This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: [Link]
This content was downloaded on 10/12/2023 at 12:26

Please note that terms and conditions apply.

You may also like:

INVERSE PROBLEMS NEWSLETTER

Kinetic Theory of Granular

Gases Emmanuel Trizac

INVERSE PROBLEMS NEWSLETTER

A Positron named Priscilla: Scientific Discovery at the Frontier / Reinventing

the Future: Conversations with the World's Leading Scientists
Jane Gregory, PUS Research Group, Science Museum, London, UK
 IOP Publishing

Quantum Mechanics
Mohammad Saleem

Chapter 1
The failure of classical physics and the advent
of quantum mechanics

Quantum mechanics has played a significant role in the development of various

disciplines of physics. It was propounded in 1925 and has reigned supreme ever
since, extending its domain over the years; of course, in its own domain it has
always been in excellent agreement with experiments. It has long since become the
language of physics and anyone who tries to understand the basic principles of
physics without having a grasp of this subject is doomed to fade away in the
darkness of ignorance. In this book an attempt has been made to provide a logical,
lucid and user-friendly treatment of this elegant and fascinating subject.

1.1 A challenge for classical physics

Looking through the lattice of history, we observe that the first quarter of the
twentieth century was a challenging period for classical physics. The interference,
diffraction and polarisation phenomena could only be explained by assuming
that light had a wave nature. But some other phenomena, such as black body
radiation, the photoelectric effect and Compton scattering, defied the wave concept
of electromagnetic radiation. The black body radiation spectrum was explained by
Planck who assumed that the atoms of the walls of the black body act as
electromagnetic harmonic oscillators. An oscillator can radiate energy only in
quanta with E = nhν, where n is a positive integer or zero, ν is the frequency of
the oscillator and h is a constant now called Planck’s constant. The remaining two
phenomena were explained by Einstein and Compton by assuming that radiation
itself, in particular light, consists of particles, called photons, each photon
possessing the energy hν. Thus, light has a dual nature, sometimes exhibiting the
behaviour of
waves and at other times showing the characteristics of particles. In 1923, a French
PhD scholar, de Broglie (pronounced de Broy), extended the idea of the duality of
light to the duality of matter. This extension of a concept to cover new realms is not

doi:10.1088/978-0-7503-1206-6ch1 1-1 ª IOP Publishing Ltd 2015

 Quantum
Mechanics

something new in physics. We remember that Newton had shown that the laws of
mechanics had the same form in all inertial frames of reference. Einstein, whose
genius, in the history of physics, is almost unparalleled in the twentieth century,
extended this idea to the entire field of physics by demanding that laws of physics
should have the same form in all inertial frames of reference. And this became one
of the two basic postulates of special relativity. However, it must be emphasised
that Newton proposed it only as a characteristic of the second law of motion but
Einstein
made it a criterion for the validity of any law of physics. de Broglie wrote that
‘after long reflection in solitude and meditation, I suddenly had the idea, during
the year
1923, that the discovery made by Einstein in 1905 should be generalised by
extending it to all material particles and notably to electrons’. He assumed that if
p is the magnitude of the three-momentum of a particle of energy E, and λ is the
wavelength and ν the frequency of the associated wave, then, in addition to E =
hν, we must have
h
p= . (1.1)
ν
According to Einstein, this hypothesis of de Broglie’s about the dual nature of
matter was ‘a first feeble ray of light on this worst of our physics enigmas’. In 1927
the experiment of Davisson and Germer, in which electrons were scattered by a
crystal surface with typical diffraction effects, confirmed this daring hypothesis
which ultimately demolished the classical picture of physics.
To get a taste of quantum theory, we analyse the photoelectric effect and
Compton scattering.

1.2 The photoelectric effect

The photoelectric effect, the emission of electrons by a metal when light falls on it,
was discovered by Hertz in 1887. Experiments showed the following
characteristics of this effect. When light falls on a metal surface in a vacuum, the
emission of electrons depends upon the frequency of the incident light. There is a
minimum frequency of light which is required for the emission of electrons from a
metal. The value of this threshold frequency varies from metal to metal. The
emission of electrons as well as the energy of the emitted electrons, photoelectrons,
does not depend upon the intