V0 .
In this case (V0  E ) is positive and the solution is given by Eq. (6.11) with
a real value of the parameter k  .
ii) The energy of the particle: E  V0 .
In this case (V0  E ) is negative and the parameter k  is complex. If we
write k   i , where  is real, then the solution is:
 II ( x )  Ce ix  De  ix (6.12)
where
2m(E  V0 )
 (6.13)
2
Let us now study the two cases separately.
CASE I: E  V0
To determine the constants A, B, C, D and F we use the properties of an
acceptable wave function for a physical system.
Let us first write down the boundary conditions. You know that for (x ) to be
d( x )
an acceptable wave function in a physical system, (x ) and must be
dx
continuous for all x. So, the wave functions in the three regions and their
derivatives must be equal at the boundaries of the potential barrier and satisfy
the following conditions:
 I ( x  0 )   II ( x  0) (6.14a)
 II ( x  L )   III ( x  L ) (6.14b)
And
d I ( x ) d II ( x )
 (6.14c)
dx ( x 0 ) dx ( x 0 )
51
d II ( x ) d III ( x )
 (6.14d)
dx ( x  L ) dx ( x  L )
We first apply the boundary condition given by Eq. (6.14a). From Eq. (6.3) we
can write  I ( x  0)  A  B . From Eq. (6.11) for  II (x ) we get
 II ( x  0)  C  D . Therefore, from Eq. (6.14a) we have:
AB C D (6.15a)
To apply Eq. (6.14b), we set x  L in Eqs. (6.11 and 6.7) for  II (x ) and
 III (x ) , respectively, and get:
 
Ce k L  De  k L  Fe ikL (6.15b)
To apply Eqs. (6.14c and d), we first write down the derivatives of  I (x ) ,
 II (x ) and  III (x ) .
d I
dx

d
dx
  
Ae ikx  Be ikx  ik Ae ikx  Be  ikx  (6.16a)
dII
dx

d
dx
  
Ce kx  De  kx  k  Ce k x  De  k x  (6.16b)
d III
dx

d
dx
  
Fe ikx  ik Fe ikx  (6.16c)
From Eqs. (6.14c and d) we can write:

ik A  B   k C  D  (6.17a)
and
 
k  Ce k L  De  k L  ikFe ikL (6.17b)
Now we have four equations (6.15a and b, and 6.17a and b), but there are five
unknowns: A, B, C, D and F. However, as you will see in the next section, we
are interested in calculating the probability of reflection and transmission, for
which we need the ratios B / A and F / A . If we divide the equations (6.15a
and b, and 6.17a and b) by A, we will be left with just four unknowns as you
can see below:
C D B
  1 (6.18a)
A A A
C k L D  k L F ikL
e  e  e (6.18b)
A A A
C D   B
k     ik  1   (6.18c)
A A  A
C  D   F
k  e k L  e  k L   ik e ikL (6.18d)
A A  A
The four unknowns now are B / A , C / A , D / A and F / A . We can determine

these constants by solving Eqs. (6.18a to 6.18d).
However, even without calculating the constants, we can see that the wave
function has an oscillatory behaviour in Regions I and III, but in Region II the
wave function is exponential. The typical wave function in all the three regions
for E  V0 , is shown in the Fig. 6.3.
52
V (x )
V0
I (x )
 II ( x )  III ( x )
I II III
x
x0 x L
Fig. 6.3: The wave function for the one-dimensional potential barrier for a
particle with an energy less than the barrier height.
In region III, the wave function is purely a travelling wave. We can calculate
the probability density of the wave in this region as:
PIII ( x )   III  ( x ) III ( x )  F 2 (which is a constant) (6.19)

In region I, the wave function is primarily a standing wave because we have a
wave travelling to the right as well as a wave travelling to the left. Note that the
amplitude of the reflected wave is necessarily less than that of the incident
wave. We can write the probability density probability density for Region I as:
PI ( x )   I  ( x ) I ( x )  A 2  B 2
 AB e 2ikx  ABe 2ikx (6.20)
So, the probability density in Region I has an oscillatory component (last two
terms of Eq. 6.20) as well as a constant component (first two terms of
Eq. 6.20). So the minimum probability density in Region I is always slightly
greater than zero as it is in Region III.
Within the barrier, although we have both the exponential terms, the
decreasing exponential term ( De k x ) is dominant. Hence, the probability
density decreases exponentially as PII ( x )  D 2 e  2k x . The probability
density for all three regions is shown in Fig.6.4 below:
P (x )
V0
I II III
x
x0 xL
Fig. 6.4: The probability density as a function of position, for a quantum
mechanical particle with an energy less than the barrier height.
53
We next calculate the reflection and transmission coefficients which are a
measure of the probability of the particle being reflected or transmitted at the
barrier.
6.2.2 Reflection and Transmission Coefficient for

E < V0
For the wave function  I (x ) ,  0 ( x )  Ae ikx is the part of the wave that is
incident on the barrier from the left, whereas  R ( x )  Be  ikx represents the part
x of the wave that is reflected back from the barrier as shown in Fig. 6.5.
Ae ikx
incident
wave
Fe ikx
transmitted
wave
Be ikx
reflected
wave x
x 0 x L
Fig. 6.5: Incident, reflected and transmitted waves at the one-dimensional

potential barrier.
As you have studied in Unit 5, the reflection coefficient is given by the ratio of
the magnitude of the probability current density associated with the reflected
wave to the magnitude of probability current density associated with the
incident wave (Eq. 5.59). So we write the reflection coefficient R as:
2
B
R (6.21)
2
A
You have studied that the intensity of a wave is proportional to its amplitude.
The reflection coefficient is, therefore, the fraction of the intensity of the
incident wave that is carried by the reflected wave. It also gives us the
probability that a particle having an energy E incident on a potential barrier V0
(such that E  V0 ) is reflected back by the barrier.
We next define the transmission coefficient which tell us the probability of a

particle being transmitted through the barrier. The transmission coefficient, T,
is the ratio of the magnitudes of the probability current densities associated
with the transmitted wave and the incident wave (Eq. 5.58), which is:
2
F
T  (6.22)
2
A
This is the ratio of the intensity of the transmitted wave to the intensity of the
incident wave.
54
Since the incident wave is either reflected or transmitted, we must also have
the sum of the reflection and transmission coefficient to be one:
R T  1 (6.23)
To calculate R and T for the system we have to calculate B / A and F / A
using Eqs. (6.18 a to d). The algebra is lengthy but straightforward and is
worked out for you in the Appendix to this unit which you should work
through. We write down the results for T and R:
1
 V0
2

T  1  sinh2 (k L ) (6.24a)
 4E (V0  E ) 
1
 4E (V0  E ) 
R  1   (6.24b)
2 2
 V0 sinh (k L ) 
where
e k L  e  k L
sinh( k L )  (6.24c)
2
and
2mL2 (V0  E ) 2mV0 L2  E
k L   1   (6.24d)
2 2  V0 
The transmission coefficient tells us about the extent of the penetration of the
barrier by the particle. It is also called the tunnelling probability.
Notice that Eq. (6.24a) tells us something that is totally at odds with classical
behaviour. For a classical particle, for E  V0 , the particle is always reflected at
the barrier. So R = 1 and T = 0. But, for a quantum mechanical particle there is
always a small but finite probability for the particle to penetrate the barrier
and appear on the other side. This phenomenon is called “barrier
penetration”. You can see that the transmission coefficient becomes
vanishingly small in the limit of large values of k L , because the factor e 2 k L in
Eq. (6.24a) would be very small.
For large values of k L , sinh( k L )  e k L / 2 and we can write the expression for
the transmission coefficient as:
1
 V0
2
 e 2k L 
T    (6.25)
 
 4E (V0  E )  4  
or
16E  E
T 1  e 2k L for k L  1 (6.26)
V0  V0 
Eq. (6.26) holds for wide barriers and large values of V0.
The transmission coefficient T decreases exponentially as e 2k L and T  1 .
The factor k L in the exponent (refer to Eq. 6.24d) is very large because
Planck’s constant is a very small number.
We define a tunnelling length:
1 
T   (6.27)
k 2m(V0  E ) 55
T is a measure of the opacity of the barrier. It is also known as the barrier
penetration depth. At a distance T into the barrier, the wave function has
fallen to 1/e of its value at the barrier edge; thus, the probability of finding the
particle is appreciable only within about T of the barrier edge. Notice that T
decreases exponentially with L, the barrier width (beyond the tunnelling
length) and the energy difference (V0  E ) .
In terms of the tunnelling length, Eq. (6.26) can be written as:
2L
16E  E  L
T  1  e T for  1 (6.28)
V0  V0  T
Let us estimate how the transmission coefficient changes with the barrier width
L for a fixed value of the tunnelling length T .
Example 6.1
Consider an electron bound inside a typical metal. Typically the effective value
of (V0  E ) that prevents the electron from escaping the metal is ~ 5.0 eV.
Calculate the tunnelling length and the ratio of the transmission coefficients for
L= 0.3 nm and 0.2 nm.
Solution : We calculate the tunnelling length using Eq. (6.27) with

V0  E  5.0 eV  8.0  10 19 J :
 1.054  10 34 Js
T  
2mV0  E  2  9.109  10 31 kg  8.0  10 19 J
 .09  10 9 m  .09 nm
Notice that the transmission coefficient of Eq. (6.28) can be written as

2L

T  f E / V0  e 2k ' L  f E / V0  e T (i)
With T  .09 nm , the transmission coefficient for L  0.3 nm is
2 3.0 nm
T1  f E / V0  e 0.09 nm  f E / V0  e 6.7
For L  0.2 nm the transmission coefficient is:

20 .2 nm

T 2  f E / V 0  e 0.09 nm  f E / V0  e 4.4
So the ratio of the transmission coefficients is:

T 2 f E / V0  e 4.4
  e 2.3  10
T1 f E / V0  e 6.7
So the transmission through a barrier of width 0.2 nm is almost ten times

more probable than a barrier of width 0.3 nm.
Let us work out another example on calculating the transmission coefficient.

56
Example 6.2
Calculate the transmission coefficient for an electron of energy 1.5 eV incident
on a potential barrier of 2.0 eV, if the width of the barrier is 0.50 nm.
Solution : We calculate first calculate T using Eq. (6.27) with
E  1.5 eV, V0  2.0 eV and then calculate the value of T from Eq. (6.26), with
1
using L = 0.50 nm and k   . So,
T
V0  E  0.5 0 eV  0.80  10 19 J .
T 
1.054  10 34 Js
2  9.109  10 31 kg 0.80  10 19 J
 0.28  10 9 m  0.28 nm
2L
E  E   T
 T  16 1  e
V0  V0 
2  0.50 nm
 1.5 eV  1.5 eV   0.28 nm
 16   1  e
 2.0 eVV  2.0 eV 
 3e  3 . 6
You may like to work through the following SAQ.
SAQ 1
Calculate the transmission coefficient for an electron in a semiconductor

(having an energy of 1.5 eV incident on a potential barrier 2.0 eV if the width
of the barrier is 0.10 nm). The effective mass of the electron is 0.22 m e .
You should not take the word ‘tunnelling’ literally. There is, of course, a finite
probability for the particle to be inside the classically forbidden barrier region
where its kinetic energy is negative. But the point is that nobody can "see" a
particle actually go through a classically forbidden region.
Particle detectors can detect only objects of kinetic energy greater than zero.
Suppose you are able to tunnel through a barrier to insert a detector inside it
to ‘see’ the particle. Then, you are not only making a hole in the potential but
also in your objective. Why so? Because the object will no Ionger belong to a
classically forbidden region, where you wanted to find it! Another way to say
this is that our effort to observe the object with any measuring instrument will
give it an uncontrollable amount of energy. This is how the uncertainty
principle works in such measurement situations!
Quantum tunnelling should be taken into consideration only in those systems

where wave particle duality is significant. In the Sec 6.3 we will take up certain
important applications of quantum tunnelling.
We now study the reflection and transmission coefficient for Case II in which
the energy of the particle E  V0. 57
6.2.3 Reflection and Transmission Coefficient for
E > V0
Let us write down the wave function for the three regions for E  V0
(Eqs. 6.4, 6.12 and 6.5) once again:
 I ( x )  Ae ikx  Be  ikx (for Region I) (6.3)
 II ( x )  Ce ix  De  ix (for Region II) (6.12)
 III ( x )  Fe ikx (for Region III) (6.5)
2m(E  V0 ) 2mE
where   and k  .
2 2
You can see that the wave function is oscillatory in all three regions. Applying
the boundary conditions Eqs. (6.14a to d) at x  0 and x  L for the wave
function and its derivatives as stated in Sec. 6.2.2, we can derive the
expressions for the transmission and reflection coefficients as:
1
 V0 2 
T  1  sin2 (L ) (6.29a)
 4E (E  V0 ) 
1
 4E (E  V0 ) 
R  1   (6.29b)
2 2
 V0 sin (L ) 
From the expressions you can see that both R and T have an oscillatory
component. Notice here that even when E  V0 , there is a finite probability for
the particle to be reflected at the barrier.
Once again this behaviour is NOT what is predicted by classical mechanics. A
classical particle which has an energy E  V0 would be transmitted and not
reflected (T = 1 and R = 0).
The typical variation of the transmission coefficient with the ratio of the particle
energy E to the value of the potential at the barrier which is V0 , is plotted in
Fig. 6.6. For E / V 0  1 , T is defined by Eq. (6.24a) and for E / V 0  1 , T is
given by Eq. (6.29a). For E / V 0  1 , you can that while there is a finite
probability of tunnelling , R  T .
T
E /V0
Fig. 6.6: The typical variation of the reflection coefficient R and transmission
2mV0 L2
coefficient T with E /V0 , for  16 .
58 2
Notice that the value of T is close to 1 for E  V0 and actually equal to 1 at
some points. These are the points at which T = 1 in Eq. (6.29a) and so we
must have:
sin 2 (L )  0  L  n for n  1,2,3... (6.30)
These points of perfect transmission, or T = 1 are called “transmission

resonances” and the energies at which they occur can be calculated from
Eq. (6.30). At these energies, a quantum particle will cross the potential barrier
without any reflection.
Let us now look at some interesting applications of quantum tunnelling which
also point towards the success of quantum mechanics.
6.3 APPLICATIONS OF QUANTUM TUNNELLING

We discuss two important applications which are alpha decay and scanning
tunnelling microscopy.
6.3.1 Alpha Decay
Alpha decay is the process in which the isotopes of certain radioactive
elements like uranium, radium and bismuth, decay by emitting alpha ()
particles. Alpha particles are helium nuclei with two neutrons and two protons.
After emitting the alpha particle, the original nucleus (parent nucleus) is
transformed into a different atomic nucleus (the daughter nucleus), with the
mass number reduced by four and the atomic number reduced by two. The
following two aspects of the alpha decay process were not explained for a
long time:
 All alpha particles emitted from the same source have almost the same
kinetic energy. If they are emitted from different sources, the kinetic
energies all lie within a narrow range of 4.0 to 9.0 MeV.
 The half-life of the radioactive element from which the alpha particle is
emitted, however varies over a very large range: for example the half-life
for polonium-214 is 160 s and the half life of uranium-238 is around
4.5 billion years. Incidentally the kinetic energies of the emitted alpha
particles are ~7.7 MeV for polonium and 4.3 MeV for uranium.
This large variation in the half-lives of the parent element in the alpha decay
process was explained by George Gamow using the concept of quantum
tunnelling. He assumed that before the alpha decay takes place, an alpha
particle exists inside the parent nucleus and is bound by the attractive
potential of the strong nuclear force. You may consider the nucleus to be a
kind of rigid spherical box inside which the alpha particle is confined. The
alpha particle is free to move between the walls of the box. It does have a
finite kinetic energy but this kinetic energy is much less that what required to
escape from the nucleus, leaving behind the daughter nucleus. So classically,
the alpha particle should remain forever remain bound inside the parent
nucleus.
We consider a somewhat simplified picture as shown in Fig. 6.7, in which the

potential function is plotted as a function of the distance from the centre of the
59
nucleus. In this the nuclear potential which binds the alpha particle is
represented by a square well. The nuclear force itself is extremely short-
ranged (~ 10 15 m) and hence it is not significant outside the nucleus.
Typically the radius of the nucleus is about 1 fm (~ 10 14 m). Suppose the
alpha particle now tunnels out of the nucleus, leaving behind the daughter
nucleus. Once the alpha particle escapes the nucleus by tunnelling through
the nuclear potential barrier, the only force acting on it is the Coulomb
repulsive force due to the (daughter) nucleus. Therefore outside the radius of
the nucleus ( r  R 0 ), the potential is modified by the Coulomb repulsion V(r)
between the alpha particle (which has a charge 2e) and the daughter nucleus
(which has a charge Ze). r is the distance between alpha particle and the
nucleus. V(r) is given by:
Ze (2e ) Ze 2
V (r )   (6.31)
4 0 r 2 0 r
The shaded region shows the forbidden region for the alpha – particle.
V (r )
Ze 2
V (r ) 
20r
E
r
R
R0
Fig. 6.7: Potential barrier for alpha-particle decay.
The kinetic energy of the alpha particle is E. At the point R, at which the alpha
particle escapes the nucleus, the kinetic energy of the alpha particle is at least
equal to the electrostatic energy between it and the daughter nucleus. So,
at r  R ,
Ze 2
E (6.32)
2 0 R
It is possible to calculate the tunnelling probability for the alpha particle,

which is just the transmission coefficient T which you have studied in
Sec. 6.2.2. Instead of the potential barrier of constant height which you have
studied earlier in this unit, here the potential barrier is described by the
function V (r ) defined in Eq. (6.31). The transmission coefficient (of
Eq. 6.26) now looks like:
R
2m
2  V ( r )E dr
2
T  e 2k L  e R0 (6.33)
This tunnelling probability, T, is the probability of emission of an alpha particle
from a nucleus. T is used to calculate the decay rate  for the nucleus, which
60 is the alpha particle emission probability per unit time. To calculate , we
multiply T by the number of times the alpha particle approaches the barrier
v
per second. This number is just , where v is the speed of the alpha
2R 0
particle inside the nucleus. v can be calculated from the kinetic energy E of
v
the alpha particle, when it escapes from the nucleus. Typically  10 21
2R0
collisions per second and the decay rate   10 21T .
The decay rate is used to calculate the half-life of the parent nucleus using the
relation T1 2  .693 /  . From Eq. (6.33), you can see that for a very small
change in E, there will be a disproportionately large change in  and hence in
T1 2 . So the more energetic α-particles have a better chance to escape the
nucleus, and, for such nuclei, the nuclear disintegration half-life will be shorter.
We now describe in brief another important device based on the tunnelling
phenomena, which is extensively used in materials science research today.
6.3.2 Scanning Tunnelling Microscope
Electrons tunnelling
between the sample
surface and the tip
sample tip
surface
Fig. 6.8: Schematic Diagram of a Scanning Tunnelling Microscope.
The Scanning Tunnelling Microscope (STM) is a type of microscope which is

used for imaging surfaces at the atomic level. It is based on the phenomenon
of field emission, in which electrons bound inside a metal are removed from its
surface by a very strong electric field. This happens by the quantum tunnelling
of electrons through the potential barrier.
The STM consists of
 a very sharp, conducting (typically tungsten, gold, or platinum—iridium) tip
(probe), which scans the surface to be imaged;
 a piezoelectric device which can control the height of the tip above the
surface to be scanned (typically 0.4 to 0.7 nm); and
 a mechanism to move the tip over the surface being studied.
There is of course a complex instrumentation that converts the inputs into a
computer imagery of the surface. Here we shall discuss only the basic
principle of the microscope.
When the tip is brought very close to the surface and a voltage difference
exists between the tip and the surface, electrons tunnel through the gap
between the tip and the surface and set up a tunnelling current. The tunnelling 61
current (which is proportional to the tunnelling probability of the electron
through the barrier) will depend on the distance between the tip and the
surface (this is the width of the barrier). So as we scan the tip over the sample
at a fixed height, the distance and hence the tunnelling current will depend on
the corrugations on the surface of the material. The variations in current are
converted into an image of the topography of the surface of the material. The
sensitivity of the microscope is such that the resolution is of the order of
0.001nm, which is even less than the typical diameter of an atom.
6.4 ONE-DIMENSIONAL FINITE POTENTIAL

WELL
Let us first define the potential function V (x ) for the one-dimensional finite
potential well:
V0 for x  a

V (x)   0 for  a  x  a (6.34)
V0 for x  a
Note that the well defined by Eq. (6.34) has the following features:
 The potential energy V0 is finite for x  a as well as for x  a .
 The potential energy is zero for  a  x  a .
 The potential function V (x ) is symmetric, i.e. V ( x )  V (  x )
This potential is sometimes referred to as the square potential well in text

books. V(x) is shown in Fig. 6.9.
V (x )
V  V0
V0
I II III
V=0
x  a x
0 x a
Fig. 6.9: A one-dimensional finite potential well. The potential energy is V0 for
x  a (Region I) and for x  a (Region III). It is zero for
 a  x  a (Region II).
If the total energy (E) of a classical particle is greater than the barrier height
V0 , it can move freely in the entire region, including the region x  a and
x  a . Its kinetic energy would be E  V0 in the region x  a and x  a and E
in the region  a  x  a . However, if E is less than V0 , the classical particle
would remain bound forever in the region  a  x  a . It would move back and
forth in the region with a constant speed (and a momentum of constant
magnitude) and the total energy of the particle could have any value E  0 .
62 Quantum mechanically, however, a bound particle can only have certain
discrete values of the energy. Further, as you have also seen in Unit 5, there
is a finite probability for the quantum mechanical particle to penetrate the
classically forbidden region i.e., x  a as well as x  a even if its
energy E  V0 .
The square shape of the potential is an over-simplification. However we study

this simpler shape because it is relatively easier to solve and it does capture
the physics for a microscopic particle bound in a region by certain forces. An
example of such a system would be a conduction electron bound inside a
metal block or a charge carrier trapped in a thin layer of semiconductor
sandwiched between two layers of a different semiconductor with larger band
gap than its own.
We now solve the time independent Schrödinger equation for a quantum
mechanical particle in this potential well.
Let us consider the motion of a quantum particle of mass m and total constant
energy E in the one-dimensional potential well described by Eq. (6.34). We
solve the Schrödinger equation in the three regions shown in Fig. 6.9. Region I
extends from   to a ; Region II from  a to a and Region III from a to  .
The central region is the potential well.
Let us first write down the time independent Schrödinger equation for the
particle in region II. Since the potential energy in this region is zero, we have
[with V(x) = 0 in Eq. 3.47]:
 2 d 2 ( x )
  E ( x ) (for Region II) (6.35)
2m dx 2
Inside the well we have a free particle. You already know the solution of the
Schrödinger equation for a free particle (Eq. 5.23). It is:
 II ( x )  A sin kx  B cos kx (for Region II:  a  x  a ) (6.36)
The total energy is

 2k 2
E  (6.37)
2m
and:
2mE
k (6.38)
2
We next write the time independent Schrödinger equation for the particle in
Regions I and III ( x  a and x  a , respectively) which is:
 2 d 2 ( x )
  V0  ( x )  E  ( x ) (6.39)
2m dx 2
You have learnt how to solve such an equation in Unit 5 and Sec. 6.2. We
define a parameter k  :
2m(V0  E )
k  (6.40)
2
63
We can rewrite Eq. (6.39) as:
d 2 ( x ) 2m
 V0  E ( x )  0
dx 2 2
d 2 ( x )
or  k 2 ( x )  0 (6.41)
dx 2
Let us write the solution for this equation for Regions I and III separately
(recall Eq. 6.10 and 6.11) where the wave functions are  I (x ) and  IIl (x ) ,
respectively:
 I ( x )  Ce k x  De k x (for Region I : x  a) (6.42)
And
 III ( x )  Fe k x  Ge k x (for Region III : x  a) (6.43)
Let us consider that the total energy of the particle E  V0 . Therefore, V0  E 
is positive and the parameter k  is real. We know that an acceptable wave
function must be finite for all values of x. Since the term De  k x in Eq. (6.41)
diverges as x   and the term Ge k x in Eq. (6.42) diverges as x   , for
the wave function be finite in all regions we must have :
D G 0 (6.44)
So, we can write:
 I ( x )  Ce k x (for Region I) (6.45)
and
 III ( x )  Fe  k x (for Region III) (6.46)
Let us summarize what we have studied so far.
Next we evaluate the constants A, B, C and F using the boundary conditions

on the wave function and its derivatives. Notice that the wave functions are
different in the three regions of the potential. In order that the wave function
and its derivatives are continuous in the entire region, we must match the
wave functions and their derivatives at the boundaries of the potential well.
Before we do that, let us examine a special property of a symmetric potential

function. This will help us simplify the solution for Schrödinger equation in
Region II.
Example 6.3
Show that, for a symmetric potential function V ( x )  V (  x ) , the parity
operator commutes with the Hamiltonian.
Solution : The commutator of P̂ and Ĥ can be written as follows:
Pˆ , Hˆ   PˆHˆ  Hˆ Pˆ (i)
To determine the value of this commutator, we operate it on the wave

function (x ) :
64
PˆHˆ  Hˆ Pˆ ( x )  Pˆ Hˆ ( x )  Hˆ Pˆ( x ) (ii)
Now
 2 d 2 ( x )
Hˆ ( x )    V ( x )( x ) (iii)
2m dx 2
In Unit 4 you have studied that the parity operator is the space inversion operator
( x   x ) and Pˆ ( x )  (  x ) . So the first term on the RHS of
Eq. (ii) is:
 2 d 2 (  x )
 
Pˆ Hˆ ( x )  
2m dx 2
 V (  x )(  x )
 2 d 2 (  x )
  V ( x )(  x ) [since V ( x )  V (  x )]
2m dx 2
 2 d 2 
   V ( x ) (  x )  Hˆ (  x ) (iv)
 2m dx 2 
For the second term on the RHS of Eq. (ii) we get

 
Hˆ Pˆ ( x )  Hˆ (  x ) (v)
So, using Eqs. (iv) and (v) in Eq. (ii) we get:

[Pˆ , Hˆ ]  0
Let us now apply this property to the potential function in Eq. (6.34).
6.4.2 Parity of the Eigenfunctions
Recall that the finite well potential function defined by defined by Eq. (6.34) is
symmetric, that is V ( x )  V (  x ) . So, as you have learnt by solving
Example 6.3, the Hamiltonian Ĥ of the symmetric potential well commutes
with the parity operator P̂ .
Now recall another result from Example 4.9 of Unit 4: If an operator Â
commutes with the parity operator P̂ , then the non-degenerate eigenfunctions
of Â have a definite parity, i.e., they are either of odd parity or of even parity.
So the solutions of the Schrödinger equation for the symmetric finite potential
well, which are the eigenfunctions of Ĥ , will have a definite parity. That is: the
eigenfunctions will either be even or odd.
Now this helps us to simplify the wave function for Region II given by
 II ( x )  A sin kx  B cos kx .
Notice that this wave function does not have a definite parity because sin kx is
an odd function [ sin k (  x )   sin kx ] and cos kx is an even function
[ cos k (  x )   cos kx ]. We have already established that the wave function
must have a definite parity. Therefore, either A or B should be zero.
If we take A = 0 we get even parity solutions:
 II ( x )  B cos kx (even parity :  (  x )   ( x ) ) (6.47)
With B = 0 we get the odd parity solutions:

65
 II ( x )  A sin kx (odd parity :  (  x )    ( x ) ) (6.48)
Both solutions are valid. Now let us apply the boundary conditions. The wave
functions must match at the boundaries of the potential well and so the
boundary conditions are:
 I ( x  a )   II ( x  a ) (6.49a)
and
 II ( x  a )   III ( x  a ) (6.49b)
Let us first work out the relation between the constants for the even parity
solution:  II ( x )  B cos kx . Substituting x  a in Eqs. (6.45 and 6.47), we
get from Eq. (6.49a):
Ce k a  B cos k ( a )  B cos ka (6.50)
Substituting x  a in Eqs. (6.46 and 6.47), we get from Eq. (6.49b):

Fe  k a  B cos ka (6.51)
The RHS of Eqs. (6.51 and 6.50) are the same, so we can write:
Ce k a  Fe k a  C  F e k a  0 (6.52)
Since e  k a  0 , the following must be true for the even parity solution:
C F (6.53)
For the odd parity solution,  II ( x )  A sin kx using Eqs. (6.49a and b)
respectively we get:
C  F (6.54)
You can verify Eq. (6.54) for yourself by solving SAQ 2.
SAQ 2
Show that for the odd parity solution  II ( x )  A sin kx , C  F .
Now we can write the complete set of solutions for the finite potential well.
The even parity solutions are (using C  F ):
 I ( x )  Ce k x (for x  a ) (6.55a)
 II ( x )  B cos kx (for  a  x  a ) (6.55b)
 III ( x )  Ce  k x (for x  a ) (6.55c)
The odd parity solutions are (using C  F ) :
 I ( x )  Ce k x (for x  a ) (6.56a)
 II ( x )  B sin kx (for  a  x  a ) (6.56b)
 III ( x )  Ce  k x (for x  a ) (6.56c)
The purpose of solving the Schrödinger equation is to obtain the

eigenfunctions of the Hamiltonian and the corresponding energy eigenvalues.
66 We have written down the even and odd parity eigenfunctions for the bound
state of the particle ( E  V0 ). However we have yet to determine the constants
B and C and the energy eigenvalues which are values of E corresponding to
these bound state eigenfunctions. Let us now determine E.
6.4.3 Energy Eigenvalues

We determine the constants B and C using the boundary conditions for the
continuity of the derivatives of the wave function.
d( x )
You know that for (x ) to be an acceptable wave function, (x ) and
dx
should be continuous for all x. Therefore the derivative of the logarithm of
(x ) which is
d
ln ( x )  1 d( x ) , should also be continuous. We now
dx ( x ) dx
write the condition of continuity for the derivatives of ln ( x ) at the boundary of
Regions I and II:
1 d I ( x ) 1 dII ( x )
 (6.57)
I ( x  a ) dx x  a II ( x  a ) dx x  a
Since
d I
dx

d
dx
 
Cek x  Ck e k x (6.58)
and for the even parity solution  II ( x )  B cos kx :
dII d
 Bk cos kx   B sin kx (6.59)
dx dx
On imposing the condition given by Eq. (6.57) on the even parity solution, we
get for x  a :
1 1
Ck e k a   Bk sin ka   k   k tan ka (6.60)
Ce k a B cos ka
Multiplying both sides of Eq. (6.59) by a we get:

ak   ak tan ka     tan  (6.61)
We have introduced the variables  and , which are defined as follows:
2m(V0  E )a 2 2mEa 2
  k a  and   ka  (6.62)
2 2
Notice that  and  are dimensionless quantities, but both  and  are
functions of the energy of the particle E.
From Eq. (6.61) we get:
2m(V0  E )a 2 2mEa 2
 2   2  k a 2  ka 2  
2 2
2mV0a 2
  R2 (6.63)
2
67
where R is a constant that depends on the depth of the well ( V0 ) and its width
2a:
2mV0 a 2
R (6.64)
2
The energy eigenvalues can now be obtained by solving Eqs. (6.61) and
(6.63) for  and  for a fixed value of R. Combining these equations we can
write:
 2   2  R 2 ;    tan  ; ,   0 for even parity states (6.65a)
Before we solve these equations you can derive the conditions equivalent to
Eq. (6.65a) for the odd parity states.
SAQ 3
Show that for the odd parity solutions given by Eqs. (6.56 a,b,c) the boundary
conditions on the continuity of the derivatives of the wave function lead to the
following conditions:
 2   2  R 2 ;    cot ; ,   0 for odd parity states (6.65b)
Eqs. (6.65a and 6.65b) are Eq. (6.65 a and b) are solved graphically as shown in Fig. 6.10. For this, we
what are known as plot  along the x-axis and  along the y-axis. Since both  and  are greater
transcendental than zero, we plot only the first quadrant. The equation  2   2  R 2
equations, where the
represents a circle with a radius R and is plotted in Fig. 6.10 for four different
variable appears both as
an argument to a values of R. For each value of R (corresponding to the value of the product
transcendental function V0 a 2 ), we get a different circle.
and elsewhere in the
equation. These equations
 A
   tan  B
are not solvable in closed     cot 
form. So we have to solve C
them numerically or R  6 .2
graphically
R  3 / 2 D
  R 2  2
R 
R  /2
Fig. 6.10: Graphical Solution of Eqs. 6.65a and b. The red, blue and black
curves represent the functions    tan  ,     cot  and
 2   2  R 2 respectively.
The equation    tan  and    cot  are plotted by the red and blue lines
respectively in Fig. 6.10. The solutions of Eq. (6.65a) are the points at which
the plot of the function  2   2  R 2 and    tan  intersect. In Fig. 6.10, for
R  6.2 , these points of intersection are A and C. At any point of intersection
68
we get a value of  , say    n . This value of  n can be used to calculate the
 2 n 2
energy eigenvalue E n using the relation En  (from Eq. 6.62).
2ma 2
Remember that this is for a particular value of R.
The solutions of Eq. (6.65b) are the points at which the plot of the function
 2   2  R 2 and    cot  intersect, which are the points B and D in
Fig. 6.10 for R  6.2 . Once again, at the point of intersection, we can calculate
the energy eigenvalue using Eq. (6.62).
Note from Fig. 6.10 that the allowed values of  and hence E are discrete and
the number of allowed values of E increases as R increases. The solutions of
Eq. (6.65a) give us the energy eigenvalues for the even parity states and the
solutions of Eq. (6.65b) give us the energy eigenvalues for the odd parity
states.
Therefore for E  V0 , the energy levels of a particle in a finite potential well are
discrete and depend upon the well parameters ( V0 ) and its width a. From
Fig. 6.10, note that:
 For a value of R lying between 0 and /2, there is just one possible
solution of the set of Eqs. (6.65a and b) and that corresponds to a state of
even parity (obtained from the intersection of the curves  2   2  R 2 and
   tan  ). We can write:
 2  22
R  R2   V0 a 2 
2 4 8m
 22
So for 0  V0 a 2  you can have just one bound state of even
8m
parity. There are no odd parity states for a value of R between 0 and /2.
 For a value of R in the range  / 2  R   , you have one solution of even

parity and one solution of odd parity, hence two bound states.
 For a value of R in the range  / 2  R  3  / 2 , you have two solutions of

even parity and one solution of odd parity, hence three bound states.
 The number of bound states depends on the value of R. The greater the
value of R (larger the value of V0 ), the greater the number of bound
states. As the value R increases, more bound states of even and odd
parities get added to the eigenfunctions.
 For the value of R = 6.2, you can see that there are four bound states. The
first intersection is at A, corresponds to a value of   1 and it gives us
the ground state energy for R  6.2 . It is a state of even parity. The next
intersection is at B, which corresponds to    2 and gives us the energy
of the first excited state, and so on.
 However small be the value of R, there will always be a solution

corresponding to an intersection of the curves  2   2  R 2 and
   tan  . This ensures there will always be at least one bound state
and that this state corresponds to an even parity state.
69
 For any value of value of R, the lowest energy state is always an even
parity state
Clearly, for E  V0 the particle is bound to the well and the energy spectrum is
discrete. What if the energy of the particle is greater than V 0 ?
In that case the particle is no longer confined to the well and the energy of the
particle varies continuously from V0 to .
In Fig. 6.11, you can see typical energy level diagram for a particle in a finite
potential well. For E  V0 , , you have discrete energy levels. For E  V0 , you
have a continuum of energies. Continuum
of Energies
V  V0 V  V0
E3
Discrete
E2 energy
levels
E1
x =  a x=a
Fig. 6.11: Typical energy eigenvalues for a finite potential well.
Once the energy eigenvalues are known, the corresponding eigenfunctions

can be determined by matching the wave function and its derivatives at one of
the boundaries x = a or x =  a . The calculation is however complex. But we
can deduce the nature of the eigenfunctions by looking at the corresponding
symmetric infinite square well (i.e. V0   ).
6.4.4 Symmetric Infinite Potential Well

The symmetric infinite potential well can be written as
 for x  a

V ( x )   0 for  a  x  a (6.66)

 for x  a
where we have replaced V0 by  in Eq. (6.1). Recall that we have solved the
infinite potential well in Unit 5. However, the potential well we studied in Unit 5
was not symmetric about x  0 .
Since there is an infinite potential for x  a and x  a , we can say that the
wave function will not extend beyond the boundaries of the well. Therefore, for
both the even and odd parity solutions:
 I ( x )   III ( x )  0
70 And for  a  x  a , we can write for the infinite symmetric potential well:
Even Parity Solution : ( x )  A cos kx (for  a  x  a ) (6.67a)
Odd Parity Solution : ( x )  B sin kx (for  a  x  a ) (6.67b)
The constants A and B are determined by the normalization condition for the
wave function. We can also write the following boundary condition for the
wave function going to zero at the boundaries of the potential well:
( x  a )  ( x  a )  0 (6.68)
Using this condition for Eq. (6.67a), we get:


( x  a )  A cos ka  0  ka  n  with n  1,3,5,... (6.69)
2
So the eigenfunction for the even parity state is:
Even Parity Eigenfunction :

nx 
 even ( x )  A cos  with n  1,3,5,.... (6.70a)
 2a 
The corresponding eigen energy for the even parity states is found using
 2k 2
E and the value of k as obtained in Eq. (6.69):
2m
 2  n  2
E n,even  with n  1,3,5,.... (6.70b)
2m  2a 
You can obtain the corresponding odd parity eigenfunctions and the eigen
energies by applying Eq. (6.68) to Eq. (6.67b). You may like to work this out
for yourself.
SAQ 4
Show that the odd parity eigenfunctions and eigen energies of the symmetric
infinite potential well are:
nx 
 odd ( x )  B sin  with n  2,4,6.... (6.71a)
 2a 
and
 2  n  2
E n,odd  with n  2,4,6.... (6.71b)
2m  2a 
You can see for yourself that the lowest possible energy is for n  1 in
Eq. (6.70b) which is:
 22
E1  (6.72a)
8ma 2
x
This corresponds to the even parity eigen function  1( x )  A cos  . This is
 2a 
the ground state and this energy is the zero point energy. Notice that the
minimum energy is not zero.
The next energy eigenvalue is found by substituting n  2 in Eq. (6.71b): 71

 2 2   22 
E2   4   4E1 (6.72b)
2ma 2  8ma 2 
x
This corresponds to the odd parity eigenfunction  2 ( x )  B sin  .
 a 
The next energy eigenvalue corresponds to a even parity state, so we

evaluate it from Eq. (6.70b) with n  3 and get:
9 2  2
E3   9E 1 (6.72c)
8ma 2
3x 
And the corresponding even parity eigenfunction is  3 ( x )  A cos .
 2a 
So as you can see, the eigenfunctions are alternately of even and odd parity.
Now that we know the form of the eigenfunctions for the infinite potential well,
we can deduce what the eigenfunctions of a finite potential well would look
like. Unlike in the infinite potential well, the wave function for a finite potential
well would not go to zero at the boundaries of the potential well. Rather the
wave function would decay exponentially on both sides of the potential well
and the form of the wave function is given by  I ( x ) and III ( x ) as described in
Eqs. (6.55a and 6.55c) or Eqs. (6.56a and 6.56c) depending on the parity of
the solution.
In Fig. 6.12(a) and (b) we plot the first three eigenfunctions for the infinite and
finite symmetric potential well respectively.
V  V0 V (x ) V  V0
V  V (x ) V 
3(x) 3(x)
2(x) 2(x)
1(x) 1(x)
x x
x =  a 0 x=a x =  a 0 x=a
(a) (b)
Fig. 6.12: The first three eigenfunctions for (a) a symmetric infinite potential
well, and (b) a symmetric finite potential well
Note that the wave function extends into the classically forbidden region on
either side of the boundary of the finite potential well.
We define what is known as the penetration depth, which is the distance to
which the wave function penetrates the region beyond the boundaries of the
potential as:
1 
  (6.73)
k 2m(V0  E )
72
At a distance  beyond the boundary of the well the amplitude of the wave
function falls to 1/e of its value at the boundary. At this distance, the wave
function in the exterior of the well is almost zero.
Also the eigen energies for a finite well are typically less than that for an
infinite square well of the same width. This is because when the wave
functions extend upto a distance of  beyond the boundaries of the well, we
can say that the effective size of the box increases. So the typical energies are
lower.
6.5 SUMMARY
 The potential energy function for a one-dimensional potential barrier of
width L is:
0 for x  0

V ( x )  V0 for 0  x  L
 0 for x  L
The time independent Schrödinger equation for a particle of mass m is:

 2 d 2( x )
  E ( x ) (for x  0 and x  L )
2m dx 2
 2 d 2  II ( x )
  V0  II ( x )  E II ( x ) (for 0  x  L ) (6.8)
2m dx 2
 2 d 2  II ( x )
  V0  II ( x )  E II ( x ) (for 0  x  L )
2m dx 2
The solutions of the Schrödinger equations are:
 I ( x )  Ae ikx  Be  ikx (for x  0 )
 II ( x )  Ce k x  De  k x (for 0  x  L )
 III ( x )  Fe ikx (for x  L )
where
2mE 2m(V0  E )
k and k 
2 2
 The transmission coefficient T is the probability of a particle being
transmitted through the barrier and is given by:
1
 V0 2 
T  1  sinh2 (k L )
 4E (V0  E ) 
where
2mL2 (V0  E ) 2mV0L2  E

k L   1  
2 2  V0 
The reflection coefficient R is the probability of the particle being reflected

at the barrier edge and is given by:
73
1
 4E (V0  E ) 
R  1  
 V02 sinh2 ( k L ) 
The sum of the reflection and transmission coefficients is 1.

For wide barriers and large values of V0 the transmission coefficient is:
1 2L
 V0
2  e 2k L  16E  E 
T     1  e T
   V 
 4E (V0  E )  4  V0  0
where the tunnelling length T is:
1 
T  
k 2m(V0  E )
The probability of finding the particle is appreciable only within about T
 The wave function for the three regions for E>V0 are:
 I ( x )  Ae ikx  Be  ikx (for x  0 )
 II ( x )  Ce ix  De  ix (for 0  x  L )

 III ( x )  Fe ikx (for x  L )
2m(E  V0 ) 2mE
where   and k  .
2 2
The reflection and transmission coefficients are:

1
 V0 2 
T  1  sin 2 (L )
 4E (E  V0 ) 
1
 4E (E  V0 ) 
R  1  
 V0 2 sin 2 (L ) 
The points of perfect transmission at which T = 1 and R = 0 are called

“transmission resonances and are given by the condition: L  n for
n  1,2,3...
 Alpha Decay
The process of emission of alpha particles from a radioactive nucleus
takes place by the quantum tunnelling. The alpha particle is initially
trapped in a potential well by the nucleus. Classically, it cannot escape
from the nucleus. However quantum mechanics allows for a finite
probability of tunnelling of the alpha particle through the potential
barrier created by the nuclear potential. The lifetime of the radioactive
nucleus is calculated from tunnelling probability.
Scanning Tunnelling Microscope
The Scanning Tunnelling Microscope (STM) is a type of microscope
which is used for imaging surfaces at the atomic level and works on the
principle of quantum tunnelling. When the tip of the STM is very close
to the surface, the voltage difference between the tip and the surface
74 causes electrons to tunnel through the gap and set up a tunnelling
current which depends on the distance between the tip and the
surface. This is used to create an image of the surface.
 The potential function for the one-dimension finite potential well is
V0 for x  a Region I

V ( x )   0 for  a  x  a Region II

V0 for x  a Region III
The time independent Schrödinger equation for the particle of mass m in
the three regions is:
 2 d 2  II ( x )
  V0 ( x )  E( x ) (for Region I and III)
2m dx 2
 2 d 2 ( x )
  E( x ) (for Region II)
2m dx 2
The solutions of the Schrödinger equation in the three regions are:
 I ( x )  Ce k x (for Region I)
 II ( x )  A sin kx  B cos kx (for Region II:  a  x  a )
 III ( x )  Fe k x (for Region III)
With:
2m(V0  E ) 2mE
k  and k 
2 2
 The potential function for the one-dimension finite potential well is
symmetric, hence the Hamiltonian commutes with the parity operator.
The eigenfunctions of the Hamiltonian have a definite parity. The even
parity solutions for E  V0 are
 I ( x )  Ce k x (for x  a )
 II ( x )  B cos kx (for  a  x  a )
 III ( x )  Ce  k x (for x  a )
The odd parity solutions for E  V0 are:
 I ( x )  Ce k x (for x  a )
 II ( x )  B sin kx (for  a  x  a )
 III ( x )  Ce  k x (for x  a )
 The energy eigenvalues can be obtained by solving the transcendental
equations:
 2   2  R 2 ;    tan  ; ,   0 for even parity states
2   2  R 2 ;     cot ; ,   0 for odd parity states
for  and  for a fixed value of R, where
2m(V0  E )a 2 2mEa 2 2mV0 a 2
  k a  ;   ka  ;R 
2 2 2
For E  V0 the particle is bound to the well and the energy spectrum is
discrete.
The lowest energy state is always an even parity state and there is
always at least one bound state. 75
The number of bound states depends on the radius of the circle, R.
The greater the value of R (larger the value of V0 ), the greater the
number of bound states.
 The potential function for the symmetric one-dimensional infinite potential
well is
 for x  a Region I

V ( x )   0 for  a  x  a Region II

 for x  a Region III
The solutions of the Schrödinger equation for a particle of mass m in the
symmetric infinite well have a definite parity:
nx 
 even ( x )  A cos  with n  1,3,5,....
 2a 
nx 
 odd ( x )  B sin  with n  2,4,6....
 2a 
The eigen energies are:
 2  n  2
E n,even  with n  1,3,5....
2m  2a 
 2  n  2
E n,odd  with n  2,4,6,.... \
2m  2a 
 22
The ground state energy is: E1 
8ma 2

1. An electron with a kinetic energy 4.0 eV is incident on a potential barrier of
height 10.0 V and width 0.80 nm. Calculate the tunnelling length and
tunnelling probability of the electron to tunnel through the barrier (use
Eq. 6.28).
2. For the electron of TQ1, calculate in which case the tunnelling probability
increases more:
i) when the width of the barrier is reduced to 0.4 nm, other parameters
remaining the same.
ii) when energy of the electron is increases to 8 eV, other parameters

remaining the same.
3. An electron and a proton with the same kinetic energy E are incident on a
potential barrier of height V0 and width L. Calculate the ratio of their
tunnelling probabilities.
4. An electron with a energy of 8.0 eV strikes a potential barrier of energy

10.0 eV. If the tunnelling probability is 2.0 percent, determine the width of
the barrier (use Eq. 6.28).
5. An electron has a kinetic energy of 8.0 eV. The electron is incident upon a
rectangular barrier of height 15.0 eV which has a thickness of 1.0 nm.
Calculate the increase in the tunnelling probability of the electron if it
76 absorbs all the energy of a photon of blue light (3.1eV).
6. Calculate the probability that an electron will tunnel through a 0.4nm gap
from a metal to the STM probe if the work function is 3.0eV. By what factor
does the tunnelling probability change if we increase the gap to 0.50 nm.
7. Explain whether the eigenfunctions of the following Hamiltonian for a

particle of mass m will have a definite parity:
p2 1
H  m2 x 2
2m 2
8. Calculate the normalization constants for the wave functions of
Eqs. 6.34a and b.
i) ( x )  A cos kx (for  a  x  a )
ii) ( x )  B sin kx (for  a  x  a )
9. Calculate the penetration depth for an electron of energy 40 eV, trapped

by an electrostatic potential of 100 eV.
10. The penetration depth for an electron is 2.0 nm. How much energy would
be required to “free” the electron from the well?

1. Following the steps of Example 6.2, we first calculate the tunnelling length
T using Eq. (6.27) with E  1.5 eV, V0  2.0 eV and m  0.22m e . So:
V0  E  0.50 eV  0.80  10 19 J
1.054  10 34 Js
T 
2  0.22  9.109  10 31 kg  0.80  10 19 J
 0.58 nm (i)
E 1.5 eV 3
Using Eq. (6.28) with   , L  0.10 nm , and T as calculated
V0 2.0 eV 4
in Eq. (i) we calculate the transmission coefficient T :
20.10 nm
3 3 
T  16      1   e 0.58 nm  3 e 0.34
4  4
2. We use the condition given in Eq. (6.49a), with  ll  x   A sin kx to get,
C e k a  A sin k  a    A sin ka  A sin ka  Cek a (i)
and applying the condition of Eq. (6.49a) we get
A sin ka  Fe k a (ii)
Equating the LHS of Eqs. (i) and (ii) we get:

 
 Ce k a  Fe k a  C  F  e k a  0
 e  k a  0 we must have C  F
3. Taking  II x   A sin kx we impose the condition 77

1 d I 1 d II
 x  a    x  a  (i)
 I  x  a  dx  II  x  a  dX
From Eq. (6.56) we know:

d I
 C k  e k x (ii)
dx
dII d
And  A sin kx   Ak cos kx (iii)
dx dx
So, using Eq. (i) and the results of Eqs. (ii) and (iii) we get:
1

C e k a
C k e    A sin1 ka Ak cos ka
 k a
 k   k cot ka (iv)
or k a   ka cot ka (v)
With   k a and   ka , we get

   cot 
And from Eq. (6.65a) 2  2  R 2
So, we can write the conditions for odd parity states as:
2  2  R 2;     cot , ,   0
4. Using  x   B sin kx and applying Eq. (6.68) we get,
 
B sin ka  0  sin ka  0 or ka  n , n  2,4,6,...
2
n
 k , n  2,4,6,...
2a
 nx 
So  n ( x )  B sin , n  2,4,6,...
 2a 
 2k 2  2  n  2
And En   , n  2,4,6,...
2m 2m  2a 
Terminal Questions
1. We first calculate the tunnelling length T using Eq. (6.27) with
E  4.0 eV,V0  10 .0 eV and m  m e :
1.054  10 34 Js
 T   0.08 nm (i)
2  9.109  10  31 kg  9.6  10 19 J
E 4.0 eV 2
With   , L  0.80 nm and T  0.8 nm in Eq. (6.28):
V0 10.0 eV 5
20.80 nm
2 2
T  16     1  e 0.08 nm
96  20
  e  3.84  e 20
5  5  25 
2. Let us denote the transmission coefficient calculated in TQ 1 by T 0 . So
78
T0  3.84 e 20 (i)
E 2
i) With L = 0.40 nm, T  0.08 nm and  in Eq. (6.28) :
V0 5
20.40 nm
 2 2
T1  16  1   e 0.08 nm  3.84  e 10 (ii)
 5  5
ii) We first calculate T with E  8.0 eV, V0  10 eV and m  m e
T 
1.054  10 34 Js  0 .14 nm
2  9.109  10 31 kg  3.2  10 19 J
E 8.0 eV 4
Then with E = 8 eV , L = 0.80nm,   and
V0 10 .0 eV 5
T  0.14 nm, we get from Eq. (6.28):
2 0.80 nm
4 4 
T2  16.  1   e 0.14 nm
 5  5 
64  11.4
  e  ( 2.56 )e 11.4 (iii)
 25 
We now calculate the ratios of the transmission coefficients T1 and T2
with respect to T0 :
T1 (3.84 )e 10
  e10
T0 (3.84 )e  20
T2 (2.56 ) e 11.4
and   (0.67 ) e 8.6
T0 (3.84 ) e  20
Since e 10  0.67e 8.6 we can say that the tunnelling probability increases
more in case (i) when width of the barrier is decreased.
3. Let us say that the tunnelling length for the electron is

 TE 
2me V0  E 
And the tunnelling length for the proton ( m P  1836 m e ) is

 TP 
2mP V0  E 
 
 
21836me V0  E  2me V0  E  1836

 Te (i)
43
The transmission coefficient for the electron is
2L
 E  E  
Te  16  1   e Te
 V0   V0 
and for the proton it is:
79
2L
 E  E   TP
T P  16  1  e
 V 0   V0 
2L 2L
   2L 
Te e Tee Te
42 
   e  Te   Te  TP
TP 2L 2L
 
e Tp e  Te 43 
4. We calculate the tunnelling length by using Eq. (6.27) with

V0  E  2 eV  3.2  10 19 J
So T 
1.054  10 34 Js  0.14 nm
2  9.109  10 31 kg  3.2  10 19 J
(i)
The tunnelling probability is calculated using Eq. (6.28) with
E 8 eV 4 1
  and T  0.14 nm . Given that T  2%  we can
V0 10 eV 5 50
write:
2L
1 4 4 
T   16   1  e 0.14 nm
50  5  5
2L 2L
 25
 e 0.14 nm  or e 0.14 nm  128
64  50
Taking the logarithm of both sides,
2L
 ln128   L  0.07  4.85  nm  0.34 nm
0.14 nm
5. We first calculate the tunnelling length using Eq. (6.27), with
V0  E  15.0  8.0  eV  11.2  10 19 J
1.054  10 34 Js
T 
2  9.109  10  31 kg  11.2  10 19 J
 0.074  10 9 m  0.074 nm (i)
The tunnelling probability T1 is calculated using Eq. (6.28) with L = 1.0 nm,
8.0 eV 1
E / V0  and T   0.074 nm
15.0 eV k
2 1.0 nm
8 8 
 T1  16    1   e 0.074 nm  ( 4.0)e 27 (ii)
 15   15 
When electron absorbs an photon of energy 3.1eV its kinetic energy
becomes
E  8.0  3.1 eV  11.1eV . So V0  E  3.9 eV  6.24  10 19 J
Now the tunnelling length is,
1.054  10 34 Js
T 
2  9.109  10 31 kg  6.24  10 19 J
 0.098 nm (iii)
80
11.1
With E / V0  and T given by Equation (iii) the tunnelling probability
15.0
is:
21.0 nm
11.1   11.1  .098 nm
T2  16    1  e  3.1 e 20.4
 15.0   15.0 
So the increase in the tunnelling probability is:
T2 3.1e 20.4  3.1  6.6

  e  570
T1 4.0  e 27.0  4.0 
So the tunnelling probability increases about 570 times.

6. Since the electron must overcome the work function, we can assume that
the value of V0  E is at least  3 eV  4.8  10 19 J
So the typical tunnelling length can be calculated using Eq. (6.27)
1.054  10 34 Js
T   0.11nm
2  9.109  10 31 kg 4.8  10 19 J
Given that E ,V0 are fixed we can write that the tunnelling probability is a
function only of the gap between the tip of the STM probe and the surface
i.e. T  f E / V0 e 2L / T .
For L  0.40 nm , the transmission coefficient is

2  0.40 nm 

T1  f E / V0  e 0.11nm   f E / V0  e  7.3
For L  0.50 nm the transmission coefficient is

20.50 nm

T2  f E / V0  e 0.11nm  f E / V0  e 9.1
T2 f E / V0  e 9.1 1
   e 1.8 
T1 f t / V0  e  7 . 3 6
The tunnelling probability reduces by 1/ 6 when the distance between the

tip and the surface increases by 0.10 nm. The tunnelling current also
changes proportionally. This is why the STM can detect variations even of
the order of 0.10 nm on the surface of a material.
7. Following the steps of Example 6.3, we can write,
Pˆ , Hˆ   Pˆ Hˆ .x   Hˆ Pˆx  (i)
where
d2 1
Hˆ   2  m2 x 2
dx 2 2
So     2 d 2 1
Pˆ Hˆ x   Pˆ 

 m2 x 2  x 
 dx 2 2 
81
d   x 
2
1
  2  m  2 x 2  x  (ii)
2 2
dx
and

Hˆ Pˆ  x   
  2 d 2 (  x ) 1
 m  2 x 2  x  (iii)
2m dx 2 2
From Eq. (ii) and (iii) we can write

   
Pˆ Hˆ   Hˆ Pˆ 
or Pˆ , Hˆ   0
 the parity operator commutes with the Hamiltonian, the eigenfunctions will
have a definite parity.
8. i) The normalization condition is
a a
n
  x   x  dx  1  A
* 2
a

cos2 kx dx  1 where k 
2a
(n  1,3,5...)
a
a
A2 A2  1 a
 1

2  1  cos 2 kxdx  1  2 
x 
2k
sin 2kx 
a
a
A2 2a  2 sin 2ka  1

  
2 2k
Given that
n n
k ( n  1,3,5...)  sin 2ka  sin 2  a  sin( n)  0
2a  2a 
A2 1
 2a  1 and A  A 
2 a
ii) The normalization condition is written as
a
n
B2  sin2 kx dx  1 where k  2a , n  2,4,6...
a
a
B2

2  1  cos 2 kx dx  1
a
B2  1 
a
or x  sin 2 kx 1
2  2k 
a
B2 2a  1 sin 2ka   1
or  
2 2k
2n
sin 2ka  sin a   sin n  0
2a
B2 1
 2a   1 or B  B 
2 a
9. We calculate penetration depth  using Eq. (6.73).
82
With V0  100 eV and E  40 eV
1.054  10 34 Js
So 
2  9.109  10  31 kg 60  1.6  10 19 J
 .025  10 9 m  .025 nm
10. The energy required to free the electron is V0  E  .
Using Eq. (6.73) for the penetration depth , we can write
V0  E 
 2 (i)
2me 2
with   2.0 nm  2.0  10 9 m
we get,
V0  E 
1.054  10  34 Js2  0.015  10 19 J
2  9.109  10  31 kg 2.0  10  9 m 
2
 0.01eV
83
APPENDIX 6A: CALCULATING THE REFLECTION AND
TRANSMISSION COEFFICIENTS
After applying the boundary conditions on the wave functions and their
derivatives for E  V0 , we have derived the following equations for A, B, C, D
and F:
AB C D (6.15a)
 
e k LC  e  k L D  e ikL F (6.15b)
ikA  ikB  k C  k D (6.17a)

 
k e k LC  k e  k LD  ike ikLF (6.17b)
Let us first calculate the value of B/A from these equations. For this we
eliminate the constants C and D from the Eqs. (6.15a,6.15b, 6.17a and 6.17b).
Multiplying Eq. (6.15a) by k  we get:
k A  k B  k C  k D (i)
Adding Eqs. (6.17a) and (i) we get:
( k   ik ) A  ( k   ik )B  2k C (ii)
Subtracting Eq. (6.17a) from Eq. (i) we get
( k   ik ) A  ( k   ik )B  2k D (iii)
Multiplying Eq. (6.15b) by ik we get:
ike k LC  ike  k LD  ike ikL F (iv)
Subtracting Eq. (iv) from Eq. (6.17b) we get
( k   ik )e k LC  ( k   ik )e  k LD  0 (v)
Which we can write as:
( k   ik )e k LC  ( k   ik )e  k LD

(k   ik )e k L
or C D (vi)
( k   ik )ek L
Eq. (vi) gives us the relation between C and D. Substituting for C from Eq. (vi)
into Eq. (ii) we get:

(k   ik )ek L
(k   ik )A  (k   ik )B  2k  D
(k   ik )ek L
which is:
(k   ik )ek L
(k   ik )A  (k   ik )B   2k D (vii)
(k   ik )e k L
84
You can see that the RHS of Eqs. (iii) and (vii) are equal. Equating the LHS of
these two equations we get:
(k   ik )e k L
(k   ik )A  (k   ik )B
(k   ik )e  k L
 ( k   ik ) A  ( k   ik )B (viii)
On simplifying, we can write Eq. (viii) as:
(k   ik )e k L (k   ik )A  (k   ik )B 
 (k   ik )e k L (k   ik )A  (k   ik )B  Notice that B/A given in
which is: Eq. (xii) is complex and it
has the general form:
(k   ik )(k   ik )e k L A  (k   ik ) 2 e k L B
B u
 ( k   ik )(k   ik )e k L A  ( k   ik )2 e  k L B 
A v  iw
Using ( k   ik )(k   ik )  k 2  k 2 we can write:

u  v  iw 
(k  2  k 2 )e k L A  (k   ik ) 2 e k L B  
v  iw  v  iw 
 (k 2  k 2 )e k L A  (k   ik )2 e k L B (ix) uv  iuw

Dividing Eq. (ix) by A we get v2 w2
B
(k  2  k 2 )e k L  (k   ik ) 2 e k L
A B
2
uv 2  uw 2

 (k 2  k 2 )e  k L  (k   ik )2 e  k L
B
(x)
A v 2  w 2 2
A
u 2 v 2  w 2 
Therefore we have 
B

( k  2  k 2 )e  k L  e k L 
(xi)
v 2  w 2 2
A ( k   ik ) 2 e k L  ( k   ik ) 2 e k L u2

which is: 
v2 w2 
B


(k  2 
 k 2 ) e  k L  e k L
(xii) With :
A ( k   k )e
2 2 k L e  k L   (2ik k )e k L  e  k L 
u  ( k  2  k 2 ) sinh(k L )
or
v  ( k 2  k  2 ) sinh( k L )
B ( k  2  k 2 ) sinh( k L )
 (xii) And
A ( k 2  k  2 ) sinh( k L )  2ik k cosh( k L )
w  2k k cosh(k L)
B2
The reflection coefficient is R  (Eq. 6.21). Using the value of B/A We get the result of
A2 Eq. (xiii).
obtained in Eq. (xii) we can write (see the margin remark):
( k  2  k 2 ) 2 sinh 2 ( k L )
R (xiii)
( k 2  k  2 ) 2 sinh 2 ( k L )  4k  2 k 2 cosh 2 ( k L )
Using the identity sinh2 (k L)  cosh2 (k L)  1 , we can simplify the denominator
of the RHS of Eq. (xiii) as
(k 2  k  2 ) 2 sinh 2 (k L )  4k  2 k 2 cosh 2 (k L )
 ( k 2  k  2 ) 2 sinh 2 ( k L )  4k  2 k 2 sinh 2 ( k L )
 4k 2k 2 cosh2 (k L )  4k 2k 2 sinh2 (k L )
 (k 2  k 2 )2 sinh2 (k L )  4k 2 k 2
Therefore
( k  2  k 2 ) 2 sinh 2 ( k L )
R
( k 2  k  2 ) 2 sinh 2 ( k L )  4k  2 k 2
1
 (k 2  k 2 )2 sinh 2 (k L )  4k 2k 2 
  (xiv)
 ( k 2  k 2 )2 sinh 2 (k L )  85
2mE
From Eqs. (6.6 and 6.9) we also know that k  and
2
2m(V0  E )
k  .
2
2mE 2m(V0  E )
So, k2  ; and k  2 
 2 2
and
2m(V0  E ) 2mE 2mV0
k 2  k 2    ;
2 2 2
2m(V0  E ) 2mE
k 2k 2   (xv)
2 2
Substituting from Eq. (xv) into Eq. (xiv) we get the result of Eq. (6.24b):
1 1
 4k 2k 2   4E (V0  E ) 
R  1    1  
2 2 2 2 2 2
 (k   k ) sinh (k L )   V0 sinh (k L ) 
To calculate T we use Eq. (6.23). Hence T = 1  R. Using the value of R from
Eq. (xiv) we get:
( k  2  k 2 ) 2 sinh 2 ( k L )
T  1 R  1
( k 2  k  2 ) 2 sinh 2 ( k L )  4k  2 k 2
4k 2 k 2

( k 2  k 2 )2 sinh 2 ( k L )  4k 2k 2
1
 (k 2  k  2 ) 2 sinh 2 (k L) 
 1   (xvi)
 4k  2 k 2 
Substituting from Eq. (xv) into Eq. (xvi) we get the following result of
Eq. (6.24a):
1
 V 2 sinh 2 (k L ) 
T  1  0 
 4E (V0  E ) 
86
Unit 7 Simple Harmonic Oscillator-I
UNIT 7
SIMPLE HARMONIC
OSCILLATOR-I
Structure
7.1 Introduction Comparison with the Classical
Expected Learning Outcomes Oscillator
7.2 Stationary State Schrödinger Examples of Simple Harmonic
Equation Oscillator Systems
Power Series Solution 7.3 Solution with Ladder Operators
Boundary Conditions and 7.7 Summary
Eigenfunctions 7.5 Terminal Questions
Energy Eigenvalues 7.6 Solutions and Answers
Parity of the Eigenfunctions
Solution of the Time-Dependent
Schrödinger Equation
7.1 INTRODUCTION
Continuing with our study of one-dimensional quantum mechanical systems,
we now study a quantum particle in a parabolic potential. The system is one
that you are all familiar with – the one-dimensional harmonic oscillator or the
simple harmonic oscillator. The system is of direct physical interest, because a
large number of systems are governed exactly or approximately by the
harmonic oscillator equation. 
F
The prototype of a classical one-dimensional harmonic oscillator is a mass m
connected to a spring, undergoing to and fro motion  (Fig. 7.1). The motion m
follows Hooke’s law according to which the force F on the particle is directly x
proportional to the displacement and is always directed towards the mean x 0
position, i.e., F  kx : k , the constant of proportionality, is known as the force Fig. 7.1: Mass on a
constant or the spring constant. It is related to the classical frequency of the spring as a one-
dimensional
oscillator and is given by k  42 2 m where m is the mass of the particle.
harmonic oscillator.
1
The potential energy of the particle at x is given by V ( x )  kx 2 or
2
1
V ( x )  m2 x 2 , where   2 is the angular frequency. If you plot the
2
potential V (x ) as a function of x, you will get a parabola, centred at x  0 .
The potential energy function for any arbitrary physical system may not be
perfectly parabolic. Yet, in the neighbourhood of a local minimum, almost any
87
potential function can be approximated by a parabola. Consider the
potential function V (x ) of Fig. 7.2.
V (x )
x
xa
Fig. 7.2: A Potential Function V(x).
Let us expand V (x ) in a Taylor series about the local minimum x  a :

dV 1 d 2V
V ( x )  V ( x  a)  ( x  a)  ( x  a )2  ...
dx x a 2 dx 2 x  a
(7.1)
dV
Since x  a is a minimum,  0 and we can write, after dropping the
dx x a
terms with powers of ( x  a ) higher than 2, and ignoring the constant term
V ( x  a)
1 d 2V
V (x)  ( x  a )2 (7.2)
2 dx 2 x a
So any arbitrary potential V (x ) can be written as a parabolic potential
1 d 2V
V (x)  k ( x  a )2 where k  . This explains why the simple harmonic
2 dx 2 x a
oscillator potential is so important.
In Sec. 7.2 we solve the stationary state Schrödinger equation for the simple
harmonic oscillator. We work out a power series solution and determine the
energy eigenfunctions and energy eigenvalues after applying the boundary
conditions for a physically acceptable wave function. The energy of a quantum
mechanical simple harmonic oscillator is found to be quantized and the
solutions for the eigenfunction can be written in terms of Hermite polynomials.
You will study some examples of simple harmonic oscillator systems like a
diatomic molecule, etc. You will compare the quantum oscillator with the
corresponding classical system and see how they are different. In Sec. 7.3
you will study another method of arriving at the simple harmonic oscillator
energy eigenfunctions and eigenvalues, which is mathematically simpler than
solving the Schrödinger equation as you do in Sec. 7.2. This method will be
used again in Unit 12 of this course.
This is the last Unit in which you solve the one-dimensional Schrödinger
equation. In the next two Units you will learn how to solve the three-
dimensional Schrödinger equation, in particular for a spherically symmetric
potential. In this unit and the next two we use the power series method for
solving ODEs. You may like to revise this from your undergraduate course
material.
88

 explain the importance of the simple harmonic oscillator problem in
physics;
 write down and solve the Schrödinger equation for a one-dimensional
simple harmonic oscillator potential;
 obtain the normalized energy eigenfunctions and energy eigenvalues for
a one-dimensional simple harmonic oscillator ;
 determine the solution of the time-dependent Schrödinger equation for the
simple harmonic oscillator; and
 obtain the energy eigenfunctions and eigenvalues of the simple harmonic
oscillator using the ladder operators.
7.2 STATIONARY STATE SCHRÖDINGER

EQUATION
The time independent Schrödinger equation for the one-dimensional simple
harmonic oscillator is:
 2 d 2 1 2 2
 2m dx 2  2 m x  ( x )  E( x ) (7.3)
 
where E is the total energy of the oscillator , which is independent of time. It is
evident from the above equation that the Hamiltonian of the system is invariant
under space inversion, i.e., it commutes with the parity operator. Hence the
eigenfunctions have a definite parity (TQ, Unit 4).
We now we define a new variable  , such that   ax, a being a constant.

With a 2  m /  and E  , Eq. (7.3) reduces to (SAQ 1)
2
d 2 (  )
 (    2 ) ( )  0 (7.4)
d 2
Notice that both  and  are dimensionless variables(check for yourself).
SAQ 1
 
Show that for a 2  m /  and E  , Eq. (7.3) reduces to
2
d 2 (  )
 (    2 ) ( )  0
d 2
We now determine the solution of Eq. (7.4) to get the eigenfunctions and
eigenvalues of the one-dimensional harmonic oscillator.
Before we obtain the solution using the power series method, we examine Eq.
(7.4) in the limits of large and small values of  to obtain a trial solution.
In the limit    , we can drop the term  in comparison to 2 in Eq. (7.4)
89
d 2()
to get the equation :  2()  0 . You can check for yourself that
d 2
2
( )  C m e   / 2 is a solution of this equation. Since a physically acceptable
wavefunction should not diverge, we can drop the term with the positive
exponent term and retain only the negative exponent term to write the solution as
2
()  Cme   / 2 .
In the limit   0 , on dropping the term 2 in comparison to  , Eq. (7.4)

d 2 (  )
reduces to :  ( )  0 which has the solution
d 2
( )  A cos    B sin  . If we now expand the sine and cosine terms as a
series in  and drop all terms of the order of  2 and higher, we get a solution of
the form ( )  A  B  .
To accommodate both limits we can assume a solution of the form

1
 2
()  h() e 2 where:
 h() approaches as  m (plus lower powers of  ) as    , and

 h() approaches A  B  (plus higher powers of  ) as   0 .
The next step is to determine h() explicitly.
7.2.1 Power Series Solution

We choose a trial solution (an ansatz) of Eq. (7.4) of the form
1
 2
(  )  h( ) e 2 (7.5)
So we get:
1
d  dh   2
  h e 2 (7.6a)
d  d 
And
1
d 2  d 2h dh   2
  2  (2  1)h  e 2 (7.6b)
d 2  d 2 d 
Substituting from Eq. (7.6b) into (7.6a) we get a differential equation for
h  h( ) :
d 2h dh
 2  (  1)h  0 (7.7)
d 2 d
We solve this equation using the power series method by assuming a solution
for h() of the form:
90

h()   cn  n (7.8a)
n  0,1,2,...
We can write down the first and second order derivatives of h() as:
dh

d n  0,1,2..

cn n  n 1 (7.8b)
d 2h

d2 n  0,1,2...

c n (n )(n  1) n  2 (7.8c)
Substituting from Eqs. (7.8a,b,c) into Eq. (7.7) we get

 
 c n (n ) ( n  1) n  2  2  cn (n ) n  (  1)  cn n  0
n  0,1,2,... n  0,1,2,.. n  0,1,2...
(7.9)
We can also write Eq. (7.9) as:
 
 cn  2 (n  2) (n  1) n  2  cn (n ) n  (  1)
n  0,1,2,... n  0,1,2,..
 cn n  0 (7.10)
n  0,1,2...
or

 cn  2 (n  2) (n  1)  2(n )c n  (  1)cn  n  0 (7.11)
n  0,1,2,..
The relation between the coefficients c n  2 and c n is

2n    1
cn2  cn (7.12)
(n  2) (n  1)
From Eq. (7.12) we get
1  3
c2  c0 ; c3  c1,
2! 3!
(1   )(5   ) (3   ) (7   )
c4  c0 ; c5  c1;.. and so on.
4! 5!
So each coefficient in the series solution can be written in terms of either c 0 or
c1 . Writing h( ) as:
h( )  c0  c 2 2  c 4 4  ...  c1  c33  ... (7.13)
and substituting for the coefficients we see that h( ) is the sum of two series:
1   2 (1   )(5   ) 4
h( )  c0 1      ...
 2! 4! 
3   3 ( 3   ) (7   ) 5
 c1       ...
 3! 5! 
 c0S1( )  c1S2 ( ) (7.14)
91
1   2 (1   )(5   ) 4
where S1()  1      ... and
2! 4!
3   3 (3   ) (7   ) 5
S2 (  )        .. and c0 and c1 are arbitrary
3! 5!
constants. Eq. (7.14) is the most general solution of Eq. (7.7), S1() and
S2 () are the two independent solutions.
The wave function for the system is:

1 1
 2  2
( )  h( ) e 2  c0S1( )  c1S2 () e 2 (7.15)
7.2.2 Boundary Condition and Eigenfunctions

We know that a physically acceptable wave function must satisfy the condition
()  0 for    and h() should approach a finite power of  ,  m (plus
lower powers of  ) as    . However from Eq. (7.13) or Eq. (7.14) we can
see that h() can have arbitrarily large powers of  . Now the behaviour of the
series solution in Eq. (7.13) is controlled by the coefficient of the lowest power
of  , that is c0 for   0 and by the coefficient c n  for    .
We now calculate the ratio of the consecutive terms in the series solution
given in Eq. (7.13) , for large values of n:
c 2n 2
limn   n  2   (7.16)
cn n2 n
Let us now look at the power series expansion of a different function
2
f ()  e :
2 2  ... 
 2   an 2 
n n
e  1  2 
2
n! 
2! n  0,1,2 n  0,1,2
n
  an n where an  1  2 ! (7.17)
n  0,2,4..
The ratio of the consecutive terms of this series for large n is :

 n  1!
a  
2  2
limn   n 1  limn    0 (7.18)
an  n ! n
 
2
So you can see that the ratio of the coefficients of consecutive terms in the
series solution for h() (Eq. 71.13 or 7.14) and in the series e  are the same
2
2
and equal to for large n. Therefore we can say that the S1() and S 2 ( )
n
2
behave as e . Therefore the series solution for ( ) is
1 1
 2  2
( )  h() e 2  c0S1( )  c1S2 ( ) e 2
1  1 2 
 
 e e 2
2  2
  e2 
 
(7.19)
92  
Clearly this solution does not satisfy the boundary condition that the wave
function goes to zero at    . Clearly an infinite series solution for h( ) of
the kind in Eq. (7.14) will not give us any allowed value for the energy E. The
1
 2
only way to ensure h( ) e  0 as    would be if h() were a
2
polynomial and not an infinite series in  . For h( ) to be a polynomial we
must have c n  2  0 for some value of n so that all consecutive coefficients
which depend on it vanish.In that case the highest power of  in h( ) is n
 1
1
 2 En   n   
and ( )  n e 2  0 as    . For that Eq. (7.12):  2
2n    1  0    2n  1, n  0,1,2,... (7.20)
For the values of   1, 5, 9,... , (even values of n) you can see that S1() is a
polynomial whereas S 2 ( ) is an infinite series. So for all even values of n if
you set c1 to zero you will get a polynomial solution. For the values of 7
n  3, E 3  
2
  3, 7,11,... , (odd values of n) S 2 ( ) is a polynomial whereas S1() is an
5
infinite series. So for all odd values of n if you set c 0 to zero you will get a n  2, E 3  
2
polynomial solution. 3
n  1, E 3  
2
So, for a physically wave function, the values of  are discrete and the 1
n  0, E0  
energy of the quantum simple harmonic oscillator is quantized: 2
1
  2n  1  2 n  ; n  0,1,2,...
Fig. 7.3: Energy level
 2 diagram of a simple
harmonic oscillator.
   1
 E  E n   n    ; n  0,1,2,... (7.21a)
2  2
The corresponding eigenfunctions are

1
 n ( )  c 0  c 2  2  c 4  4  ..  c 4  n e 2
 2
for n  0,2,4...
(7.21b)
1
 c1  c 3  3  c 5  5  ..  c n  n e 2
 2
for n  1,3,5...
(7.21c)
SAQ 2
Tabulate  n () and E n for n  0,1,.2,3,4 .
For   2n  1, Eq. (7.7) is the Hermite equation:
d 2h dh
 2  2nh  0 (7.22)
d 2 d
The solutions of this equation are familiar to you, these are the Hermite
polynomials H n () with n  0,1,2,3... :
A few of the lower order H n () are

H 0 ( )  1 (7.23a)
93
H1( )  2 (7.23b)
H 2 ( )  4 2  2  21  2 2  (7.23c)
2
H 3 ()  8 3  12  12    3  (7.23d)
 3 
 4 
H 4 ( )  16 4  48 2  12  121  4 2   4  (7.23e)
 3 
Note that:
1 1
 2  2
 0 ( )  c 0 e 2  H 0 ()e 2 with c 0  1 (7.24a)
1 1
 2  2
1( )  c1e 2  H1()e 2 with c1  2 (7.24b)
1 1
 2 ()  c 0 1  2 2 e 2
 2  2
 H 2 ()e 2 with c 0  2 (7.24c)
1 1
2  2  2
 3 ()  c1    3  e 2  H 3 ()e 2 with c1  12 (7.24d)
 3 
1 1
4  2  2
 4 ()  c 0 1  4 2   4 e 2  H 4 ()e 2 with c 0  12
 3 
(7.24e)
We write the general solution for the eigenfunctions of the simple harmonic
oscillator problem as:
 n ( x )  Nn Hn () exp( 2 / 2), n  0,1, 2,... (7.25a)
Where Nn is the normalization constant. These eigenfunctions form an

orthonormal set

*
  n ( x )n ( x )dx  nn (7.25b)

The following relations are also known for the Hermite polynomials:
  2 n!

2 n
 H n ()H n () exp(  )d   nn (7.26a)

H n  ()  2nH n 1() (7.26b)
And
H n 1()  2H n ()  2nH n 1() (7.27)
With   ax, we can write
 n ( x )  N n H n (ax ) exp( a 2 x 2 / 2), n  0,1, 2,... ` (7.28)
94
N n is the normalization constant to be determined by the normalization

condition on the eigenfunctions:   *n ( x ) n ( x )dx  1. Using Eq. (7.26a) we

can determine N n :
1
 a 2
Nn   (7.29)
 2 n!  
n
The normalized eigen function is therefore

1/ 2
 a 
n (x)    H n (ax ) exp( a 2 x 2 / 2), n  0,1, 2,...
  2 n n! 
(7.30a)
And
 1
Hˆ  n ( x )  E n  n ( x ); where E n   n    ; n  0,1,2,...
 2
(7.30b)
Notice also that all the stationary state eigenfunctions are real and so:
 *n ( x )   n ( x ) (7.30c)
SAQ 3
a) Check the orthogonality of the wave functions:
i)  0 ( x ) and 1( x ) ii) 1( x ) and  2 ( x ) .
b) Calculate the normalization constants N 0 and N1 for the eigenfunctions.

i)  0 ( x )  N 0 H 0 () exp(  2 / 2) ii) 1( x )  N1 H1() exp(  2 / 2)
1/ 2
 a 
With  n ( x )    H n (ax ) exp( a 2 x 2 / 2), n  0,1, 2,... we can write
  2 n n! 
down the first four eigenfunctions which are plotted in Fig. 7.4 below.
Table 7.1: Simple Harmonic Oscillator Wavefunctions
n Eigenfunction Energy Eigenvalue
 a 
1/ 2 
0 0 (x)    exp( a 2 x 2 / 2), 2
 
 a 
1/ 2 3 
1  1( x )    (2ax ) exp( a 2 x 2 / 2) 2
2 
 a 
1/ 2 5 
2 2 (x)    2a 2 x 2  1exp( a 2 x 2 / 2) 2
8  
 a 
1/ 2 7 
3
3 (x)    8a 3 x 3  12ax exp(a 2 x 2 / 2) 2
 48  
95
 a 
1/ 2 9 
 4 4 3
 4 ( x )     2a x  6a 2 x 2   exp( a 2 x 2 2
4 6    2
1( x )
0 ( x )
ax
ax
(a)
(b)
2(x) 3 ( x )
ax ax
(c) (d)
Fig 7.4: Eigenfunctions a)  0 ( x ) ; b)  1( x ) ; c)  2 ( x ) ; and d)  3 ( x ).
7.2.3 Energy Eigenvalues

The energy eigenvalues of a simple harmonic oscillator are quantized, and the
values are:
  3  5  1
, , ,...,  n    , ..
2 2 2  2
The integer n is known as energy quantum number. Since our simple
harmonic oscillator is one-dimensional system we have only one quantum
number.
The energy levels of quantum mechanical oscillator are equally spaced and
the difference between consecutive energy eigenvalues is   . This is a
characteristic of some parts of the molecular and nuclear experimental
spectra. The harmonic oscillator provides a good model description of these
spectra, so much so that the spectra are referred to as vibrational spectra.
There are also excitations in solids called phonons that fall in the same
category.
For each eigenvalue, there will be only one eigenfunction, there is no
degeneracy. This property seems to be a common characteristic of bound
sates for one-dimensional potentials that remain finite for all finite values of x.
96 We have the zero point energy given by

E0  for n = 0 (7.31)
2
This zero point energy is again a consequence of the Heisenberg uncertainty
p2 1 2
principle. This can be seen as follows. Since E   kx , E can be zero
2m 2
only when p and x both are equal to zero simultaneously. Under such a
circumstance p and x will become definite (equal to zero) simultaneously. This
will violate uncertainty principle. Therefore, the lowest eigenenergy has to be
non-zero.
7.2.4 Parity of the Eigenfunctions
As you know from TQ 7 of Unit 6, the eigenfunctions of a symmetric potential
function have a definite parity. The eigenfunctions of simple harmonic
oscillator are also eigenfunctions of the parity operator. It is evident form Eq.
(7.28) that the eigenfunctions corresponding to zero or even values of n are of
even parity (  n (  x )   n ( x ) ) . On the other hand odd parity eigenfunctions
 n (  x )   n ( x ) correspond to odd values of n. From Fig. 7.4 you can see
that the wave functions  0 ( x ) and  2 ( x ) have even parity and the
wavefunctions 1( x ) and  3 ( x ) have odd parity.
7.2.5 Solution of the Time-Dependent Schrödinger

Equation
The general solution for the time dependent Schrödinger equation of the
simple harmonic oscillator can be written in terms of its stationary state
eigenfunctions:
iEnt

 ( x, t )   a n  n ( x, t )  an  n ( x )e 
n  0,1,2... n  0,1,2...
1
 i ( n  )t
  an  n ( x )e 2 (7.32)
n  0,1,2...
From Eq. (7.32) we know that we can write the initial state ( t  0 ) of the
system (x,0) as:
( x,0)   an  n ( x ) (7.33)
n 0,1,2...
The initial state is just a linear superposition of the stationary state
eigenfunctions of the quantum oscillator . To find the coefficients an we do the
following:
 
 n* ( x ) ( x,0) dx   ak   *n ( x )  k ( x ) dx
 k  0,1,2... 
  ak  nk  an (7.34)
k  0,1,2...

Therefore : an    *n ( x ) ( x,0) dx .

The probability density of finding the the oscillator in the state  n (x ) at t  0
is a n 2 . The probability density of finding oscillator in the state  n (x ) at a
97
2
later time t is also an . If the initial state of the system is normalized, we
have

  * ( x,0) ( x,0) dx  1   an  1
2
(7.35)
 n  0,1,2...
   iEn t   iE t
 k 
Then   * ( x, t ) ( x, t ) dx   

 an*  *n ( x )e     ak  k ( x )e  dx
 k  0,1,2... 
   n  0,1,2...
i En  En t 
   *
an a k e 
  *n ( x ) k ( x ) dx
n  0,1,2... k  0,1,2... 
i En  En t
   an* ak e   nk
n  0,1,2... k  0,1,2...
  an 2 (7.36)
n  0,1,2...
From Eq. (7.35) we know that  an 2  1 and therefore

n  0,1,2...

  * ( x, t ) ( x, t ) dx  1 (7.37)

So if the initial state of the system is normalized, the wave function is

normalized for all t.
SAQ 4
The initial wave function for a simple harmonic oscillator is
1 2
( x,0)  1( x )  i 2( x )
5 5
where 1,  2 ,..., are the normalized eigenfunctions of the simple harmonic
oscillator.

i) Show that   * ( x,0) ( x,0) dx  1

ii) Determine ( x, t )
iii) Determine the expectation value of Ĥ in the state (x,0) .
7.2.6 Comparison with the Classical Oscillator

Let us now compare the quantum oscillator with the classical oscillator. We
first take up the question of time dependence. Classically, the simple harmonic
oscillator oscillates in such a manner that the position of the particle
represented by the oscillator changes from one moment to another in a
definite manner. Quantum mechanics, on the other hand, tells us that for any
state of energy E, although there is a distribution of probabilities for various
positions, this distribution does not change with time; (these probabilities are
‘frozen’ in time). This is the usual meaning of energy eigen states being
stationary. Is it possible to reconcile these two very different pictures?
98
The answer lies in considering not one single eigenfunction but a
superposition of eigenfunctions as in a wave packet. Consider for example,
the following superposition ( x, t ) of the first two oscillator eigenfunctions:
1
 ( x, t )  exp( iE 0 t /  )  0 ( x )  exp( iE1t /  )1( x ) (7.38a)
2
And  * ( x, t ) 
1
exp(iE0t / )  0* ( x )  exp(iE1t /  )1* ( x ) (7.38b)
2
So ( x, t ) 2   * ( x, t )( x, t )
1 *
2

 0  0  1* 1   0* 1 exp( it )  1*  0 exp(it )
 
1

  0 2  12  2 0 1 cos t
2
 (7.39)
1

Notice that ( x, t  0) 2   0 2  12  2 0 1
2

And ( x, t 
2 2 1


)   0 2  12  2 0 1
2

2 2
Therefore ( x, t  0) 2  ( x, t  ) . Clearly, the probability density

oscillates with time with just the frequency of the harmonic oscillator, as
2
expected classically. If we plot ( x, t ) , we get Fig. 7.5, where the plot is
made for four different values of time. It is therefore reasonable to except that
when we take a superposition of a large number of oscillator eigenfunctions
we will get a classical behaviours.
2 2
( x, t ) ( x, t )
ax ax
2 2
( x, t ) ( x, t )
ax ax
Fig. 7.5: The probability corresponding to the superposition of the first two
oscillator eigenfunctions of equal amplitude of Eq. (7.39) (with their
time dependence included) plotted at four different times a) t  0 ;
99
b) t   / 2 ; c) t   ; and d) t  3  / 2 .
You should be very clear, that the quantum solution of the harmonic oscillator
is radically different from that the classical oscillator. In classical mechanics,
the oscillator is forbidden to go beyond the potential, beyond the turning points
where its kinetic energy turns negative. But clearly, the quantum wave
functions extend beyond the potential, and thus there is a finite probability for
the oscillator to be found in a classically forbidden region.
SAQ 5
a) Determine the classical turning points of a simple harmonic oscillator with
1
an energy given by E n   n     .
 2
b) Show that the probability distribution of a particle in a harmonic oscillator
2
potential returns to its original shape after the classical period T  for

any general harmonic oscillator state.
As an example, let us compare the quantum and classical probabilities for the
states corresponding to n = 0 , n = 1 and n = 2. The quantum probabilities are
easily calculated by taking the square of the appropriate wave functions,  0 ,
1 and  2 .
The expression for the probability of finding a classical harmonic oscillator of
mass m and energy E, governed by the equation x  A sin t in a region x
is given as
1 1
P ( x )  x  x (7.40)
2A (1  x / A 2 )1 / 2
2
where A  (2E / m 2 )1 / 2 . As expected, the classical probability is non-zero

only for  A  x  A; the oscillator is confined within the turning points. For
1
x  A , it is clear that the potential energy m2 x 2  E, and classically this is
2
impossible.
P1( x )
P0 ( x )
2
P2 ( x )  0 (x )
2 2
 2 (x ) 1(x )
Fig. 7.6: Comparison of the quantum and classical probabilities for n = 0, n = 1

and n = 2.
We can determine the classical amplitude corresponding the nth eigen state by
100 equating the maximum energy of the classical oscillator to its energy
1 2 1
eigenvalue in that state so m 2 An  E n  n    and derive the
2  2
corresponding classical probability for the nth eigen state Pn (x ) by replacing
A by An in Eq. (7.40).
The quantum and classical probabilities are compared in Fig. 7.6 for n = 0,
n = 1 and n = 2, respectively. In both cases, the quantum probability does not
vanish in the classically forbidden region. For n = 1, the classical probability is
maximum at the turning points. But the quantum probability reaches the
maximum much closer to the point of equilibrium. For large n, the average of
the quantum mechanical probability distribution is given by the classical
probability curve.
Let us now consider an application of the simple harmonic oscillator to a
physical system: the vibrations of the two atoms of a diatomic molecule.
7.2.7 Examples of Simple Harmonic Oscillator Systems
Diatomic Molecule R1 R2
The simple harmonic oscillator serves as an important model for the
vibrational motion in a diatomic molecule. Let us assume that in a particular A c.m B
diatomic molecule, both the atoms in the molecule execute simple harmonic
Fig. 7.7: Vibration of
motion about their equilibrium positions. Hence, they satisfy the following two atoms A and B in a
equations diatomic molecule.
d 2R1
M1   k (R  R e ) (7.41a)
dt 2
and
d 2R2
M2  k (R  Re ) (7.41b)
dt 2
where R1 and R2 are the distance of the two atoms A and B from their centre
of mass and R  R1  R 2 (Fig. 7.7). The masses of the two atoms are M1 and
M 2 k is the force constant and at equilibrium, the distance between the two
atoms is Re . Considering the moment about A (see Fig. 7.7) we obtain
M 2R  (M1  M 2 )R 1 (7.42)
Putting Eq. (8.42) in Eq. (8.41a) we obtain
M1M2 d 2R
 k (R  Re ) (7.43)
M1  M2 dt 2
or
d 2x
  k x (7.44)
dt 2
M1M 2
with x  R  Re and   .
M1  M 2
Thus we have reduced a two body problem of masses M1 and M 2 to a one-
body problem of mass  and the whole molecule behaves as a simple
harmonic oscillator of mass  (reduced mass of the molecule) having force
constant k. Hence the eigenfunctions and eigenvalues of the molecules are
given by Eq. (7.30a) and Eq. (7.30b), respectively. These equations have
been very useful in understanding the vibrational spectrum (obtained in the
near infrared region of the electromagnetic waves) of diatomic molecules. The 101
analysis of the experimental spectrum has yielded force constants of a large
number of heteronuclear molecules.
The quantum oscillator has implications even beyond the simple diatomic
molecule. It is the foundation for the understanding of complex modes of
vibration in larger molecules, for example polyatomic molecules can also be
modelled by coupled harmonic oscillators. The atoms in large molecules are
treated as point masses and the bonds connecting then act (approximately)
like springs, the motion of atoms in a solid lattice, the theory of heat capacity,
etc. Let us consider a simple example of such a coupled system.
Coupled spring-mass system
Consider the system shown in Fig. 7.8 where two blocks of mass m are
 k 
connected by springs of spring constant k     . The displacement of the
 m 
two masses are x A and x B respectively.
The Hamiltonian of the system is:
p 2 p 2 1

H  A  B  m2 x A 2  x B 2  ( x A  x B ) 2
2m 2m 2
 (7.45)
x  xA x  xB
m m
x
x 0
Fig. 7.8: Coupled spring-mass system.

By making a transformation of coordinates:
1 1
x1   x A  x B ; x 2  x A  xB  (7.46)
2 2
The Hamiltonian of Eq. (7.45) reduces to:
p2 p 2 1 1
H  1  2  m2 x12  m 3 x 2 2
2m 2m 2 2
  (7.47)
Now this Hamiltonian corresponds to two decoupled simple harmonic

oscillations with two different frequencies and can be solved accordingly.
Would you like to apply the ideas discussed so far to some concrete situation?
Try the following SAQ.
So if the initial state of the system is normalized, it is normalized for all t.
SAQ 6
a) Consider a proton as a bound oscillator with a natural frequency of
3  10 21 Hz. Calculate the energy of its ground state and first excited state.
b) Calculate x and p x for the ground state harmonic oscillator
eigenfunction.
102
There is another method of solving the simple harmonic oscillator problem in
quantum mechanics. This method is sometimes called the algebraic method,
the method explained in Sec. 7.3 is also called the analytic method. We now
discuss the algebraic method in brief.
7.3 SOLUTION USING LADDER OPERATORS

We start once again with the stationary state Schrödinger equation
Hˆ ( x )  E( x ) . For the simple harmonic oscillator
pˆ 2 1
Hˆ   m2 xˆ 2  pˆ  (mxˆ )2 
1 2
(7.48)
2m 2 2m
So Hˆ ( x )  pˆ  (mxˆ )2 ( x )  E( x )

1 2
(7.49)
2m
Notice that Ĥ can also be written as:
1
Hˆ  mxˆ  ipˆ mxˆ  ipˆ  (7.50)
2m
We now define two operators â and â  as follows
1 1 d
aˆ  (mxˆ  ipˆ )  (mxˆ   ) (7.51a)
2m 2m dx
1 1 d
aˆ †  (mxˆ  ipˆ )  ( mxˆ   ) (7.51b)
2m 2m dx
Notice that the operators â and â  are the adjoints of each other. Now
1
aˆ aˆ †  (ipˆ  mxˆ ) ( ipˆ  mxˆ )
2m

1
pˆ 2  (mxˆ )2  impˆ xˆ  xˆpˆ 
2m

1
pˆ 2  (mxˆ )2  imxˆ, pˆ  (7.52)
2m
Using xˆ, pˆ   i in Eq. (7.52), we get:
aˆ aˆ † 
1
2m

pˆ 2  (mxˆ )2  m 
1  pˆ 2  1  1  1
     m2 xˆ 2   (7.53)
  2m    2  2
With this we get:

pˆ 2 1  1
 m2 xˆ 2  Hˆ   aˆ † aˆ   (7.54)
2m 2  2
So the stationary state Schrödinger equation (Eq. 7.49) for the simple
harmonic oscillator reduces to
 1
 aˆ † aˆ   ( x )  E( x ) (7.55)
 2
You can verify for yourself that the commutator of â and â  is:
[aˆ, aˆ † ]  1 (7.56) 103

SAQ 7
Show that [aˆ, aˆ  ]  1 .
Now let us consider the action of the operator Ĥ on aˆ † ( x ) .

 1  1 
Hˆ (aˆ †  )    aˆ † aˆ  (aˆ †  )    aˆ † aˆ aˆ †  aˆ †  
 2  2 
 1
  aˆ †  aˆ aˆ †    (7.57)
 2 
From the commutator relation of Eq. (7.53) we can write

aˆ aˆ †  aˆ † aˆ  1  aˆ aˆ †  aˆ † aˆ  1 (7.58)
Thus Eq. (7.57) reduces to

 1   1 
Hˆ (aˆ †  )   aˆ †  aˆ † aˆ  1     aˆ †   aˆ † aˆ     
 2   2 

 aˆ † Hˆ     (7.59)
From Eq. (7.49), Ĥ   E , so Eq. (7.59) reduces to
 
Hˆ (aˆ †  )  E   aˆ †  (7.60)
Clearly â †  is a stationary state of the Hamiltonian with an eigen energy

greater than that of the stationary state ( ) by  .
ˆ (x) .
Now let us consider the action of the operator Ĥ on a
 1  1 
Hˆ (aˆ  )    aˆ † aˆ  (aˆ  )    aˆ † aˆ aˆ  aˆ   (7.61)
 2  2 
From the commutator relation of Eq. (7.56) we can also write

aˆ aˆ †  aˆ † aˆ  1  aˆ † aˆ  aˆ aˆ †  1 (7.62)
Thus Eq. (7.61) reduces to

  

  1  1
Hˆ (aˆ  )    aˆ aˆ †  1 aˆ  aˆ    aˆ   aˆ †aˆ     
2
 2    

 aˆ Hˆ     (7.63)
Using Ĥ   E , Eq. (7.63) reduces to
Hˆ (aˆ  )  E   aˆ   (7.64)
Clearly â is a stationary state of the Hamiltonian with an eigen energy less
than that of the stationary state  by  .
Eqs. (7.60) and (7.64) therefore show that given any stationary state  , we
can generate eigen states of the Hamiltonian with higher or lower energies by
operating on  by the operators â† or â respectively.
104
â and â† are called the ladder operators, because they help us climb down
or climb up to energy states with lower or higher energies.
Does this then mean that using the operator â on any stationary state,  ,
repeatedly, we can find energy eigen states with lower and lower energies,
even with energy less than zero? That is not the case, because we know that
the eigen energy of the simple harmonic oscillator must be positive definite.
Therefore, there must be a state, which we can denote as the lowest energy
state or ground state  0 ( x ) , for which
aˆ  0 ( x )  0 (7.65)
We use Eq. (7.65) and Eq. (7.51a) to write the following differential equation
for the ground state eigenfunction:
1   d  mx   ( x )  0  d 0   m x (7.66)
  0 0
2m  dx  dx 
Eq. (7.66) can be solved to get:
m
 x2
 0 ( x )  N 0 e 2 (7.67)
Where N 0 is an arbitrary constant. We can determine N 0 by normalizing the


wave function using the normalization condition   0* ( x )  0 ( x ) dx  1 . This is a

calculation we have already done (SAQ 3b ) and we write the result
straightaway as:
1 / 4  m x 2
m 
 0 ( x )    e 2 (7.68)
  
We can also derive the ground state energy using Eq. (7.65). Note that:
Hˆ  0 ( x )  E0  0 ( x )
 1
   aˆ †aˆ    0 ( x )  E00 ( x )
 2

or aˆ  aˆ  0 ( x )   0 ( x )  E0  0 ( x ) (7.69)
2
Since aˆ  0 ( x )  0 , Eq. (7.69) gives us the value of the ground state energy as:
1
E0   (7.70)
2
Now, we can operate â† on  0 ( x ) to get the eigenfunction which has an
1 3
energy E       . This is the eigenfunction 1( x ) of Sec. 7.2 .
2 2
So:
 1
1( x )  N1(aˆ † )  0 ( x ); E1   1  
 2 
 3  1
 2 ( x )  N2 (aˆ † ) 1( x )  N2 (aˆ † )2  0 ( x ); E2   1     2  
 2  2
 5  1
 3 ( x )  N3 (aˆ † )  2 ( x )  N3 (aˆ † )3  0 ( x ); E3  1     3  
 2  2 105
(7.71)
where N1 , N 2 , N 3 are the respective normalization constants.

So the nth eigenfunction and the corresponding eigen energy is:
 1
 n ( x )  Nn (aˆ † )n  0 ( x ); En   n   (7.72)
 2
So Eq. (7.72) is a method of determining the eigenfunctions (upto a
normalization constant) and eigen energies of the simple harmonic oscillator.
The normalization constant may always be calculated using the normalization
condition.
You may like to calculate the eigenfunctions 1( x ) for practice.
SAQ 8
Use Eq. (7.72) to obtain 1( x ) . Is the wave function normalized?
The ladder operators are an elegant way of arriving at the eigenfunctions of

the simple harmonic oscillator and we shall use this method once again in Unit
12 of this course.
7.4 SUMMARY
 The simple harmonic oscillator potential is important in physics
because in the neighbourhood of a local minimum (say at x = a),
almost any arbitrary potential function V (x ) can be approximated by
a parabolic potential
1 d 2V
V ( x )  k ( x  a ) 2 where k 
2 dx 2 x  a
 The time independent Schrödinger equation for the one-dimensional

simple harmonic oscillator is:
 2 d 2 1 2 2
 2m 2  2 m x  ( x )  E( x )
 dx 
where E is the total energy of the oscillator. In terms of the
 
dimensionless variables   ax, with a 2  m /  and E  , the
2
Schrödinger equation is
d 2 (  )
 (    2 ) (  )  0
d 2
 The general solution of the Schrödinger equation is

1 1
 2  2
( )  h( ) e 2  c0S1()  c1S2 ( ) e 2
where c0 and c1 are arbitrary constants, and S1() and S 2 ( ) are
series which contain even and odd powers of  respectively, with:
1   2 (1   )(5   ) 4
S1( )  1      ...
2! 4!
106
and
3   3 (3   ) (7   ) 5
S2 (  )        ..
3! 5!
On applying the boundary condition ( )  0 for    ,
the values of  are found to be discrete and hence the energy of the
quantum simple harmonic oscillator is quantized and:
1
 energy eigen values are E n   n    ; n  0,1,2,...
 2
 eigen functions are

1/ 2
 a 
 n ( x )    H n (ax ) exp( a 2 x 2 / 2), n  0,1, 2,...
n 
  2 n! 
where H n (ax ) are the Hermite polynomials of order n.
 The eigenfunctions of the simple harmonics oscillator potential are real

and have a definite parity. The eigenfunctions corresponding to zero or
even values of n are of even parity  n (  x )   n ( x ) . On the other
hand odd parity eigenfunctions ((  n (  x )   n ( x ) ) correspond to odd
values of n.
 The general solution for the time dependent Schrödinger equation for
the simple harmonic oscillator is:
1
 i ( n  )t
( x, t )   an  n ( x )e 2
n  0,1,2...

where an    *n ( x ) ( x,0) dx , (x,0) being the wave function of the

system at t = 0.
 The energy eigenvalues and eigenfunctions of a simple harmonic
oscillator can be determined using ladder operators â and â† which
are defined as follows:
1
aˆ  ( mxˆ  ipˆ ) ;
2m
1
aˆ †  (mxˆ  ipˆ )
2m
â and â  are called the ladder operators, because they help us climb
down or climb up to energy states with lower or higher energies
respectively:
Hˆ (aˆ  )  E   aˆ   ; Hˆ (aˆ †  )  E   aˆ †   

The nth eigenfunction is given by:
 n ( x )  Nn (aˆ † )n  0 ( x )
Where  0 ( x ) is the ground state eigenfunction.
107

1. An oscillator consists of a tiny object of mass of 10−12kg. It vibrates on the
end of a thin fiber with a maximum amplitude of 10−3 m and a frequency
104Hz. Calculate (i) the approximate quantum number for the oscillator
when its energy is equal to its classical energy (ii) its energy in its lowest-
energy state and (iii) its classical amplitude of vibration if it were in its
lowest-energy state.
2. Show that the average value of x for a simple harmonic oscillator in the nth
quantum state is zero.
3. Calculate x̂ 2 and pˆ x 2 for the simple harmonic oscillator in its ground
state.
4. Calculate the mean kinetic and potential energies of a simple harmonic
oscillator in its ground state.
5. The potential energy of a particle of mass m is given by:
1
V (x)  m2 x 2 for x  0
2
 for x  0
Show that the eigen energies are given by:
 3
E 2k  1  2 k   for k  0,1,2,...
 4
6. For what value of  and  is the following wave function
x 2

( x )  (x 2  1) e 2 a solution of the stationary state Schrödinger
equation for the simple harmonic oscillator ?
7. Using the results of TQ 2 and SAQ 7(b) calculate x and  p x for  0 ( x )
. Show that x p x   / 2 .
8. Using the method outlined in Sec. 7.3, obtain the wave function  2 ( x ) .
9. The initial state of the simple harmonic oscillator of mass m and frequency
 is:
( x,0) 
1
3
0 ( x )  i 
2 1( x )
where  0 , 1 are the normalized eigenfunctions for the simple harmonic

oscillator. Determine
i) ( x, t ) and the probability density ( x, t ) 2 . Show that the probability
density returns to its original shape after the classical period T  2 / 
ii) the expectation value of x̂ and p̂ in the state ( x, t ) . Show that x̂
oscillates with time. Calculate the frequency and amplitude of its
oscillation.
1
10. Consider a particle of mass m in a potential V (r ) m2r 2 , also called
2
108 the 3D isotropic harmonic oscillator. Write down the Schrödinger equation
for the system in Cartesian coordinates and determine the energy
eigenvalues.
11. A carbon monoxide molecule has an effective mass of 6.85 amu and an
effective spring constant of 1860 Nm 1. Calculate the zero point energy
and the energy required to excite the carbon monoxide molecule to its first
excited state.

d d  d d d 2 d 2
1. Since   ax we can write:  a ;  a2
dx d dx d dx d 2
So Eq. (7.3) is
 2 a 2 d 2  1 m 2 2
     E (  ) (i)
2m d 2 2 a 2
2m
Multiplying Eq. (i) by we get:
 2a 2
d 2 (  ) m 2  2 2 2m
   (  )  E (  )
d 2  2a 2  2a 2
d 2 (  ) m 2  2 2 2m
   (  )   E (  )
d 2 2
 a 4  2a 2
m  m 22
Since a 2   a4  we get:
 2
d 2 (  ) 2E
  2 (  )   (  )
d 2 
d 2 (  ) 2E
Or     2 ()  0 with  
d 2 
2. From Eq. (7.14) we can write for n  0,1, 2, 3, 4
1
1   2 (1   )(5   ) 4   2
 n ( )  c 0 1      e 2 for n  0,2,4
 2! 4! 
1
3   3   2
 c1    e 2 for n  1,3
 3! 
  2n  1  1  n ( )
n En   n    
 2
0 1  1
 2
2  0 ( )  c 0 e 2
1 3 3  1
 2
2 1()  c1e 2
109
2 5 5  1
 2 ()  c 0 1  2 2 e 2
 2
2
3 7 7  4   2
1

2  3 ()  c1   3 e 2
 3! 
1
2  2
 c1    3  e 2
 3 
4 9 9  8 32 4   2
1
2  4 ( )  c0 1   2   e 2
 2! 4! 
1
4  2
 c0 1  4 2   4 e 2
 3 
 
a
 
2axe  a x dx  0 , since the integral is
2 2
3. a) i)  0 ( x ) 1( x ) dx 
To evaluate the 
2  
integrals in this unit,
odd.
you may Table 3.1 of
Unit 3 , Block 1.  
a 2 2
ii)   1( x )  2 ( x ) dx   [2ax(4a
2 2
x  2) ]e a x dx

4  
 8a 3 x 3  4ax  e  a x
a 2 2
 dx  0
4  
(since the integral is odd)

b) i) Using the normalization condition:
 
  e a x
2 2
N 02  02 ( x ) dx  1  N 02 dx  1
 

 N02 1
a2
1 1
 a  2  m  4
 N0     
    
ii) The normalization condition is
 
  4a 2 x 2e  a x
2 2
N12  12 ( x )dx 1  N12 dx  1
 

2 a 2 x 2 1  2  2
 4a 2N12 x e dx  1  4a 2 N12 1 N1  1
2 a6 a

1 1
 a  2  m  4
 N1     
2    4  
1 2
4. i)  * ( x,0)  1* ( x )  i  *2 ( x )
5 5
 1 * 2 *  1 2 
And  * ( x,0)( x,0)   1( x )  i  2 ( x )  1( x )  i  2 ( x )
 5 5  5 5 
110
 
1 4 
   * ( x,0) ( x,0) dx    5 1* 1  i 2 5  *2 2  dx
 
 
1 4

5  1* 1 dx 
5   *2 2 dx  1
 
 
ii) Since 1,  2 are normalized eigenfunctions,  1* 1 dx    *2 2 dx  1
 
The wave function at time t is (Eq. 7.32):

1 2
( x, t )  1( x )e  iE1 t /   i  2 ( x ) e  iE2t / 
5 5
i ( 3t ) i ( 5t )
1  2 
 1( x )e 2 i 2 ( x) e 2
5 5

iii) Hˆ    * ( x,0)Hˆ ( x,0)dx

 1 2 
Hˆ ( x,0)  Hˆ  1( x )  i  2 ( x )
 5 5 
3 5
You know that: Hˆ 1  E1 1   1 ; Hˆ  2  E 2  2   2
2 2
1 3  2 5 
So Hˆ     1( x )  i     2 ( x )
5 2  5 2 

 1 * 2 *  1 3  2 5  
 Hˆ   5
1  i
5
2  
  5
    1  i
 2  5
   2  dx
2  

 
1 3  4 5 

5 2  
   1* 1 dx      *2 2 dx
5 2  
 
 
25  * 23  * 23
52  
i    1 2 dx  i      21 dx  .
52  10
 
(  1* 1* dx    *2  2dx  1 and  1*  2dx    *2 1dx  0 )

5. a) Let the classical turning points be  x 0 . At these points the energy of
1
the classical oscillator is m2 x 02 and we can write:
2
1/ 2
1 1
m2 x 02  E n   n     x 0   (2n  1)

2  2  m 
1
We can also write this as x 0  (2n  1)1 / 2  ax 0  (2n  1)1 / 2 .
a
Writing  0  ax 0 we can write the classically allowed region as:
 0  (2n  1)1 / 2 .
b) A general harmonic oscillator state is

111
1 it
 i ( n  )t 
 ( x, t )   an  n ( x )e 2 e 2  an  n ( x )e  int
n  0,1,2... n  0,1,2...
(i)
2
With T  , we can write

i( t T )

 ( x, t  T )  e 2  an  n ( x )e  in(t T ) (ii)
n  0,1,2...
Now e in( t T )  e int e inT  e int e in( 2)  e int (iii)

Using Eq. (iii) in Eq. (ii) we get
iT   it  iT
 
( x, t  T )  e 2 e 2


an  n ( x )e  int   e 2 ( x, t )

n  0,1,2...
(iv)
iT
 * ( x, t  T )  e 2  * ( x, t ) (v)
So
2 2
( x, t  T )  ( x, t ) and the probability distribution has a time
period T.
1
6. a) E n   n    , so for the ground state n  0, and
 2 
1 1 1
E0    h   6.626  10 34 Js 3  1021 Hz
2 2 2
 9.939  10 13 J
For the first excited state, n  1
3 3
E1    h  3E 0  2.982  10 12 J
2 2
b) For the ground state harmonic oscillator wave function
 
 a 

2 2
x  dx 0* ( x )x 0 ( x ) dx     x exp( a x ) dx
   

The integrand is an odd function. Hence the integral over the interval
will be zero.
 x 0


Similarly, p x  i   0* ( x ) x  0 ( x ) dx


 a   a2x 2   a2x 2 
  i
 
 exp 


2 
 ( a 2 x ) exp 
 2 
 dx


a3
i
 

 x exp ( a 2 x 2 ) dx
112 Again the integrand is an odd function of x,

 px  0
 1 1 
7. [aˆ, aˆ † ]   (mxˆ  ipˆ ), (mxˆ  ipˆ )
 2m 2m 

1
m 2 2 xˆ, xˆ   pˆ, pˆ   impˆ, xˆ   imxˆ, pˆ 
2m
1
 im i   imi   1
2m
since xˆ, xˆ   pˆ, pˆ   0 and xˆ, pˆ   i.
1/ 4 mx 2
8. 1( x )  N1(aˆ † )  0 ( x ) 
N1    d  mx   m  e  2 (i)
  
2m  dx    
m m
   d  e  2 x  mxe  2 x
2 2
and   (ii)
 dx 
Therefore Eq. (i) reduces to
1 / 4  mx mx 2
2
1/ 4
N1 m  m  2m 
1( x )  2mx    e 2    xe 2
2m       

To determine N1 we must calculate  1* ( x ) 1( x ) dx :

  m 2
m  2m   x
 1* ( x ) 1( x ) dx  N1 2 
   
  x 2e  dx  1  N1  1
 
So the wave function is normalized.
Terminal Questions
1. m  10 12 kg;   2  6.28  10 4 s 1 ; Amplitude A  10 3 m
1 1
m2 A 2   10 12  6.28  10 4   10  3   19.7  10 10 J
2 2
E
2 2
(i)
E n   n       n   6.054  10  34  6.28  10 4 
1 1
 2  2
 
 38.01 10 30  n  


1
2
(ii)
n  .5  10 20
1
ii) The zero point energy    19.0  10 30 J
2
1
iii) We have m2 A2  19.0  10 30 J
2
38.0  10 30
 A2   1034 m2

10 12  6.28  10 
4 2
or A  10 17 m .
113

2. x  (  n , x n )    n* ( x ) n ( x ) dx

Since  n (x ) is of definite parity, it is either an odd or an even function of x.
in either case, the product  n* ( x )  n ( x ) will be even. Since x is odd, the
integrand will be an odd function of x and hence the integral will be zero.
3. The expectation value of x̂ 2 in the ground state is
 
a a 1   1
xˆ 2   x 2e  a x
* 2  
 0 ( x ) x 0 ( x ) dx 
2 2
dx 
     2 a 6  2a 2


 xˆ 2 
2m

d 2 0
pˆ x 2    0* ( x ) p x2  0 ( x ) dx and pˆ x2  0 ( x )   2
dx 2

1/ 2 a2x 2
d 0  a  
2 2
   (a x ) e
dx  
1/ 2
d 2
2
 a 
   a2
d  a 2 x 2 2 
 xe
dx 


a5 / 2
1 / 2

e  a x / 2  a 2 x 2e  a x / 2
2 2 2 2

dx   ( )

 e  a x 
 2a 3 1 2 2 1
 a 2 x 2 e  a x dx
2 2 2 2
pˆ x2    a   m
  2 2
4. The average value of the potential energy V is given by


1
V    0* ( x ) 2 kx 2  0 ( x ) dx .

  
Using the results of TQ3    0* ( x ) x 2  0 ( x ) dx  we get
 2m  
  
k    m2    1
V        .
2  2m  2  2m  4
1
Since the ground state eigen energy is E0  
2
1 1
 K .E.  E 0     .
4 4
  pˆ 2 
pˆ x 2
(You can also evaluate K .E.     0* ( x ) x   0 ( x ) dx ).
2m  2m 
  
5. The potential function is identical to the SHO potential for x > 0. However
since V (x )   for x  0, the wave function must satisfy the condition:
( x  0)  0 (as in a rigid box), the only solutions that are acceptable are
those for which ( x  0)  0 .
 Only the odd parity (n = odd) eigenfunctions of the SHO will be

114 eigenfunctions of this potential. i.e., ( x ),  3 ( x ),  5 ( x ) etc.
We can label these eigenfunctions as  2k 1( x ), where k  0,1, 2 and
 1  3
E k   2k  1     2 k   
 2  4
2 2 2
6. We use   ( x 2  1) e  x  x 2e x  e  x in the stationary state
Schrödinger equation (Eq. 7.3).
d
 2(  ) xe  x  2x 3 e  x
2 2
(i)
dx
d 2 2 2
 2(  ) e x  (10  42 ) x 2e x  4 2 x 4 e  x
2
(ii)
2
dx
Substituting from Eq. (ii) into Eq. (7.3) we get
  2  5 2 2 22 1   2 22 m2 
e  x 2  (   )  x 2    m2  x 4   
 m  m m 2   m 2 
   
2
 E (x 2  1) e x
(iii)
Setting the coefficient of x 4  0 in Eq. (iii) we get
2 2 2 m2 m 2 2 m
  2   (iv)
m 2 4 2 2
Substituting for  in Eq. (iii) we get
d 2 1 2
2      5   
 2  m2 x 2   e  x    x 2  
2 2  2 
dx  m 2
5    x 2  2  1 2 
 e x   5  m 
2
1 2 2 m
Setting   1, we get  
5 m 5 
 2m x 2  1
  
 5 
 2 d 2 1 5
And   m2 x 2   
2m dx 2 2 2
5
 E 
2
7. Using the results of SAQ 2b and TQ 2
 2  m
x  xˆ 2  xˆ 2  ; p x  pˆ x 2  pˆ x 
2m 2
 m
x  p x   /2
2m 2
8. Using Eq. (7.72) and 1( x ) from SAQ 8
 1  mx 2
† N2  d   m  4 2m   2
 2 ( x )  N2 a 1( x )     m x     xe
2 m  dx      
 
115
1 1
mx 2 mx 2 
  

N2  m  4
 2m  2   d  xe  2   mx 2 e  2 
   
2m        dx   
   
1
mx 2 mx 2 
 m  4  m 2  
 N2   2 x e 2  e 2 
     
 
1  a2 x 2
m  4
 2a 2 x 2  1e
 a   2   2 / 2
 N 2  2  N2   2 2  1 e
    22  
9. i) Following SAQ 4, we write
it i 3t
1  2  2
( x, t )  2 
e 0(x)  i e ( x )
3 3
it i 3t
1 2
 * ( x, t )  e 2 0 ( x )  i e 2
3 3
( x, t ) 2 
1
3
2
 0 ( x ) 2  1( x ) 2 
3
i 2  i t
3
e 
 e it  0 1 
1 2 2 2 2
    2 sin t  0 1
3 0 3 1 3
1 2 2 2 2
  0  1  2  0 1 sin t
3 3 3
2
2 1 2  2  1 2 2 2
at t  0 ( x,0)   02   12   x, t    0  1
3 3   3 3

ii) xˆ    * ( x, t ) x  ( x, t ) dx

1  
2 ( x ) dx  2 x 2 ( x ) dx
  x 
3  0 3  1
 
2  it 2 it
i
3
e  x 0 ( x ) 1( x ) dx  i 3
e  x  0 ( x ) 1( x ) dx
 

 i
3

2 it
e  e  it   x0 ( x ) 1( x ) dx


2
2 sin t  x  0 ( x ) 1( x ) dx
3 
So x  x 0 sin t
116

2 2
where x0 
3  x 0 ( x ) 1( x ) dx

1

4a 2 2 a 2 x 2 4a 2  1   2 2   2
 x e dx  .  
 3a  3  m 
3   3   2 a6 
10. The stationary state Schrödinger equation is
 2 2 2 2 2 2
 2m 2  2m 2  2m 2  2 m x  y  z  ( x, y , z )  E( x, y , z )
1 2 2 2 2 
 x y z 
We can write the wave function as: ( x, y , z )   x ( x ) y ( y ) z ( z ) to get
  2 d 2 x 1    2 d 2 y 1 
   m2 x 2  x  y  z     m2 y 2 y  x  z
2
 2m dx 2 2   2m dy 2 
  2 d 2 y 1 
   m 2 z 2  z  x  y  E x  y  z
 2m dz 2 2 
 
We can reduce these equations to the following three equations each

corresponding to the Schrödinger equation for a one-dimensional quantum
oscillator:
 2 d 2 x 1
  m2 x 2  x  E x  x
2m dx 2 2
 2 d 2 y 1
  m2 y 2  y  E y  y
2m dy 2 2
 2 d 2 z 1
  m2 z 2  z  E z  z
2m dz 2 2
With the condition: E  E x  E y  E z
Using the results for the eigen energies of the simple harmonic oscillator,
we can write
1 1 1
E x   n x    ; E y   n y    ; E z   n z    
 2  2  2
Where n x , n y , n z  0,1,2,3....
So that
3 
E  E x  E y  E z    n x  ny  nz   
2 
117
k 1860 Nm 1 1860
11. 2    s2
 6.85 amu  27
6.85  1.66  10
   40.5  1013 s 1
1 1
E    6.054  10 34  40.5  1013 J  .1330 eV
2 2
3
E1    .4 eV  E1  E0  0.27 eV
2
118
Unit 8 Spherically Symmetric Potentials
UNIT 8
SPHERICALLY SYMMETRIC
POTENTIALS
Structure
8.1 Introduction 8.3 Space Quantization
Expected Learning Outcomes 8.4 Radial Eigenfunctions
8.2 Three Dimensional Schrödinger 8.5 Summary
Equation for a Central Potential 8.6 Terminal Questions
Schrödinger Equation in Spherical 8.7 Solutions and Answers
Polar Coordinates
Eigenfunction and Eigenvalues of
L̂z Spherical Harmonics
Parity of the Spherical Harmonics
8.1 INTRODUCTION
In the previous three units you have obtained the eigenfunctions and
eigenvalues of a number of one-dimensional systems. In this unit, we shall
extend our study to a three-dimensional system. Thus now there will be three-
independent variables, which can be x, y, and z, in the Cartesian coordinates
system or r ,  and  in the spherical polar coordinate system. Hence, the
number of degrees of freedom of the particle will increase from one to three
and the time independent Schrödinger equation will now be a three-
dimensional differential equation.
In general, the potential in which a particle moves in a three-dimensional
space is a function of all three coordinates. However, in the present unit we
shall confine ourselves to those potentials which depend only upon the radial
coordinate r and are independent of the polar coordinates  and . Such
potentials are known as spherically symmetric potentials and the
corresponding physical systems are called spherically symmetric systems.
1
The isotropic harmonic oscillator potential of TQ 9, Unit 7 V (r )  m2 r 2  is
 2 
an example of a spherically symmetric potential. Another example would be a
particle moving freeing in a spherical box of radius R0 :
 0 r  R0
V (r )  
 r  R0
The Coulomb potential governing the motion of electrons in a Hydrogen-like

atom is another example of a spherically symmetric potential:
119
Ze 2
V (r )  
r
In this unit we look at the properties of the motion of a particle in a spherically
symmetric potential. In the next unit we extend these ideas to the specific
example of the hydrogen atom.
In Sec. 8.2 we write down the steady state Schrödinger equation for a
spherically symmetric potential in spherical polar coordinates and separate the
equation into its radial and angular components. The solution of the radial
equation depends on the explicit form of the potential but the angular part
does not. We solve the angular part of the steady state Schrödinger equation
and obtain the eigenfunctions and eigenvalues of L̂2 and L̂z using the power
series method and applying the boundary conditions. The eigenfunctions of
are called the spherical harmonics. We describe the properties of the spherical
harmonics like their parity etc. In Sec 8.4 we introduce the concept of space
quantization for the angular momentum. In Sec 8.5 we describe the effective
potential energy for a particle in a spherically symmetric potential and
introduce the concept of the centrifugal barrier.
 separate the time independent Schrödinger equation for a spherically
symmetric system into its radial and angular parts;
 determine the eigenfunctions and eigenvalues of the z-component of the
angular momentum operator L̂z ;
 obtain the eigenfunctions and eigenvalues for L̂2 and show that the
angular momentum is a constant of motion for such systems;
 explain the concept of space quantization; and
 discuss the properties of the radial wave function.
8.2 THREE DIMENSIONAL SCHRÖDINGER

EQUATION FOR A CENTRAL POTENTIAL
Let us consider the three-dimensional motion of a particle of mass  in a
spherically symmetric potential. For its stationary states, the time independent
Schrödinger equation is given by
 2 2   
 2   V (r ) (r )  E(r ) (8.1)
 
where E is the total energy of the particle, V (r ) is its potential energy and

(r )  (r , , ) is the wave function of the particle ,that depends on all three of
its coordinates. Note that V (r ) is independent of the polar angles  and . The
 
   
force F acting on such a particle will be directed along r F  V . So
    
classically, the torque  on the particle is equal to r  F . Since F and r are

in the same direction,  is equal to zero. Furthermore, the torque is equal to

the rate of change of the angular momentum L . Hence, the angular
momentum of a particle moving under a spherically symmetric potential (which
120
is also known as a central potential) will not change with time, it is constant in
magnitude and direction and is a constant of motion for the object. In your
mechanics undergraduate courses you have studied that the angular
momentum is a constant of motion for all central force motion. Recall that you
used this property to obtain Keplerian orbits for planetary motion around the
sun.

However, a constant L means that all its three components Lx , Ly , and Lz are
constant simultaneously. This is not possible in quantum mechanics because
the three components of the angular momentum operator L̂ , Lˆx , Lˆy , and L̂z do
not commute among themselves (see Unit 4). Thus, there is a difference
between the classical and quantum interpretations of angular momentum.
Instead of relating angular momentum to torque (as in classical mechanics), in
quantum mechanics we find that the Hamiltonian can be written in such a way
that it depends only on the angular momentum. This in how angular
momentum makes its entry into the scheme of quantum mechanics.
We now write down the Schrödinger equation in spherical polar coordinates.
8.2.1 Schrödinger Equation in Spherical Polar Coordinates
In spherical polar coordinates,  2 is given by
1   2   1     1 2
2   r    sin    (8.2a)
r 2 r  r  r 2 sin      r 2 sin 2  2

Putting this expression for  2 in Eq. (8.1) and writing (r )  (r , , ) we get
   r 2    2 (E  V (r )) r 2  
 r  r   2 
 1   1 2 
  (sin  )   (8.2b)
 sin    sin2  2 

Eq. (8.2b) suggests that (r , , )) can be separated in the variables r ,  and
 as follows:
(r , , )  R (r ) Y (, ) (8.3a)

The normalization condition for  * (r ) is a volume integral over the
coordinates r , , 
    2
3 2
  * (r )(r )d r      * (r , , )(r , , )r sin drdd  1
r  0  0  0
(8.3b)
which is
    2 

 R * (r )R (r )r 2 dr 
 0 
   Y * ( , )Y (, ) sin dd  1
  0   0 
 
(8.3c)
Putting Eq. (8.3) in Eq. (8.2) and using the method of separation of variables
we get the following two equations
121
1   2 R (r )   2 K 
r    2 (E  V (r ))  2 R (r )  0 (8.4)
2 2
r r  r    r 
and
1   Y (, )  1  2Y (, )
 sin    K Y (, ) (8.5)
sin      sin2  2
where K is a constant. We can further show that the operators L̂2 and L̂z in
spherical polar coordinates are given by
 1     1 2 
Lˆ2   2   sin    (8.6a)
 sin      sin 2   2 
 
Lˆz  (8.6b)
i 
In fact, you can like to do this exercise yourself. Try the following SAQ.
SAQ 1
a) Prove Eqs. (8.6a) and (8.6b).

b) Show that L̂2 and L̂z commute with the Hamiltonian.
Using Eq. (8.6a), we can write Eq. (8.2b) as:
  r 2    2 (E  V (r )) r 2     Lˆ2  (8.7a)
 r   
 r   2 2
And Eq. (8.5) as:

Lˆ2 Y (, )  K 2 Y (, ) (8.7b)
From Eq. (8.7b) we can readily see that Y (, ) is an eigenfunction of the
operator L̂2 with the eigenvalue K 2 . We now obtain the value of K in Eq.
(8.7b) and determine the form of Y ( , ) .
You have established in SAQ 1 that L̂2 commutes with the Hamiltonian, so
Lˆ2 , Hˆ   0 (8.8a)
 
Recall Eq. (4.59) of Unit 4 which tells us that for any operator D̂ , if Dˆ , Hˆ  0 ,
d Dˆ
then  0, i.e., D̂ is constant. Applying this result to Dˆ  Lˆ2 we get
dt
L̂2  constant (8.8b)
So in quantum mechanics, the expectation value of the square of the angular

momentum is a constant of motion for a central potential.
8.2.2 Eigenfunctions and Eigenvalues of L̂z

We can solve for Y ( , ) by separating the variables  and  and writing
Y (, )  P ()  () . (8.9a)
Substituting Eq. (8.9a) and Eq. (8.6a) in Eq. (8.7b), and using the method of
122 separation of variables, we get
 sin  d  dP ( )  2  1 d 2
 P () d  sin  d   K sin     d 2 (8.9b)
 
Eq. (8.9b) shows that both sides are equal to the same constant. Let us
denote this constant by ml2 . So we get
1 d 2
 ml2 (8.10)
 d 2
and its solution is
()  Ae iml  (8.11)
Where A is a constant. Now () also has to be single valued for all  , hence
we must have
()  (  2) or Ae iml   Ae iml (   2) (8.12)
angles   0 and   2 are actually the same. So: Ae ( 2iml )  1 and therefore
ml has to be an integer:
ml  0,1,2,....... (8.13)
To determine the constant A , we apply the normalization condition
2
  * (  ) (  ) d   1 (8.14)
0
or
2
 iml  1
e iml d  1  A 
2
A e 2
(8.15)
0
1
So finally:  ml ()  e iml  (8.16a)
2
Notice that we have labelled the function () by the index ml where the
allowed values of ml are given by Eq. (8.13). You can check that when the
operator L̂z (Eq. 8.6b) operates upon  ml () (Eq. 8.16a) we get
Lˆz  ml ()  m l   ml () (8.16b)
Thus,  ml () is an eigenfunction of the operator L̂z with the eigenvalue

ml  .
8.2.3 Special Harmonics

Now that we have derived  ml () , we can write Eq. (8.9a) as:
Y (, )  P ()  ml ( ) (8.16c)
The differential equation for P () is
d  dP ( ) 
sin   sin    K P () sin 2   m l2 P () (8.17)
d  d 
Eq. (8.17) can be solved analytically by the power series method. We
introduce a variable v  cos  and write the ODE of Eq. (8.17) in terms of this
new variable v as (SAQ 3):
2
d (1  v 2 ) dP (v )   K  ml P (v )  0 (8.18)
  
dv d   (1  v 2 ) 
123
SAQ 2
Derive Eq. (8.18) from Eq. (8.17).
All that we know about Eq. (8.18) is that ml is an integer. To determine the
function P (v ) , we take two particular cases.
Case I: ml  0
For ml  0 Eq. (8.18) reduces to:
d  2 dP (v ) 
dv (1  v ) dv   KP (v )  0 (8.19a)
 
d 2P (v ) dP (v )
or (1  v 2 )  2v  KP (v )  0 (8.19b)
2 dv
dv
Notice that this equation has regular singularities at v  1 but is analytic at
v  0 . Therefore in the range  1  v  1 ( 0    ) , we can choose a power
series solution to this equation of the form

P (v )   ck v k (8.20a)
k  0,1,2,...
dP d 2P
So  
c k k v k 1;
dv k  0,1,2..

dv 2 k  0,1,2..

c k k (k  1) v k  2 (8.20b)
Substituting from Eq. (8.20a and b) into Eq. (8.19a) and collecting the
coefficients of v k we get the relation between the coefficients as:
k ( k  1)  K
ck  2  ck (8.20c)
(k  2) (k  1)
and
K K 2 K 2 K 6
c2  
c0 ; c3   c1; c 3   c1; c 4   c0
2 6 6 12
(8.20d)
The general solution is
P (v )  c 0 S1 (v )  c1S 2 (v ) (8.21)
where c 0 and c1 are arbitrary solutions and S1 (v ) , S 2 (v ) are series in even

and odd powers of v respectively. S1 (v ) , S 2 (v ) are linearly independent
solutions. Let us check how the series behaves for large values of k. As
k :
1 K
1(1  ) 
c k k 1
limk  k  2  lim 1 (8.22a)
ck 0 2 1
k (1  ) (1  )
k k
 
And P (v )  c0 1  v  v 2  .. .  c0 
 1 

 1 v 
(8.22b)
124
The series in Eq. (8.22b) diverges at v  1. The only one can get a physically
acceptable solution for P (v ) is if the series were to terminate at some
value of k, i.e. c k  2  0 for at some value of k. So
k (k  1)  K
ck  2  c k  0  K  k (k  1) (8.22c)
(k  2) (k  1)
where k  0,1,2,... . Suppose we write K  l (l  1) , the series terminates when
l (l  1)  k (k  1) for some integer value of k  0 . We can just pick l  k to
get the allowed values of l to be 0,1,2.... Thus the values of K are quantized,
just like the values of ml .
Suppose the series terminates at a particular value of k  l . If this value of l is
 even ( l  0,2,4,... ) then the even series S1 (v ) will terminate with the highest
value of v in S1 (v ) being v l but the odd series, S2 (v ) will still be an infinite

series. Therefore for even values of l we must have c1  0 .
 odd ( l  1,3,4,... ) then the odd series S2 (v ) will terminate with the highest
value of v in S2 (v ) being v l but the even series, S1 (v ) will still be an infinite
series. Therefore for odd values of l we must have c 0  0 .
The resulting solutions for different values of k are called the Legendre
polynomials. We can label the Legendre polynomial of order l by Pl (v ) and
it is a polynomial of degree k. You have studied about Legendre polynomials
in Unit 4 of MPH-001.The first few Legendre polynomials are
l  0  P0 (v )  1 (8.23a)
l  1  P1(v )  v (8.23b)
l  2  P2 (v ) 
1
3v 2  1 (8.23c)
2
l  3  P3 (v ) 
1
5v 3  3v  (8.23d)
2
You have already studied the properties of Legendre polynomials in Unit 4 of
MPH-001 and you know that:
1
2
 Pl (v )Pl  (v )dv  2l  1  ll  (8.24)
1
Therefore for ml  0 , relabeling Y (, ) as Yl ,m l (, ) in Eq. (8.16b) to

account for the values of l and ml we write
Yl ,ml  0 ( , )  Yl ,0 (, )  APl (cos )  ml  0 ()

1 
 Al Pl (cos )  (8.25)
 2
Where Al is the normalization constant. We can get the value of Al using Eq.
(8.24). So
1/ 2
2l  1 
Yl ,0 (, )    Pl (cos ) (8.26)
 4 
125
Where l  0,1,2,...
Next we solve for the more general case .
Case II: ml  0
Let us first consider the case of ml  0 . We start by writing the solution of

Eq. (8.18) as
P (v )  1  v 2 
ml / 2
g (v ) (8.27)
Differentiating once with respect to v we get:
 
 
ml
dP (v ) d  1  v 2 2 g (v )

dv dv  
  (8.28a)
   
ml ml
dg (v ) 1
 1 v 2 2  ml v 1  v 2 2 g (v )
dv
Multiplying Eq. (8.28a) by 1  v 2 we get  

1  v dPdv(v )  1  v   m v 1  v 
ml ml
2 2 2  1 dg (v ) 2 2
l g (v ) (8.28b)
dv
d  
     
ml ml
d  2 dP (v )  2 2 1 dg (v ) 2 2 
And  1  v  1  v  m v 1  v g (v ) (8.28c)
dv  dv 
l
dv  dv 
 
   
ml 2 ml
2 2 1 d g dg
 1 v  2ml  1v 1 v 2 2
2 dv
dv
  
ml
1
 ml ml  1v 2  1 1  v 2 2 g (v ) (8.28d)
Substituting from Eq. (8.28d) into Eq. (8.18) we get:
 
ml m m
1  v    
2 l l
2 2 1 d g dg 1
 2ml  1v 1  v 2 2  ml ml  1v 2  1 1  v 2 2 g (v )
2 dv
dv
ml
 
 ml2 
 K   1  v 2 2 g (v )  0 (8.29a)

 (1  v 2 ) 


ml

Multiplying Eq. (8.29a) by 1  v 2 2 we finally get:
1  v  ddv g  2m  m m  

2 1
dg
2
l  1v l l  1v 2  1 1  v 2 g (v )
2 dv
 ml2 
 K   g (v )  0 (8.29a)
 (1  v 2 ) 
which gives us
(1  v 2 )
d 2g
dv 2
 2(ml  1)v
dg
dv
 K  ml (ml  1) g  0   (8.29b)
126
Now when ml  0 , Eq. (8.29b) is just Eq. 8.19b ( with g (v )  Pl (v ) ) and the
solution is the Legendre polynomial Pl (v ) . So for ml  0 , g (v )  Pl (v ) .

Taking the derivative of Eq. (8.29) we get:
d 3g d 2g
(1  v 2 )  2( ml  1  1)v
dv 3 dv 2
  
 K  m l  1 ( m l  2)
d 2g
dv 2
0 (8.30)
dg
This is the same equation with g (v )  and ml  ml  1 . So if Pl (v ) is a
gv
dPl (v )
solution of Eq. (8.29b) with ml  0 , then is a solution of Eq. 8.30 or
dv
d 2Pl (v )
Eq. 8.29 with ml  1 . In the same way is a solution of Eq. (8.29b) with
dv 2
ml  2 and so on. In general, therefore the function:
d ml
Plml (v )  1  v 2 
ml / 2
Pl (v ) (8.31)
dv ml
is a solution of Eq. (8.18) for arbitrary ml .
The functions defined in Eq. (8.31) are called the associated Legendre
functions Plml (v ) which are obtained by differentiating the Legendre
polynomial ml times and multiplying the result by 1  v 2 

ml / 2
. Note that
Plml (v ) is not Pl (v )ml . This definition holds for ml  0 , with Pl0 (v )  Pl (v ) .
Since the highest power of v in Pl (v ) is v l , taking the derivative ml times
reduces the highest power of the polynomial to l  ml . So the function Plml (v )
is non zero only for ml  l . The limiting value of ml is therefore l .
So far we have only considered only values of ml  0 . Notice that Eq. (8.18)
contains only ml2 and not ml , so the ODE is invariant under the
transformation m l  m l . So the solutions for  ml are proportional to each
other and it can be shown that:
 ml  (l  ml )!  ml
Pl (v )   1ml  Pl (v ) (8.32)
 ( l  ml )! 
ml
where Pl (v ) is:
d ml
(v )  1  v 2 
ml ml / 2
Pl Pl (v ) (8.33)
dv ml
with v  cos  and ml  l . So we have
 l  ml  l (8.34a)
Alternatively Eq. (8.33) can also be written as ( with v  cos  ) :
ml d ml
Pl (cos )  sin   ml Pl (cos ) (8.34b) 127
dv ml
Quantum mechanically acceptable solutions of Eq. (8.7) with K equal to
l (l  1) are obtained when the l is equal to one of the integers
l  0,1, 2,..... (8.34c)
And for a given integer l there will be following (2l  1) values of ml :
 l ,  l  1,  l  2,...,0,..., l  1, l (8.35)
The general normalized solution of Eq. (8.7) for ml  0 is:

1/ 2
 2l  1 (l  ml )! 
Yl ,ml (, )  ( 1)ml    Pl ml (cos ) e iml 
 4  ( l  ml )! 
(8.36)
And for ml  0
Yl ,ml (, )   1ml Yl ,ml (, )* (8.37)
SAQ 3
a) Use the Legendre polynomial P1 ( x )  x and Eq. (8.34b) to construct
P10 ( x ) and P11( x ) .
b) Use Eqs. (8.36 and 8.37) to construct Y1,0 (, ) , Y1,1 (, ) and Y1,1 (, ) .
Being eigenstates of Hermitian operators with different eigenvalues, spherical

harmonics with different l and ml subscripts are automatically orthogonal.
Therefore
 Y *l ,ml (, )Yl ,ml (, )d   ll   mm (8.38)
Where d is the integration over the solid angle.

The eigenvalue equations for L̂2 and L̂z are
Lˆ2 Yl ,ml ( , )  l (l  1) 2 Yl ,ml (, ) (8.39a)
Lˆz Yl ,ml (, )  ml  Yl ,ml (, ) (8.39b)

The normalization constant is conveniently derived using ladder operators and
the method is elaborated in Unit 13. Here we have just quoted the results for
completeness.
The functions given by Eq. (8.36) and (8.37) are known as spherical
harmonics. The (2l  1) eigenfunctions Yl ,ml (, ) corresponding to the same
l but different ml (ranging from  l to l ) have the same eigenvalues for L̂2 .
Hence we can say that these eigenfunctions are ( 2l  1) fold degenerate
eigenfunctions of L̂2 . Yl ,ml (, ) are the angular part of the steady state
eigenfunctions of all particles moving under spherically symmetric
potentials. They form an orthonormal set and any function of  and  can be
expressed as a linear combination of Yl , m l (, ) . We give here the explicit
forms of some of the lower order Yl , m l (, ) for ready reference.
1 1 / 2
Y00    (8.40a)
 4 
1/ 2
3 
Y10    cos  (8.40b)
128  4 
3 1/ 2
Y1,1     sin  e  i (8.40c)
 8 
5 1 / 2
Y20    (3 cos 2   1) (8.40d)
 16 
1/ 2
15
Y2,1     sin  cos  e  i (8.40e)
 8 
15 1 / 2
and Y2,2    sin 2  e  i 2 (8.40f)
 32 
The squares of some of these functions (which also represent the angular part
of the wave function of the hydrogen atom) are shown in Fig. 8.1. You may
now like to perform an exercise based on the ideas discussed so far.
(a) (b) (c)
(d) (e) (f)
2 2 2 2
Fig 8.1: Polar plots of Yl ,ml ( , ) (a) Y0 ,0 ( , ) (b) Y1,0 (, ) (c) Y1,1 ( , )
2 2 2
(d) Y2 ,0 ( , ) (e) Y2 ,1 ( ,  ) and (f) Y2 ,2 ( , ) .
SAQ 4
a) Show that Yl ,ml (, ) is an eigen function of Lz . Determine its
eigenvalues.
b) Show that Y2,2 is normalised and is orthogonal to Y2,2 .
8.2.4 Parity of the Special Harmonics

Before studying further, it would do us well to examine the parity of the
spherical harmonics. For this, reflect, Yl , ml (, ) about the origin. It such a
reflection  changes to    and  changes to    . Now
129
ml d ml
Yl ,ml (, )  Constant Pl ml (cos ) e iml  with Pl cos   sin  ml
dv ml
Pl (cos ) . Now
e iml (   )  ( 1)ml e iml  (8.41a)

Further sin(   )  sin  and cos(  )   cos  . Therefore for even values of
l, the function Pl (cos ) will be of even parity and for odd values of l, Pl (cos )
will be of odd parity. So
parity transformation
Pl (cos )       ( 1) l Pl (cos ) (8.41b)
ml th th
And since Pl (cos ) contains the ml derivative w.r.t. v  cos  , the ml
 ml
order derivative will give a factor of ( 1) . Hence,
ml ml
d parity transformation  ml d
      ( 1) (8.41c)
ml ml
dv dv
Therefore for the spherical harmonic we have
parity transformation l  ml  ml
Yl , ml ( , )       ( 1) Yl , ml (, )
 ( 1)l Yl , ml (, ) (8.42)
Therefore, the parity of Yl ,ml ( , ) depends only on the value of l and is given
by (1)l . We shall make use of these concepts in the next unit.
SAQ 5
Use the expression given by Eq. (8.40e) for Y2,1(, ) and verify that it is of
even parity.
To sum up, so far we have obtained the solutions of the angular part of the
stationary states of a particle moving in a spherically symmetric potential.
These are nothing but the eigenfunctions of the angular momentum operator
L2 . Their exact functional dependence on  and  is given by spherical
harmonics. The eigenvalues of the operator L2 are l (l  1) 2 where l takes
discrete integral values.
Let us now try to understand what these results mean physically, in terms of
what is called space quantization.
8.3 SPACE QUANTIZATION

You have shown in SAQ 4(a) that the spherical harmonics Yl ,ml ( , ) are
eigenfuctions of L̂z with eighenvalues ml  . Thus we can determine exact
values of L̂2 and L̂z simultaneously. However, since Lˆx , Lˆy , Lˆz cannot be
determined simultaneously according to the uncertainty principle, L̂x and
L̂y will be uncertain. Thus, we are confronted with some surprises about the
130
quantum mechanical angular momentum as compared with its classical
counterpart.
Classically, for the same magnitude of the angular momentum, we can obtain
an infinite number of states by changing the direction of the agular momentum
vector. But quantum mechanically, for each value of angular momentum, there
are only a finite number of states characterised by l and ml . Moreover, in
quantum mechanics, the components of L̂ in two of the three directions being
uncertain, we do not describe a state by specifying the directions of the
angular momentum vector. Instead, we give the component of the angular
momentum along a specific direction. We conveniently choose this direction to
be along the z-axis. So how do we visualise this situation?
There is a useful pictorial way to communicate these quantum mechanical
results – this the so called vector model of the angular momentum. In this
model we represent the angular momentum of the particle in motion is given

by a vector L of length [l (l  1)]1/ 2 . The angular momentum vector precesses

around the z-axis in such a way that the magnitude of L (hence L2 ) and

Lz (projection of L on z-axis) are constants (see Fig. 8.2a).
z
ml  2
6
ml  1
ml  0
Lz 
L
ml  1
ml  2
(b)
(a)

Fig.8.2: a) Precession of L about z-axis; b) space quantization for l  2 . The
radius of the circle is [ 2( 2  1)]1 / 2  6 . The multiplicity of states is 5.
Since for a given value of l , the eigenvalues of L̂z are ml  with integer values

of ml (ranging from  l to l ) , the component of L along the z-axis is
quantized. A measurement of Lz will yield only the 2l  1 quantized values,

with a maximum value l less than the magnitude of the vector L for l  0 .

Further, the vector L can
 make only certain quantized angles with the z-axis;
the angle , between L and Lz can take only discrete values given by
ml
cos   (8.43)
[l (l  1)]1 / 2

This phenomenon of the quantization of the direction of L with respect to one
of the coordinate axes, is known as space quantization. Since ml is always

less than l (l  1) (except for l  0) , the vector L can never be along z-axis. For
l  2, the values of ml are 2, 1, 0, –1 and – 2 as shown in Fig. 8.2b.
131
Furthermore, although L̂x and L̂y are uncertain, Lˆ2x  Lˆ2y being equal to Lˆ2  Lˆ2z
have definite non-zero values unless l  0; however, the values of Lx and
Ly are not quantized. Thus we can visualise the angular momentum vector
sweeping out in all possible direction in the xy plane.
Having analysed the angular part of the wavefunction and some of its
implications, let us now consider the radial part of the eigenfunctions for a
spherically symmetric potential.
8.4 RADIAL EIGENFUNCTIONS

Putting K  l (l  1) in Eq. (8.4), we obtain the differential equation for the radial
function R (r ).
 2 1 d  2 dR (r )    2 l ( l  1) 
l 3   r   V (r )   R(r )  E R(r )
2 r 2 dr  dr   2 r 2 
(8.44a)
l 2
This is a one-dimensional eigenvalue equation for the radial eigenfunction
R(r ) . The actual solution depends upon the form of potential energy function
V (r ) . However, the effective potential energy of the particle is
l 1
 2 l (l  1)
Veff (r )  V (r )  (8.44b)
 2 r 2
r
Thus, there is an extra term in the form of repulsive potential energy
l (l  1)
Fig. 8.3: The variation of (l (l  1) 2 / 2r 2 ) which increases with l (see Fig. 8.3, where we plot
l ( l  1) r2
with r for some
r2 as a function of the radial distance). You can see that this term decreases the
values of l. probability of finding the particle near the centre of force. This term is also
known as the centrifugal potential energy, or the centrifugal barrier.
The origin of the centrifugal term can be understand in the following manner
using classical correspondence. For a particle of mass  moving in a circular
orbit of radius r , classically, there is a centrifugal force directed radially
outward. The magnitude of the force is v 2 / L2 r 3 , where L  vr for a
circular orbit. The potential corresponding to such a force is
L2 / 2r 2 (since F  V / r ) . In quantum mechanics we must replace L2 by its
eigenvalue l (l  1) 2 , hence we obtain the quantum mechanical expression for
the centrifugal potential.
For bound particles (such as a simple harmonic oscillator), the values of
E (eigenvalues) are discrete. Otherwise E varies in a continuous manner. But
whatever be the form of V (r ) , as long as it is spherically symmetric, the
angular part of the eigenfunction of the particle is given by the spherical
harmonics Yl ,ml (, ) .
In the next Unit we shall take V (r ) to be the Coulomb potential energy,

appropriate to a hydrogen atom and shall obtain eigenfunctions and
eigenvalues for the hydrogen atom.
132
8.5 SUMMARY
 In general, the potential in which a particle moves in, in a three-
dimensional space is a function of all three spatial coordinates.
However, a spherically symmetric potential depends only upon the
radial coordinate r and is independent of the polar coordinates  and 
( V (r , , )  V (r ) ).In this unit we have discussed the quantum
mechanical behaviour of a particle having constant total energy and
moving under a three-dimensional, spherically symmetric
potential.

 In classical mechanics, the angular momentum L of such a particle is
a constant of motion. However, in quantum
 mechanics all the three
components Lx , Ly and Lz of the vector L cannot be constants of
motion simultaneously due to the fact the these three components
 do
not commute among themselves. 
However, the magnitude of L or L2
and one of the components of L (which we have taken as Lz ) can be
a constant of motion.
 For a spherically symmetric potential V (r ) the steady state
Schrodinger equation is
  r 2    2 (E  V (r )) r 2  
 r   
 r   2
 1   1 2 
  (sin  )  
 sin    sin 2   2 
Where   (r , , ) .
The angular momentum operators L̂2 and L̂z commute with the
Hamiltonian. Therefore the expectation value of the square of the
angular momentum is a constant of motion for a central potential.
 (r , , ) can be separated in the variables r ,  and  as follows:
(r , , )  R (r ) Y (, )
Where R(r ) is the radial component and Y ( , ) is the angular
component of the wave function.
The stationary state Schrodinger equation reduces to the following two
ODEs:
1   2 R(r )   2 K 
r    (E  V (r ))  R (r )  0
r 2 r 2  r    2 r2
And
1   Y (, )  1  2Y (, )
 sin    K Y (, )
sin      sin 2   2
where K is a constant.
 The solutions Y (, ) of the angular equation are the spherical
harmonics Yl ,ml (, ) which are the eigenfunctions of L̂2 and L̂z :
Lˆ2 Yl ,ml (, )  l ( l  1) 2 Yl ,ml (, ) with l  0,1,2,..... 133
Lˆz Yl ,ml (, )  m l  Yl ,ml (, )
For a given integer l there will be the following (2l  1) values of ml :
 l ,  l  1,  l  2,...,0,..., l  1, l
Hence we can say that the eigenfunctions Yl ,ml (, ) are ( 2l  1) fold
degenerate. The parity of Yl ,ml (, ) depends only on the value of l and
is given by (1)l . Hence it is odd for odd values of l and even for even
values of l.
 For ml  0
1/ 2
 2l  1  (l  ml )! 
Yl ,ml (, )  ( 1)ml    Pl ml (cos ) e iml 
 4  (l  ml )! 
Pl ml (cos ) are the associated Legendre functions and
ml d ml
Pl (cos )  sin   ml
Pl (cos )
dv ml
Where Pl (cos ) are the Legendre functions.
And
Yl ,ml (, )   1ml Yl ,ml (, )*
 In the vector model of the angular momentum  the angular momentum
of the particle in motion is given by a vector L of length [l (l  1)]1 / 2 .

Since L̂z is quantized, the vector L can make only certain quantized

angles with the z-axis; the angle , between L and Lz can take only
discrete
 values. This phenomenon of the quantization of the direction
of L with respect to one of the coordinate axes, is known as space
quantization.
 The effective potential energy of the particle in a spherically
symmetric potential is
 2 (l  1)
Veff (r )  V (r ) 
2 r 2
The repulsive potential energy (l (l  1) 2 / 2r 2 ) which increases with l
decreases the probability of finding the particle near the centre of force
is called the centrifugal barrier.

1. Prove the following commutation relations
i) [Lˆz , xˆ ]  i  yˆ (ii) [Lˆz , pˆ x ]  i  pˆ y (iii) [Lˆz , rˆ 2 ]  0 (iv) [Lˆ2 , Lˆz ]  0
2. Consider a spherically symmetric rigid rotator with for which the classical
L2 
Hamiltonian is H  where L is the angular momentum and I is the
2I
moment of inertia. For the quantum rigid rotator, calculate the energy
eigenfunctions and eigenvalues and the degeneracy of the first three
energy eigenvalues.
3. Use the expressions given by Eq. (8.) to show that

134
2
5

2
Y2,m l ,   
m l  2
4
Note that in general

l 2 ( 2l  1)
 Yl ,ml ( , ) 
4
ml   l
4. Show that the spherical harmonics Y1,0 (, ) , Y1,1(, ) and Y1,1(, ) are
normalized and orthogonal to each other.
1
5. Use the Legendre polynomial P2 ( x )  3 x 2  1 to construct
2
Y2 ,0 (, ) , Y2 ,1 (, ) and Y2 ,2 (, ) .

1. a) We know that Lˆ  rˆ  pˆ  reˆ r  ( i )
  1  1 
In spherical polar coordinates   eˆr  eˆ  eˆ
r r  r sin  
 1  1 
Therefore, Lˆ  ireˆ r  eˆ   eˆ  (eˆ r  eˆ r  0)
 r  sin   
 1  ˆ 1    1 
 ir eˆ   e   i eˆ   eˆ 
 r  r sin      sin   
  1    1  
Hence Lˆ2  Lˆ..Lˆ   2  eˆ   eˆ  .  eˆ   eˆ   (i)
  sin      sin   
        1  
  2  eˆ . eˆ    eˆ . eˆ 
          sin   
(ii)
 1      1   1  
  eˆ . eˆ    eˆ . eˆ 
 sin        sin     sin   
To simplify Eq. (ii), the partial derivatives are carried out first and then
the scalar product is taken. The unit vectors eˆ r , eˆ  , eˆ  are orthonormal
and derivatives of the unit vectors are given as
  
eˆ  eˆr ; eˆ  eˆ cos  ; eˆ   0;
  

eˆ  (eˆr sin   eˆ cos ) (iii)

Now using Eq. (iii) we get
 2  eˆ    2
 ˆ   ˆ  
 e . e   eˆ  .eˆ   
 eˆ  .   (iv)
       2      2
135
 eˆ  1    1  
 ˆ   ˆ
 e . e
1  
 
  eˆ  .    eˆ   
  sin   
    sin       sin  
 eˆ  1 
  eˆ  .  
  sin  
 eˆ  .eˆ  
  1  
  
  sin   


  eˆ.eˆr  sin1    eˆ.eˆ    sin1   
 
0
 1      1 eˆ   1 2 
 eˆ  . eˆ    eˆ  .  eˆ  
 sin        sin    sin   
 1  1 2 
 eˆ  . (eˆ r sin   eˆ  cos )  eˆ  
 sin   sin   
cos  

sin  
 1   1    1 eˆ  1 2 
 eˆ . eˆ   eˆ.   eˆ 
 sin     sin     sin 2   sin 2 2 
 1 1 2 
 eˆ. 
eˆ cos 


 eˆ 
 sin 2  sin 2 2 
1 2

sin 2 2
Substituting in the equation for L̂2 above, we get

 2  1 2 
Lˆ2   2   cot   
  2  sin2   2 
 1     1 2 
  2   sin    
 sin      sin 2   2 
  1  
And Lˆz  eˆ z .Lˆ  (eˆr cos   eˆ  sin ).   i eˆ  i eˆ 
  sin   

 i

b) Now
  2 2 
[Hˆ , Lˆ2 ]    V (r ), L2 
 2 
 2 1   2   Lˆ2 
   r    V (r ), Lˆ2  (Using Eq. 8.6a)
 2 r r  r  2r
2 2

 0 [r ,  and  are independent variables and L̂2 commutes

with itself]
136
Similarly,
 2 1   2   Lˆ2 
[Hˆ , Lˆz ]   r   V (r ), Lˆz   0( [Lˆ2 , Lˆz ]  0)
 2 r r  r  2r
2 2

d dv d d
2. We can write    sin 
d d dv dv
and
d d d d
sin    sin 2   (1  cos 2 )  (1  v 2 )
d dv dv dv
Therefore Eq. (8.17) is:
d  dP (v ) 
 (1  v 2 )   (1  v 2 )   K P (v ) (1  v 2 )  ml2 P (v )
dv  dv 
or (1  v 2 )
d 
dv 
(1  v 2 )
dP (v ) 
dv 
 
 K (1  v 2 )  ml2 P (v )  0
Dividing by (1  v 2 ) , we get
2
d (1  v 2 ) dP (v )   K  ml P (v )  0
  
dv dv   (1  v 2 ) 
d ml
( x )  1  x 2 
ml ml 2
3. a) Pl Pl ( x ) gives us
dx ml
P10 ( x )  1  x 2  P1( x )  x
0
d
P11( x )  1  x 2  P1( x )  1  x 2 
1 2 1 2
dx
b) From SAQ 4a, we know that
P10 (cos )  cos 
P11(cos )  sin 
Using l  1 , ml  0 and Eq. (i) in Eq. (8.36) we get
3 1 / 2
Y1,0 (, )    cos 
 4 
Using l  1 , ml  1 and Eq. (ii) in Eq. (8.36) we get
1/ 2
 3  (0)!  3 1/ 2
Y1,1 (, )      sin  e i    sin  e i
 4  ( 2)!   8 
And using Eq. (8.37) we get
137
3 1/ 2
Y1,1(, )   11Y1,1(, )*    sin  e  i
 8 
4. a) Let Yl , ml (, ) be an eigenfunction of L̂z with the eigen value c, so
LˆzYl ,ml (, )  c Yl ,ml (, )
i 
or  Yl ,ml (, )  c Yl ,ml (, )
2 
L̂z operates only on the  component of Yl ,ml (, ) which is e iml  .

Hence, using Eq. (8. 36 ), we get
i
 iml Yl ,ml (, )  c Yl ,ml (, )
2
 c  ml h / 2  ml 
b) We have to show that

 2
 Y2*,2 (, )Y2,2 (, ) d  1 or   Y2*,2 Y2,2 sin  d d  1

0 0
 2 
15 
Let I1    Y2*,2 Y2,2 sin  d d   
2 sin 4  sin  d
 32 
0 0 0
1
15

16  (1  2  2   2 ) d where   cos 
1
15  4 2  15 (15  10  3)
or I1  2     1
16  3 5 8 15
Hence proved.

Next we have to show that Y2*,2 (, )Y2,2 (   ) d  0 .
 2  2
15
Let I2    Y2*,2 Y2,2 sin  d d 
32   sin 4  e  4i d d
0 0 0 0
2
Now  e  4id  0,  I2  0 .
0
Hence proved.
5. Replacing  by    and  by    we obtain
1/ 2
15
Y2,1(   ,   )    sin (   ) cos (   ) e i (   )
 8 
1/ 2
15
   sin  cos  e  i  Y2,1(, )
 8 
138
Hence the parity of Y2,1(, ) is even.
Terminal Questions
1. i) [Lˆz , xˆ ]  [ xˆ pˆ y  yˆ pˆ x , xˆ ]  [ xˆ pˆ y , xˆ ]  [ yˆ pˆ x , x ]   yˆ [ pˆ x , xˆ ]  i yˆ
ii) [Lˆz , pˆ x ]  [ xˆ pˆ y  yˆ pˆ x , pˆ x ]  [ xˆ pˆ y , pˆ x ]  [ yˆ pˆ x , pˆ x ]  [ xˆ, pˆ x ] pˆ y  i  pˆ y
iii) [Lˆz , rˆ 2 ]  [Lˆz , xˆ 2 ]  [Lˆz , yˆ 2 ]  [Lˆz , zˆ 2 ]
 xˆ [ xˆpˆ y  yˆ pˆ x , xˆ ]  [ xˆ pˆ y  yˆ pˆ x , xˆ ] xˆ
 yˆ [ xˆ pˆ y  yˆ pˆ x , yˆ ]  [ xˆpˆ y  yˆ pˆ x , yˆ ] yˆ
 zˆ[ xˆ pˆ y  yˆ pˆ x , zˆ ]  [ xˆ v y  yˆ pˆ x , zˆ ] zˆ
 ( xˆ yˆ  yˆ xˆ ) i  ( yˆ xˆ  xˆ yˆ ) i   0
iv) [Lˆ2 , Lˆz ]  [Lˆ2x  Lˆ2y  Lˆ2z , Lˆz ]  [Lˆ2x , Lˆ y ]  [L2y , Lˆ z ]
 Lˆ x [Lˆ x , Lˆz ]  [Lˆ x , Lˆz ] Lˆ x  Lˆ y [Lˆ y , Lˆz ]  [Lˆ y , Lˆ z ] Lˆ y
 Lˆ x [ i L y ]  i  Lˆ y Lˆ x  Lˆ y i Lˆ x  i  Lˆ x Lˆ y  0
Lˆ2
2. Let the eigenfunction be (r , , )  R (r ) Y (, ) . Now Hˆ  , the steady
2I
state Schrodinger equation is:
Hˆ (r , , )  E(r , , )
where E is the energy eigenvalue. As you know, the operator L̂2 acts only
on the angular part of the wave function, so
Lˆ2 Lˆ2
Hˆ (r , , )  (r , , )  R(r ) Y (, )
2I 2I
And Eq. (i) reduces to ( with R(r )  constant )
Lˆ2
Y (, )  EY (, )
2I
You already know that the solutions of the ODE Lˆ2 Y (, )  EY (, ) are
the spherical harmonics
Lˆ2 Yl ,ml (, )  l ( l  1) 2 Yl ,ml (, ) with l  0,1,2,.....
Therefore the solutions to Eq. (ii) are
Lˆ2 l ( l  1) 2
Yl ,ml (, )  Yl ,ml (, )  EYl ,ml (, )
2I 2I
139
So the energy eigenvalues are
l (l  1) 2
El  with l  0,1,2,.....
2I
And the energy eigenfunctions are Yl ,ml (, ) with l  0,1,2,..... For each
value of l, (2l  1) values of ml :  l ,  l  1,  l  2,...,0,..., l  1, l .
So while the energy depends only on the value of l, each energy eigen
state is (2l  1) -fold degenerate.
For l  0 , ml  0 , degeneracy is 1.
For l  1 , ml  1, 0,1, degeneracy is 3 .
For l  2 , ml  2,  1, 0,1, 2 , degeneracy is 5 .
2 2
3.  Y2,m;l ( , )
ml  2
15 15 5
2 sin2  cos2   2 sin4   (3 cos2   1)2
8 32 16
5 5
 [6 cos4   6 cos2   4  6 sin2  cos2 ] 
16  4
1

2
You may note that the sum Yl ,m l (, ) is always spherically
m l  1
symmetric.
4. From Eq. (8.38) we know that we have to show
 2 2
  Yl ,ml ( , ) sin dd  1
  0  0
For Y1,0 ( , ) we have
 2  2
2 3
  Y1,0 ( , ) sin dd   cos2  sin d  d
  0  0 4
0 0

3  1 
  cos3  02
4  3 0
3  2
  2  1
4  3 
For Y1,1( , ) we have

 2 2 3
 2
3
  Y 1,1 (  ,  ) sin  d  d  
8   sin  d   d
  0  0 0 0
 1  cos  sin  d   d 
 2
3 2

8
140 0 0

3  1 
   cos   cos3  02
8  3 0
3  4
  2  1
8  3 
For Y1,1( , ) we have
 2 2 3
 2
3
8   d  1
  Y 1,  1 (  ,  ) sin  d  d   sin  d 
  0  0 0 0
For orthogonality of the eigen functions Y1,0 ( , ) and Y1,1( , ) we have to

 2
show:   Y *1,0 ( , )Y1,1( , ) sin dd  0
  0  0
 2  2
3 2 i
  Y *1,0 ( , )Y1,1( , ) sin dd  
4 2 
 cos  sin d   e d
  0  0 0 0
 2
3 1 3  i
  sin   e d  0
4 2  3 0 0
Similarly:
 2 3
 2
2  i
  Y *1,0 ( , )Y1,1( , ) sin dd  
4 2 
 cos  sin  d   e d
  0  0 0 0
 2
1
3 3   i
  3 sin   e d  0
4 2  0 0
 2 3
 2
3  2i
 Y *
 1,1 ( ,  )Y1,1( ,  ) sin  d d   
8   sin d   e d
  0  0 0 0
2
   2i 
3 1 3  e
   cos   cos     2i   0
4 2  3  0   0
d ml
( x )  1  x 2 
ml ml 2
5. Using Pl Pl ( x ) we get
dx ml
1
P20 ( x )  1  x 2  P2 ( x )  3 x 2  1
0
2
P21( x )  1  x 2  P2 ( x )  31  x 2 
d
1 2 1 2
x
dx
2 2 d2
P22 ( x )  1  x 2  P2 ( x )  31  x 2 
dx 2
So,
141
P20 (cos )  3 cos 2   1 ; P21( x )  3 sin  cos  ; P22 ( x )  3 sin 2 

1
2
5 1/ 2 1 1/ 2
Y2 ,0 (, )    3 cos 2   1   5  3 cos 2   1
 4  2  16 
1/ 2
 5  (1)! 
Y2 ,1 (, )      3 sin  cos  e i
 4  (3)! 
1/ 2
15
    sin  cos  e i
 8 
1/ 2 1/ 2
 5  (0)!  15 
Y2 ,2 (, )     3 sin 2  e 2i    sin 2  e 2i
 4  ( 4)!   32 
142
Unit 9 The Hydrogen Atom
UNIT 9
THE HYDROGEN
ATOM
Structure
9.1 Introduction Degeneracy of the Eigenfunctions
Expected Learning Outcomes Radial Probability Density
9.2 Schrödinger Equation for 9.4 Spectra of the Hydrogen Atom
the Hydrogen Atom Quantum Numbers and Constants of
9.3 Solution of the Radial Equation Motion
Asymptotic Properties of the Radial 9.5 Summary
Wave Function 9.6 Terminal Questions
Power Series Solution 9.7 Solutions and Answers
Eigenfunctions
Bound States and Continuum States
9.1 INTRODUCTION
When quantum mechanics was developed in the 1920s, one of its first (and
also one of the most important) applications was to understand hydrogen and
hydrogen like atoms (atoms with one valence electrons). In this unit, our main
focus will be on the hydrogen atom. In the previous Unit we discussed the
motion of a particle in a spherically symmetric potential. We had separated the
Schrödinger equation for a spherically symmetric potential V (r ) into radial and
angular coordinates. As we saw, the spherical harmonics are the
eigenfunctions of the angular part of the Schrödinger equation. The solution of
the radial equation depends on the particular potential function.
In this Unit we extend the ideas developed in the previous unit to the hydrogen
atom. As you know, a hydrogen atom consists of a proton and an electron
moving in the Coulomb potential of the proton. The motion of an electron in
the Coulomb potential of the nucleus is also referred to as the Kepler problem
of quantum mechanics – it is exactly solvable. You know that the Coulomb
e2
potential of the proton at a distance r is V (r )   . Hence the potential is
4  0 r
spherically symmetric.
In Sec. 9.2 we write down the stationary state Schrödinger equation for the
Hydrogen atom which is a two body system comprising a proton and an
electron. The equation is then reduced to two three-dimensional Schrödinger
equations in the centre-of mass and relative coordinates. The general solution
for the stationary state wave function comprises the radial and angular parts.
In Sec. 9.3 we solve the stationary state Schrödinger equation for the radial 143
part of the three-dimensional Schrödinger equation to obtain eigenfucntions
and eigen energies for the stationary states of the hydrogen atom. As for the
simple harmonic oscillator, we rewrite the Schrödinger equation in terms of
dimensionless variables and study the differential equations in its asymptotic
limits to arrive at a trial solution considering the properties of a physically
acceptable wave function. We use the power series method to obtain the
energy eigenfunctions and eigenvalues. We study the degeneracy of the eigen
states and also discuss the continuum states of the hydrogen atom. In
Sec. 9.4 we discuss the Hydrogen atom spectra and the significance of the
quantum numbers.
 reduce the two-body hydrogen atom system to two one-body systems;
 solve the stationary state Schrödinger equation for the radial wave
function for the hydrogen atom;
 obtain the eigenfunctions and energy eigenvalues for the stationary states
of a hydrogen atom;
 explain the spectra of the hydrogen atom; and
 specify the constants of motion and the corresponding quantum numbers
for the hydrogen atom problem.
9.2 SCHRÖDINGER EQUATION FOR THE

HYDROGEN ATOM
Let us consider the hydrogen atom as an example of a three-dimensional
quantum mechanical system. As you know, a hydrogen atom consist of a
proton and an electron. Thus, it is a two-particle system. The Hamiltonian for
two-body motion in a central force field is given as
p12p2
H  2  V (r1, r2 ) (9.1a)
2M 2m
Thus the stationary states of the hydrogen atom are the solutions of the
following time independent Schrödinger equation
 2 2 2 2 e2 
 1  2      (r1, r2 )  ET (r1, r2 )
 2M 2m 4 0 r1 r2 
(9.1b)

where M and m are the masses of the proton and the electron, respectively; r1

and r2 are the position vectors of the proton and the electron, respectively,
with respect to an origin O . ET is the total energy of the system and 0 is the
permittivity constant. Thus we are required to solve a six-dimensional
differential equation to obtain the eigenfunction ( r1, r2 ) and eigenvalue ET .
However, we can reduce the above equation into two three-dimensional
equations in the following manner.

Let R be the coordinate of the centre of mass of the atom. Then
 Mr1  m r2
144
R (9.2)
M m
The separation between the proton and the electron is given by
  
r  r1  r2 (9.3)
   
Solving for r1 and r2 in terms of R and r we get
  m 
r1  R  r (9.4)
M m
  M 
and r2  R  r (9.5)
M m
Now you know that
  X  x
  (9.6)
x1 X x1 x x1
  
where x1, X and x are the x -components of r1, R and r , respectively. Hence
Eqs. (9.2) to (9.4) yield for the x -component
 M  
  (9.7)
x1 m  M X x
Hence, in three-dimensions
 M  
1  R   (9.8)
mM
where
   
  iˆ  jˆ  kˆ (9.9)
x y z
  
and ( x1, y1, z1), ( x, y , z ) and ( X ,Y , Z ) are the components of r1, r and R,
respectively. Similarly, from Eq. (9.5) we obtain
 m  
2  R   (9.10)
mM
Eqs. (9.8) and (9.10) yield
M 2 2  M   
12     R  2  R .   2 (9.11a)
mM  mM 
and
m 2 2  m   
 22    R  2   R .   2 (9.11b)
mM  mM 
Putting the expressions for 12 and  22 from Eqs. (9.11a and b) into Eq. (9.1b)
we get
 2 2 2  1 1  2 e2 
 2(M  m )  R  2  m  M    r  (R, r )  ET (R, r )
   
(9.12)
where for our convenience we have replaced e 2 / 4 0 by only e 2 with
e 2  2.31 10 28 J m. Eq. (9.12) is separable in the coordinates R and r.
Taking
(R, r )  (R ) (r ) (9.13) 145
we find that (R ) and (r ) are the respective solutions of the following three-
dimensional differential equations
2
  2 (R )  E H (R ) (9.14)
2(M  m ) R
2 2 e2
and   ( r )  ( r )  E ( r ) (9.15a)
2 r
1 1 1
where      (9.15b)
m M 
and
ET  E  E H (9.16)
As you know,  is the reduced mass of the system.
Eq. (9.14) shows that a particle of mass (m + M), which is the total mass of the
hydrogen atom, is moving freely in a three-dimensional space and its total
energy is EH (with zero potential energy). This is a problem that you have
already solved in Unit 5. Its eigenfunctions are given by plane waves
 
(R )  e iK .R (9.17)
 2K 2
With  EH (9.18)
2( m  M )
The eigenvalue E H and the corresponding quantum number K vary in
continuous manner.
On the other hand, Eq. (9.15a) describes the motion of a particle of mass
 having potential energy  e 2 / r with respect to a fixed centre. Thus, by the
above procedure we have reduced a two-body system into two one-body
systems, one of mass (m + M) which moves freely in space and other of mass
 and charge e which moves under an attractive potential  e 2 / r . You should
note that in the present model the relative motion of electron and proton with
respect to each other has been replaced by the motion of a particle of mass
 with respect to a fixed centre of force.
Now Eq. (9.15a) is exactly the same as Eq. (8.1) with V (r )  e 2 / r . Hence
the eigenfunctions of the particle of mass  , which are also known as the
eigenfunctions of the hydrogen atom, are given by
(r , , )  R (r )Yl , ml (, ) (9.19)
Substituting K  l (l  1) and V (r )  e 2 / r in Eq. (8.5), the radial function
R(r ) is the solution of the following one-dimensional differential equation
2 1 d  2 dR (r )   e 2  2 l (l  1) 
 r      R (r )  ER ( r )
2 r 2 dr  dr   r 2 r 2 
(9.20)
The above equation has been obtained from Eq. (8.44a) by taking
V (r )  e 2 / r . The effective potential energy in this case is
e 2  2 l (l  1)
Veff (r )    (9.21)
146 r 2 r 2
 2 l (l  1)
Near the origin is much larger than  e 2 / r .
2 r 2
It is possible to get both bound and continuum states for the Coulomb
potential. For E  0 one gets states with continuous values of E, which
describe electron-proton scattering. But with E  0 the states have discrete
values of E, which are the bound states of the hydrogen atom.
9.3 SOLUTION OF THE RADIAL EQUATION

We rewrite Eq. (9.20) as
1 d  2 dR (r )   2e 2 1 l (l  1)  2E
 r    2   R(r )  2 R(r )
2
r dr  dr    r r 2
 
(9.22)
1 d  2 dR( r )   2e 2 1 l (l  1)  2E

Or  r    2   rR(r )  2 rR(r )
r dr  dr    r r 2
 
(9.23)
We define a new variable :
u  rR(r ) (9.24)
and with this Eq. (9.23) reduces to (SAQ 1):
d 2u  2E 2e 2 1 l (l  1) 
    u (9.25)
dr 2   2 2 r r2 
Which can be rewritten as:
 2 d 2u  2e 2  2 1  2 l (l  1) 
  1  u (9.26)
2E dr 2   2 2E r 2E r 2 
2E
Defining a constant k   , which as you can see is real for the bound
2
states because E < 0, we introduce a dimensionless variable   kr and:
du du d du d 2u d 2 u d d 2u 2E d 2u
 k ; k  k2 
dr d dr d dr 2 d 2 dr d 2  2 d 2
(9.27)
Therefore Eq. (9.26) is
d 2u  2e 2 1 l (l  1) 
 1   u (9.28)
d 2   2k  2 
2e 2
We define  0  to write Eq. (9.28) as
 2k
d 2u   0 l (l  1) 
 1   u (9.29)
d 2   2 
SAQ 1
With u  rR(r ) , show that Eq. (9.23) reduces to Eq. (9.25).
147
9.3.1 Asymptotic Properties of the Radial Wave
Function
We now examine the properties of the solutions of Eq. (9.29) for    and
  0.
 For   
d 2u   0 l (l  1)   d 2u
 1    u    u (9.30a)
d 2   2  d 2
And the general solution is:
u  Ae   Be  (9.30b)
Since the solution should not diverge, we must have A  0 and therefore
u  Be  (9.31)
 For   0
d 2u   l (l  1)  0 d 2u l (l  1)
 1  0   u     u (9.30a)
d 2    2  d 2 2
And the general solution is:
u  C l  D l 1 (9.30b)
Eq. (9.30a) does not Since the solution should not diverge as   0 , we must have C  0 (see
hold at ρ=0 , though the margin remark) and therefore
Eq. (9.31) still holds)
u  D l 1 (9.31)
Now that we know the asymptotic behaviour, we can write the general solution
for Eq. (9.29) as
u  e  l 1F () (9.32)
The first factor ensures that the    solution does not diverge and the
second factor ensures that the   0 solution is finite. Also, an
exponential function decays faster than any polynomial and therefore in
the limit    the solution is dominated by the exponential factor.
Next, we try to determine the function F () . Substituting from Eq. (9.32) into
Eq. (9.29) we get the following ODE for F () (SAQ 2):
d 2F dF
  2(l  1)  2   0  2(l  1)F  0 (9.33)
d 2 d
SAQ 2
Show that the ODE for F () is
d 2F dF
  2(l  1)  2   0  2(l  1)F  0
d 2 d
9.3.2 Power Series Solution

We now look for a power series solution for F () of the kind:

F ( )   cmm (9.34a)
m  0,1,2,...
148
The first and second order derivatives are
dF d 2F
 
c m m  m 1 ;
d m  0,1,2..

d 2 m  0,1,2..

c m m (m  1) m  2 (9.34b)
Substituting from Eqs. (9.34 a and b) into Eq. (9.33) we get
 c m  m(m  1)  2(l  1)m m 1

m  0,1,2..
 0  2(l  1)  2m   cmm  0 (9.35)

m  0,1,2..
And
2(l  1)  2m   0 
c m 1  cm (9.36)
(m  1)(m  2l  2)
Which gives us:
2(l  1)   0 
c1  c0 ;
2l  2
2(l  1)  2   0  2(l  1)  2   0 2( l  1)  0 
c2  c1  c0 …
2(2l  3) 2(2l  3)2l  2
So all the coefficients of the power series can be determined in terms of c0 .
Let us now check how the series solution behaves for large values of m.
c 2m 2
lim m   m  1  lim m    lim m   (9.37)
cm m(m  1) m 1
Now if we consider the recursion relation of Eq. (9.37) to be the actual one (in
place of Eq. 9.36) and apply this to the series inEq. (9.34a) we have:
2
c m 1  cm
m 1
22 2 23
 c1  2c 0 ; c 2  c1  2c 0  ;c3  c2  c 0 ... (9.38)
2! 3 3!
2m
In general you will find c m  c 0 and
m!
 
2m m 2m
F ( )  c 0  m !
  c0  m!
c 0 e 2 (9.39)
m  0,1,2,... m  0,1,2,...
So for large values of m, the power series solution for F () reduces to an
exponential function and we would have
u  e   l 1F ()  c 0 e   l 1e 2  c 0 e   l 1 (9.40)
which would diverge for    . Such a solution is clearly unacceptable for a
physical wave function, (remember that this is the solution that was dropped
from the general solution of Eq. 9.30b) since such a wave function is not
normalizable. The only way out is therefore that the power series solution for
F () terminates at some point (as for the simple harmonic oscillator solution
h( ) ). Thus in Eq. (9.36) we must have cm 1  0 for some value of m, which
we denote as mmax :

c m 1  0  2(l  1)  2mmax   0  0  mmax  0  (l  1)
2
(9.41) 149
0
Since l and mmax are integers, we see that must also be an integer. We
2
0
designate this integer value of by n which is called the principal quantum
2
number. So
mmax  n  (l  1)  n  mmax  (l  1) (9.42a)
n is a positive integer and for a given value of l , n ranges from

n  l  1, l  2, l  3,... . The lowest value of mmax and l is 0 ( remember from
Unit 8 that l  0,1,2,... ) so the possible values of n are n  1,2,...
This value of m  mmax is the highest power of  in the power series for
F ( ) :
mmax
F ( )   cmm (9.42b)
m  0,1,2,...
And mmax depends on n and l.
Recollect the definition of  0 :
2e 2 2e 2 2e 2

0   k  (9.43)
 2k  2k  20
2E
with k   . Using  0  2n in Eq. (9.43) we get:
2
2e 2 e 2  1 
k k   (9.44)
 2 ( 2n ) 2  n 
We define the parameter a0 as
2
a0  (9.45)
μ e2
It is interesting to note that a0 is equal to the radius of the first orbit of the
electron in a hydrogen atom for the model proposed by Bohr, provided  is
replaced by the rest mass of the electron m. Since the ratio of m /  is very
close to unity (1.0005), we shall take a0 to be equal to the first Bohr radius
with 0.529  10 10 metre as its value. With this
1
k (9.46a)
a0 n
2E 1  1 
or k2      where n  1,2,... (9.46b)
 2 a02  n 2 
Therefore the energy E of the hydrogen atom can take on discrete values
corresponding to the values of n  1,2,... Designating the eigen energy of the
hydrogen atom corresponding to any value of n by En , we get from Eq.
(9.46b)
150
2  1     e4
En     (9.47)
2 μa02  n 2  2 2 n 2
Under the approximation m /   1 the eigen energy E n is given by
R
En   (9.48a)
n2
me 4
where R (9.48b)
2 2
is the Rydberg constant.
SAQ 3
a) Obtain the value of the Rydberg constant in the units of electron volts and
m 1 .
b) Write down the energy for the ground state and first three excited states of
the hydrogen atom.
For each value of n the solution of the radial equation ( u  e   l 1F () )
designates an eigen state of the system defined by an eigenfunction which we
now write down.
9.3.3 Eigenfunctions
With  0  2n , the radial equation Eq. (9.29) reduces to
d 2u  2n l (l  1) 
 1   u (9.49)
d 2   2 
The solution of Eq. (9.49) ( u  e  l 1F () ) clearly depends on the value of n
and l. For each set of values of n and l the solution of Eq. (9.49) can be
r
written as u n,l () where   kr  . Also from Eq. (9.24) u n,l () can be
a0n
written as
u n,l ()
u n,l ()  rR n,l (r )  R n,l (r )  (9.50)
r
where for a given value of n:
l  n  1, n  2, n  3,...,0 (9.51)
This is in addition to what we had obtained for the spherical harmonics in Unit
8, where we found that the values of l are 0,1,2,3... Eq. (9.51) therefore sets
the upper value for l which is decided by the value of the principle quantum
number n for any given eigen state of the hydrogen atom. So for each value of
n, l  0,1,2,...( n  1) and we can write
e l 1F () e l 1 e l 1 n l 1

mmax
Rn,l (r ) 
r

r
 m c  m

r
 cm m (9.52a)
m 0,1,2,... m 0,1,2,...
r
with   and n  1,2,... and for each value of n , l  0,1,2,...( n  1)
a0n 151
F () is a polynomial of degree mmax  n  (l  1) and the recursion relation for
coefficients of the power series is
cm 1 
2(l  1)  2m  2n  cm (9.52b)
(m  1)(m  2l  2)
All the coefficients in Eq. (9.52b) upto mmax  n  (l  1) are written in terms of
c0 and c0 can be determined by applying the normalization condition
on each function R n,l ( r ) .
Let us write down the first of these eigen functions. For n  1 , the only
r
possible value of l is l  0 . With n  1 ,   and mmax  0 . So the power
a0
series for F () will have just one term and F ()  c 0 . So
e r / a0   r  c 0  r / a 
R1,0 ( r )  c 0   e 0 (9.53)
r  a0  a 0
To determine the value of c 0 we carry out the integration:
 
2
 R1,0 (r )R1,0 (r )r 2 dr  1  c 02 e  2r / a0 r 2 dr  1  c 0 

0 0
a0
And write
e r / a0   r  2
R1,0 (r )  c 0   e  r / a0  (9.54a)
r  a0  a03
The higher order radial eigenfunctions can be found in the same way.
A few more of the lower normalized radial eigenfunctions of the hydrogen

atom are given by
1  r   r / 2a0
R 2,0 (r )  2  e (9.54b)
(2a0 )3 / 2  a0 
1 r
R2,1(r )  e  r / 2a0 (9.54c)
3/2 a0
3 (2a0 )
3/2 
 1  2r 2 r 2   r / 3a0
R3,0 (r )    21   e (9.54d)
 3a0  2
 3a0 27 a0 
3/2
 1  4 2 r  r   r / 3a0
R3,1(r )     1  e (9.54e)
 3a0  9 a0  6a0 
3/2
 1  2 2 r 2  r / 3a0
R3,2 (r )    e (9.54f)
 3a0  27 5 a02
152
In general the radial eigenfunctions of Eq. (9.52a) are given by
l
 r   r 
R n,l (r )  N n l exp      Gn l (r / a0 ) (9.55)
 na 0   a0 
where Gn l (r / a0 ) are the associated Laguerre polynomials and N n l is the

normalisation constant. The normalization constant is determined using the
properties of the associated Laguerre polynomials.
The radial part of some eigenfunctions are shown in Fig. 9.1.
Rn,l (r )
R1,0 (r )
R 2,0 (r )
R3,0 (r )
R2,1(r )
Fig. 9.1: Radial eigenfunctions for the Hydrogen Atom.
SAQ 4
a) Using Eqs. (9.52 a and b) , determine R2,0 (r ) and R2,1(r ) .

b) Normalize the eigenfunctions obtained in SAQ 4(a).
c) Show that R20 (r ) is orthogonal to R10 (r ) .
Finally, the complete eigenfunction of the hydrogen atom is
 nlml (r , , )  R n, l (r )Yl ,ml (, ) (9.58)
where R n, l (r ) and Yl , ml (, ) are given by Eq. (9.46) and Eq. (8.36),
respectively. These eigenfucntions form an orthonormal set, i.e.,
  n* l m (r )  n l m (r ) r 2dr sin  d d   nn  ll   m m

l l l l
(9.59a)
where  jj   1 for j  j  and zero otherwise.

Also: Hˆ nlml E  n nlml (9.59b)
Lˆ2 nlml  l ( l  1) 2 nlml
(9.59c)
Lˆz  nlml  ml  nlml
(9.59d)
153
SAQ 5
a) List the eigenfunctions of the hydrogen atom for n  1, 2 .
b) Using Eqs. (9.54), (9.58) and (8.40) write down the eigenfunctions
100 ,  200 ,  211 .
100 is the eigenfunction for the lowest energy state or ground-state of the
hydrogen atom. A few hydrogen atom wave functions are given in the table
below:
Table 9.1: Hydrogen Atom Wave Functions
1 1
100  e  r / a0  200  2  r / a0 e r / 2a0
a03 / 2 3
4 2a0 / 2
1 1
 210  r / a0 e  r / 2a0 cos   211  r / a0 e  r / 2a0 sin e  i
4 2a03 / 2 8 a03 / 2
 r2  2
 300 
1
 27  18
r
 2  e  r / 3 a0  310  r / a0 6  r / a0 e r / 3a0 cos 
81 3a03 / 2  a0 a02  a03 / 2
 311 
1
3 / 2
r / a0 6  r / a0 e r / 3a0 sin e  i  320  1
3 / 2
r / a0 2 e r / 3a 3 cos2   1  
81 a0 81 6a0
1 1
 32 1  r / a0 2 e  r / 3a0 sin  cos e  i  322  r / a0 2 e  r / 3a0 sin2 e  2i
81 a03 / 2 162 a03 / 2
9.3.4 Bound States and Continuum States

We can now make some observations about the Coulomb potential energy
problem in relation to the hydrogen atom. Eq. (9.46) tells us that the bound
state eigenfunctions (E < 0) for the Coulomb potential go to zero as r goes to
infinity. Notice that this potential gives an infinite number of bound states
starting at energy  e 4 / 2 2 and ending at 0. The eigenenergy given by
Eq. (9.44) varies in a discrete manner. The difference between the energies of
the two consecutive energy states decreases as n increases (see Fig. 9.2).
For large n, the energy difference becomes quite small. The states with high n
are called Rydberg states. Finally at n   , the eigenenergy becomes zero
154 and the hydrogen atom is ionized into a proton and an electron with zero total
energy. The eigen states with E > 0 are continuum states. They are shown by
the shaded portion of Fig. 9.2. The eigenfunctions of such states do not go to
zero as r goes to infinity and E varies in a continuous manner. The
eigenfunctions of continuum states of a hydrogen atom are coulomb waves.
The atomic electrons having l  0,1, 2, 3,... are known as s, p, d , f ,... electrons,
respectively. It is evident from Eqs. (9.55) and (9.58) that only for s electrons,
for which l  0, the eigenfunction  nlm l is finite at r  0 , which is practically
the position of the nucleus. Hence only s electrons have a finite probability of
their existence at the nucleus whereas, for the electrons having non-zero
angular momentum (l  0) the probability is zero. Such a behavior can be
understood from Eq. (9.23). The centrifugal potential energy
( 2 / 2) l ( l  1) / r 2 ) for l  0 does not allow p, d , f ,... electrons to come very
close to the nucleus n  1 state is known as the ground state of the hydrogen
atom while n   corresponds to its lowest ionised state. Thus we require one
Rydberg energy to ionise a hydrogen atom.
E n in eV
Continuum States
0
 0.85 n4
 1 .5 n3
 3 .4 n2
 13.6 n 1
l 0 l 1 l 2 l 3
Fig. 9.2: Eigen Energies of the Hydrogen Atom.
9.3.5 Degeneracy of the Eigenfunctions

There is another interesting feature worth commenting upon: we have a
degeneracy in the spectrum, the l  degeneracy. The energy does not depend
on l . but only on n; yet for a fixed n, possible l values are l  0,1, 2,..., n  1. In
addition to the l  degeneracy, there is also the ml  degeneracy, the result of
spherical symmetry. For each l, ml goes from  l to  l degenerate levels. For
any n, the total degeneracy, then is
n 1
 (2l  1)  n 2 (9.60)
l 0
And if we take into account the two-valuedness called spin, which you will
study in Unit 14, the total degeneracy is 2n 2 .
155
9.3.6 Radial Probability Density
The probability of finding an electron in the  nlml state in a certain

region of space dV inside the hydrogen atom is
2
Probability   nlml dV . (9.61a)
The volume element dV  r 2 sin drdd And we know that when this
probability is integrated over all space, the value is 1:
  2 2
    nlm l r 2 sin drdd  1 (9.61b)
r  0  0  0
2
P nlml (r , , )   nlml
is the probability density or the probability per unit
volume. In general (Eq. 8.36), the wave function  nlml is
 nlml (r , , )  R n, l (r )Yl ,ml (, )  CRn, l (r ) Pl ml (cos ) e iml 

(9.62)
Where C is a constant. When the electron is in the l  0 state (zero angular

momentum) the wave function is  n 00  R n, 0 ( r )Y0,0 (, ), the wave function
has no dependence on  or  and is spherically symmetric. In such
2
cases, the probability density P n 00 (r , , )   n 00 depends only on the
radial distance and not on  or . The probability of finding an electron
between r and r  dr , when the electron is in the state  n 00 , which is the
probability of finding the electron within a thin spherical shell of volume
4r 2 dr can be written as:
Pn 00 (r )dr   n 00 2 4r 2 dr (9.63)
Pn 00 (r ) is called the radial probability density function or radial

probability distribution function which is the probability per unit length.
For example the radial probability density for the 100 state is:
2
 1  4r 2  2r / a0 
P100 (r )dr  4r 2  n 00 2
 r 2  e  r / a0    e (9.64)
 a 3  a03
 0 
In TQ 3 you will calculate the most probable location of the electron in the
100 state and see that it is equal to one Bohr radius.
The radial probability distribution functions for a few zero orbital angular
momentum states is shown in the figure below.
156
P100
P200 P300
Fig. 9.3: Radial probability distributions for the 100 ,  200 and  300 states.
SAQ 6
Calculate the probability of finding the electron within a distance r  a0 and

r  2a0 from the proton in the ground state 100 of the hydrogen atom.
The discussion so far helps us to beautifully explain the spectra of the

hydrogen atom.
9.4 SPECTRA OF THE HYDROGEN ATOM

When the electron in a hydrogen atom makes a transition from its excited
states (n > 1) to a lower excited state or to the ground state (n = 1), it emits
electromagnetic radiations of characteristic frequencies of the hydrogen atom.
The energy difference between two eigenstates of the hydrogen atom is given
by
 1 1 
E  R   (9.61)
 n2 n2 
 2 1 
If we take R in the units of m1 wave number of emitted radiation is given by
 1 1 
  1.097  107  2  2  m 1 (9.62)
n 
 2 n1 
For n 2  1 and n1  2, 3, 4,... we get a series of electromagnetic radiation of

different wavelengths. This series is known as Lyman series and lies in the
ultraviolet region of the electromagnetic spectrum. The Balmer series
correspond to the transitions from n1  3, 4, 5,... to n2  2 . Similarly, Paschen
and Brackett series are produced due to transitions from n1  4, 5,6,... to
n 2  3 and n1  5, 6, 7,... to n 2  4 respectively. Thus, the theoretically
obtained eigenenergy spectrum successfully explains the observed line
spectra of the hydrogen atom.
157
However, when we compare the energy spectrum with very accurate
experimental data, we find some discrepancies. This is because in the real
hydrogen atom there are other interaction that we have neglected here. We
will now discuss an important concept regarding such systems – the concept
of quantum numbers that are constants of motion and characterise the state of
such a system.
9.4.1 Quantum Numbers and Constants of Motion

In the study of the hydrogen atom you have come across three integers
namely n, l and ml . These integers are known as quantum numbers. Since n
is connected with the eigenenergy of the system (see Eq. 9.48) it is known as
energy quantum number of principal quantum number. Its existence is due
to the fact that that the energy is a constant of motion, i.e., the states are
stationary states. The energy quantum number exists for continuum states
also with the difference that now it varies in a continuous manner. In the
continuum, it is usually denoted by k .
Due
 to spherically symmetric potential the angular momentum of the object
L becomes a constant of motion. However, due
 to non-commutability of
Lx , Ly and Lz the angular momentum vector L is not a constant of motion in
quantum mechanics but as you have seen, L2 is a constant of motion. This
gives rise to another quantum number l , which varies in a discrete manner and
is a positive integer. Since l is connected with the orbital motion of the object
it is called orbital quantum number or the azimuthal quantum number.
From Eq. (9.45) we can show that l is less than n and for a given value of n it
takes the following values:
0,1, 2,..., n  1 (9.63)
Since a spherically
 symmetric potential is also axially symmetric, the
z component of L also becomes a constant of motion and gives rise to a third
quantum number ml . It can take negative as well as positive number n .
However, it can be shown that if the atom is placed under a magnetic field, its
energy depends upon ml . Hence ml is known as magnetic quantum number.
For a given value of l , the permissible values of ml are
 l ,  l  1,...., 0,1, 2,..., l  1, l (9.64)
The existence of three quantum numbers is also a consequence of the fact

that the time independent Schrödinger equation contains three independent
variables r,  and . We have one quantum number for each space coordinate.
SAQ 7
Show that for n  3 there are 9 degenerate eigenfunctions for a hydrogen

atom.
Let us now summarise what you have studied in this unit.
158
9.5 SUMMARY
 A hydrogen atom in a two-particle system consisting of a proton(of
mass M) and an electron(of mass m) and its stationary state
Schrödinger equation is a six-dimensional differential equation.
However, it can be separated into two three-dimensional differential
equations:
-One corresponding to the motion of a free particle having mass
( m  M ) . Its solutions are plane waves e iKR , where the energy
quantum number K varies in a continuous manner and is related to
E H by Eq. (9.18b).
-The second three-dimensional differential equation describes the
motion of a particle of mass  (reduced mass of the system) having a
charge  e in a spherically symmetric potential (Coulomb potential) due
to a fixed center of force having charge  e . This differential equation
again separates into three one-dimensional differential equations, one
for each spherical polar coordinate r ,  and  . These degrees of
freedom given rise to three quantum numbers n, l and ml .
 The nature of the radial wave function R(r ) depends upon whether the
state is a bound state or a continuum state. For bound states the
 2 1 / 2 r 
eigenfunction varies as exp   E  at large value of r. Hence
2 
   a0 
the probability of finding the particle goes to zero as r increases to  .
On the other hand, for continuum states the probability remains finite
even as r goes to infinity. At small and intermediate values of
r , different spherically symmetric potentials give rise to different radial
functions. In this unit we have considered Coulomb potential,
appropriate to a hydrogen atom.
 The angular eigenfunctions of the hydrogen atom are again spherical
harmonics and the radial eigenfunctions for bound states are given in
terms of the associated Laguerre polynomials. The quantum number n
takes only positive integer values excluding zero. For a given value of
n, the angular momentum quantum number l takes n values given
by
l  0,1, ...., n  1
For each value of l the magnetic quantum number ml takes the
values
ml  1,  l  1,.... 0, l , ...l  1, l
 The energy E corresponding to different eigenfunctions

 nlml (r , , ) depends only upon the principal quantum number n .
Hence for a given n there are n 2 degenerate eigenfucions
corresponding to different permissible values of l and ml . For states
with E  0 , the radial eigenfunctions are Coulomb waves and the
energy quantum number K varies in a continuous manner. Now energy
states are no longer discrete and we talk in terms of energy states per
unit energy range. 159
 The quantum mechanical treatment of the hydrogen atom explains the
production of various electromagnetic series experimentally observed
in the spectrum of the atom.

1. According to the virial theorem the average value of the potential energy of
a particle subjected to a Coulomb potential in any stationary bound state is
two times its total energy. Verify the above theorem for the ground state of
the hydrogen atom. Further show that the average kinetic energy is equal
to the magnitude to total energy.
2. Use the uncertainty relation to show that the dimension of the most stable
ground state hydrogen atom is of the order of the first Bohr radius a0 .
3. Obtain the most probable value and expectation value of r for the ground
state of a hydrogen atom.
4. Consider an initial state of the hydrogen atom:

(r ,0) 
1
10
 3100   200  2 210  2 211 
Determine a) the expectation value of the energy at t = 0 and b)the wave

function (r , t ) .
5. Calculate r 2 for the ground state of the hydrogen atom.
6. Determine the radial eigenfunctions and energy eigenvalues for the

case l  0 , for a particle confined in the infinite spherical well:
 0 r  R0
V (r )  
 r  R0
7. The wave function for the electron in the hydrogen atom is
 1

(r )  3 200  7 211  2 211  4 311
6

Calculate the expectation value for L̂2 and L̂z .
8. Consider an electron in the state


(r )  N 100  2 2  210  4i  21  1 
Determine the normalization constant C and the expectation values of
the energy, of L̂2 and Lˆz .
9. Deuterium is a stable isotope of Hydrogen with a proton and a
neutron in the nucleus. Given that the reduced mass of a deuterium
atom is 9.106909  10 31kg, calculate the ground state energy and
using Eq. (9.47).

du d dR dR du
1.  rR(r )  r R  r  R (i)
160 dr dr dr dr dr
dR du
 r2 r  rR (ii)
dr dr
d r 2 dR   d r du   d rR   r d 2u  du  du  r d 2u
So  dr  dr  dr  dr
dr dr 2 dr dr dr 2
(iii)
With Eq. (iii), and u  rR(r ) , Eq. (9.23) is:
1  d 2u   2e 2 1 l (l  1)  2E
 r     u 2 u (iv)
2
r  dr    2 r r 2
 
d 2u  2E 2e 2 1 l (l  1) 
or     u (v)
dr 2   2 2 r r2 
2. Using u  e  l 1F () we can write
du dF
 e    l 1F ()  (l  1)e    l F ()  e    l 1 (i)
d d
d 2u dF
 e   l 1F  (l  1)e   l F  e   l 1
d 2 d
dF
 (l  1)e   l F  l (l  1)e   l 1F  (l  1)e  l
d
dF dF d 2F
 e    l 1  (l  1)e l  e    l 1
d d d 2
 e   l   2(l  1)  l (l  1) 1 F
dF  d 2F 
 e   l 2(l  1)  2  e    l   (ii)
d  d 2 
Substituting in Eq. (9.29) we get:
e   l   2(l  1)  l (l  1) 1 F
dF  d 2F 
 e   l 2(l  1)  2  e    l  
d  d 2 
  l ( l  1) 
 e    l  1 1  0   F
   2 
d 2F dF
or   2(l  1)  2   0  2(l  1)F  0
d 2 d
me 4
3. a) R   2.18  10 18 Joules  13.6 eV ( e 2  2.31 10 28 J m)
2 2
me 4 1
also in units of m1, R  2 ch
 1.10  107 m1 an accurate value of
2
the Rydberg constant is
R  1.09737373  10 7 m 1 .
b) Using the results of SAQ 3a and Eq. (9.48a), we get the ground state
energy to be 161
R
The ground state energy E1    13.6 eV
12
R
The first excited state energy E 2    3.4 eV
22
R
The second excited state energy E 3    1.5 eV
32
R
The third excited state energy E 4    0.85 eV
42
e    l  1 n  l 1 m
4. a) We use Rn,l (r ) 
r
 c m  (Eq. 9.52a) with n  2 , l  0 ,
m  0,1,2,...
r r
n  l  1  2  0  1  1 and    to write
a0 n 2a0
e r / 2a0   r 
1
e r / 2a0    r 
R 2,0 (r ) 
r

 2a0
  cm m 
m  0. 2a0 c 0  c1 2a
  0


(i)
Using Eq. (9.52b) we get n  2 , l  0 , m  0
2  4
c1  c o  c1  c o (ii)
2
Using Eq. (ii) in Eq. (i) we get
e r / 2a0    r  e r / 2a0   r 
R 2,0 (r )  c 0  c 0  2a   c 0 2  
2a0   0  4 a 0  a 0
e    l  1 n  l 1 m
Using Rn,l (r ) 
r
 c m  (Eq. 9.52a) with n  2 , l  1 ,
m  0,1,2,...
r r
n  l  1  2  1  1  0 and    we get
a0 n 2a0
2
e  r / 2a0   r  0 re  r / 2a0 
R 2,1(r ) 
r
 
 2a0  m  0.

c m  m c 0
4a02
(iv)
b) For R 2,0 (r ) we determine c 0 using the normalization condition


c 02   r  2
2
 R 22,0 (r )r 2 dr  1  e  r / a0   2 
  r dr  1
2
16a0  a0 
0 0
To evaluate the
integrals in this unit, So
you may Table 3.1 of
Unit 3 , Block 1. c 02   r4 r3 
e  r / a0   4r 2 
 4 dr  1
16a02  a02 a00 
0  
c 02
 8a03   1 or c 0 
2
(i)
16a02 a0
So the normalized eigenfunction R 2,0 (r ) is

162
2 e r / 2a0   r  e r / 2a0   r 

R 2,0 (r )  2   2   (ii)
a0 4a0  a0  2a0  3 / 2  a0 
For R2,1(r ) we determine c0 using the normalization condition


c 02 
 R 22,1(r )r 2 dr  1 
16a04  e  r / a r 4dr  1
0
0 0
c 02
So 24a05   1 or c0 
2
(iii)
16a04 3a0
 2  re  r / 2a0  1 re  r / 2a0 
So R 2,1( r )     (iv)
 3a  4a02 2a0 3 / 2 3a0
 0 
 
1  r 
c)  * (r )
R 20 R10 (r ) r 2 dr 

2a06
1/ 2
    2  a0  r 2e  (3r / 2a0 ) dr
0 0
 
 
1  2! 1 3! 
 2  0
2 a03   3 
3 a0  3  4 
    
  2a0   2a0  

5. a) For n  1, l  0 and ml  0 so the eigenfunction is 100
For n  2 and l  0,1. For l  0 , ml  0 . For l  1 , ml  1,0,1. So
the wave functions are
 200 ,  211,  210 and  211 .
2  1 1 / 2 
b) 100  R1, 0Y0,0  e  r / a0    
a03  4  
1
 e  r / a0 
a03
 1  r   r / 2a0   1 1 / 2 
 200  R 2, 0Y0,0   2  e    
 4  
3/2  a
 ( 2a 0 ) 0 
1  r   r / 2a0
 2  e
32a03  a0 
 1 r    3 1/ 2 
 211  R2,1Y1,1   e  r / 2a0     sin  e i 
  8 
( 2a )3/2 a 3
 0 0  
1/ 2
 1  r  r / 2a0
   e sin  e i
 64a 3  a0
 0 
6. The probability of the electron being within a distance r from the proton in
the 100 state is
163
r r
4

P  P100 (r )dr 
a03  r 2e  2r / a dr
0
0 0
4 a3 a a2 a3 
  0  0 r 2 e  2r / a0   0 re  2r / a0   0 e  2r / a0  
a03  4 2 2 4 
  r  r2 
 1  e  2r / a0  1  2   2  (i)
  a0   a02 

The probability of finding the electron within a distance r  a0 is
obtained by substituting r  a0 in Eq. (i) :
P (r  a0 )  1  e 2 1  2  2  1  5e 2  .323
The probability of finding the electron within a distance r  2a0 is

obtained by substituting r  2a0 in Eq. (i) :
P (r  2a0 )  1  e 4 1  4  8  1  13e 4  .761
7. Since n  3, the permissible values of l are 0, 1 and 2.

For l  0 , the permissible value of ml is ml  0
For l  1 , the permissible values of ml are ml  1, 0,1
For l  2 , the permissible values of ml are ml  2,  1, 0,1, 2
Hence, there are nine permissible values of ml . Each combination of

n, l and ml gives rise to an eigenfunction. However the energy depends
only upon n . Hence all these nine eigenfunctions are degenerate.
Terminal Questions
e2
1. Since V (r )   ,
r
 e2  2
V (r )   100 ( r ) 2  
 r
 r dr sin  d d


1

a03
4 ( e 2 )  e  2r / a0 r dr
0
4e 2 e2  2 4

1
   2 E1
 a0   and E1   e 
a03  2  2 a0  e 2 2 2 

 
 a0 
Since kinetic energy + potential energy = total energy
e2 e2 e2
K .E.  E1  V (r )      E1
2a0 a0 2a0
2. Let the size of the atom be R0 . since the electron is inside the atom, the
uncertainty in the momentum is p   / R0 . The linear momentum of
magnitude p can be in any direction so its components can have values
164
from  p to p . Hence the uncertainty in momentum is also approximately
p . Hence we take p  p . Now we take
K .E. 
p 2 
2
 K .E.
2 2 R02
2 e2
E  K .E .  V   at R0 .
2  R0 2 R0
dE 2 e2
Now    0 for a stable atom.
dR0  R03 R02
Hence
2
R0   a0 .
 e2
Hence the size of the most stable atom is the first Bohr radius itself.
3. The probability of finding the electron between r and r + dr is given by
1 4 2  2r / a 0
100 (r ) r 2 4dr  e  2r / a0 r 2 4dr  r e dr .
a03 a03
Hence to determine the most probable value of r we differentiate

r 2 e 2r / a0 with respect to r and equate the result to zero. Thus we get
 2 2  
r    2r  e  2r / a0  0 or r  a0
  a0  
However the average value of r is given by
r   100 (r ) 2 r r 2 dr sin  d d

1

4 e  2r / a0 r dr
3

3
a0
0
4 6 3
  a0 .
a03  2 
4 2
 
 a0 
 
4. a) 
E  Hˆ   * (r ,0)Hˆ (r ,0)d 3 r

Hˆ (r ,0) 
1
Hˆ  3 100   200  2 210  2 211 
10

1
 3E1100  E 2 200  2E 2  2E 2 211 
10
Also

 * (r ,0) 
1
 *
3100   *200  2 *210  2 *211 
10
Using the property of orthogonality of the eigenfunctions (Eq. 9.58) and the
value of the eigen energies (Eq. 9.49a) we get
E
1
10 
 3100
*
  *200  2 *210  2 *211 
 3E1100  E2 200  2E 2  2E2 211 d 3r  165
1 1
 3E1  E2  4E 2  2E 2   3E1  7E2 
10 10
R  7
   3    6.46 eV
10  4
iHˆ t
  
b) (r , t )  e  (r ,0)
iHˆ t
e

  1
 10  
3100   200  2 210  2 211 

iE t iE t
3  1  2  1 1 2 
 e  100  e    200   210   211 
10  10 10 10 
1
5. For the ground state 100  e  r / a0  . Therefore
a03
  2
r2     r 2 100
2 
r 2 sin drdd 
r  0  0  0
1   2
  
a03 r  0   0   0
 
 r 4e  2r / a0  sin  dr d d
4  4  2r / a0 
 r e dr  3a02
a03 r  0
6. The wave function vanishes for r  0 . The radial equation for r  R 0

is (using u  rR(r ) and V (r )  0 in Eq. 9.25) is
d 2u  2mE l (l  1) 
   u
dr 2   2 r 2 
For l  0 the equation is:
d 2u 2mE
  u
dr 2 2
With the solution: u(r )  A cos( kr )  B sin( kr )
2mE
Where k  . Therefore
2
u(r ) cos( kr ) sin( kr )

R( r )  A B
r r r
cos( kr )
Since the function diverges as r  0 , we must set A  0 so
r
that R(r ) is a square integrable function. So the solution is
sin( kr )
R( r )  B
r
The boundary condition is R(R0 )  0

166
Therefore
sin( kR0 ) n
 0  kR0  n  k  with n  1, 2, 3...
R0 R0
So the energy eigenvalues are:

2
 2  n 
En    with n  1, 2, 3...
2m  R 0 
And the eigenfunctions are
 nr 
sin 
 R0 
R( r )  N n
r
where N n is the normalization constant.
7. The expectation values are:

   

Lˆ2   * (r )Lˆ2 (r )d 3 r
and
Lˆz    * (r )Lˆz (r )d 3r
From Eq. (9.59 c and d)) we get

 1
6
 
Lˆ2(r )  Lˆ2  3 200  7 211  2 211  4 311 



1
6

3(0) 200  7 ( 2 2 ) 211  2( 2 2 ) 211  4(2 2 ) 311 
 1
6
 
Lˆz (r )  Lˆz  3 200  7 211  2 211  4 311 



1
6

3(0) 200  7 ( ) 211  2(  ) 211  4( ) 311 


6
 7 211  2 211  4 311 
Also
 1

 * (r )  3 *200  7 *211  2 *211  4 311
6
* 
So L̂2 
1
36
 3
7(2 2 )  4(2 2 )  16(2 2 )   2
2

And
L̂z 

7  4  16  19 
36 36
8. To evaluate the normalization constant N we use the condition :

  3
  * (r )(r )d r  1
Using Eq. (9.59a), Eq. (ii) reduces to

N 2   *100 100d 3r  8   *210  210d 3 r  16   *211 211d 3 r  1 
167
1
 N 2 1  8  16   1 or N 
5

So (r ) 
1

5 100

 2 2  210  4i  21  1 
 1

And  * (r )   *100  2 2  * 210  4i  * 21  1
5

 1
5
 
Hˆ (r )  Hˆ  100  2 2  210  4i  21  1 


1

 E1100  2 2E 2  210  4i E 2 21  1
5

1
And Ĥ  E1  8E2  16E2   1 E1  24E2    7 R
25 25 25
L̂2 
2
25

(0)  8(2)  16(22 ) 
48 2
25
 
And
L̂z 

0  0  16   16 
25 25
9. With n  1 ,   9.106909×10 31 kg and e 2  2.31 10 28 Jm in

Eq. (9.47) we get that the ground state energy of deuterium is
9.106909×10 31
E1   2.31 10  28 2 J

2 1.054  10  342
 21.87  10 19 J  13.67 eV
168
Unit 8 Observables and Operators
FURTHER READINGS
1. Introduction to Quantum Mechanics, David J. Griffiths, 2nd Edition,

Pearson Prentice Hall (2004).
2. Quantum Mechanics Theory and Applications Ajoy Ghatak and S.

Lokanathan, 2nd Edition, Macmillan Publishers India Limited (2004).
3. Principles of Quantum Mechanics, R. Shankar, 2nd Edition,Springer

(1994).
4. Modern Quantum Mechanics, J. J. Sakurai and Jim Napolitano, 3rd

Edition, Cambridge University Press (2021)
169
Block 2 TABLE OF PHYSICAL CONSTANTS
Applications of Quantum Mechanics
Symbol Quantity Value

c Speed of light in vacuum 2.998  108 ms1
0 Permeability of free space 1.257  10 6 N A 2
0 Permittivity of free space 8.854  10 12 C 2 N1 m 2
1/40 8.988  109 Nm2C2

e Charge of the proton 1.602  10 19 C
e Charge of the electron 1.602  10 19 C
h Planck’s constant 6.626  10 34 Js
 h / 2 1.054  10 34 Js
me Electron rest mass 9.109  10 31 kg
 e/me Electron charge to mass ratio 1.759  10 19 C kg1
mp Proton rest mass 1.673  10 27 kg
mn Neutron rest mass 1.675  10 27 kg
a0 Bohr radius 5.292  10 11 m
NA Avogadro constant 6.022  10 23 mol 1
R Universal gas constant 8.315 J mol 1K 1
kB Boltzmann constant 1.381  10 23 J K 1

G Universal gravitational constant 6.673  10 11 N m 2kg 2
170

Block 2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Block 2

Uploaded by

Copyright:

Available Formats

MPH-004

Course Design Committee

Block Preparation Team

Course Coordinators: Dr. Subhalakshmi Lamba, Dr. M. Boazbou Newmai

Further Readings 169

In Unit 6 entitled “Simple One-Dimensional Potentials-II”, we solve the Schrödinger

In Unit 8 we solve the three-dimensional Schrödinger equation in spherical polar coordinates

We wish you all success!

Expected Learning Outcomes

5.2 A FREE PARTICLE

The eigenvalues of Ĥ can form a discrete spectrum or have continuous

1 ( x )  e ikx ;  2 ( x )  e ikx (5.9)

For the solution (x ) to be physically acceptable, k cannot have an imaginary

To understand, let us take k to be complex: k    i . In that case,

1 ( x )  e i (   i ) x  e ix e x (5.11)

Since any non-negative value of E is allowed, the energy spectrum of a free

 1 ( x )1 ( x )dx   e e dx   dx

If the quantum mechanical free particle is in a stationary state (x ) which

( x, t )  Ae i kx t   Be  i kx  t  (5.16)

which is independent of time t as well as position x. In other words, at any

Show that 1 ( x )  e ikx and  2 ( x )  e  ikx are eigenfunctions of the momentum

Using 1 ( x )  e ikx in Eq. (5.35) we get:

So pˆ x 1x   k 1x  . So 1 ( x ) is an eigenfunction of the momentum

Using  2 ( x )  e ikx in Eq. (5.35) we get:

 i   ike ikx   ke ikx

Since pˆ x  2 x    k  2 x  , we can say that  2 ( x ) is also an eigenfunction of

In the classical picture, a free particle has “unbounded motion” because it is

5.3 PARTICLE IN A BOX

On substituting the second order derivative in Eq. (5.31), we get:

Fig: 5.5: The first three eigenfunctions for a particle in a box.

So how do we reconcile the quantum mechanical picture with our experiences

The minimum energy is the zero point energy:

8me L2 8me L2 89.109  10  31 kg 1.0  10 10 m2

5.3.3 Time Dependent Wave Function

To understand, let us work through the following example.

Calculate the probability of finding the particle between x  L / 4 and

5.4 QUANTUM DOT

A real quantum dot is definitely more complicated than what we have

 Quantum dots (sometimes called artificial atoms) are typically semiconductor

We now write the time independent Schrödinger equation for a particle in a

5.6 ENERGY (E) OF THE PARTICLE IS LESS

Region I:    x  0 for which V  0

1 ( x )  A e ik1x  B e  ik1x (5.50)

So, the solution of Eq. (5.47a) is a linear combination of wave functions

Region II:  0  x    for which V  V0

so that we can write Eq. (5.51b) as:

The solution of Eq. (5.51d) is:

In Eq. (5.52), the term Ce x represents an exponentially decreasing wave

1 ( x )  A e ik1x  B e  ik1x for Region I: x < 0 (5.50)

 2 ( x )  Ce  x for Region II: x > 0 (5.52)

The next step, as you know, is to apply the boundary conditions at x  0 to

 Aik1  Bik1  C

Subtracting Eq. (5.53b) from Eq. (5.53a), we get:

Equating the LHS of Eqs. (5.54a and 5.54b) we can write:

Thus, we have obtained the constants B and C in terms of A. Let us now

5.6.1 Reflection and Transmission Coefficients

Solution : We use Eq. (5.56) to determine Jr .

 d Be  ik1x  d Be  ik1x 

Determine the probability current density (Ji ) of the incident wave.

The reflection coefficient (R) is the probability that an incident particle is

Since the incident wave is either reflected or transmitted without any

3. The wave function penetrates into classically forbidden region, x > 0.

5.7 ENERGY (E) OF THE PARTICLE IS GREATER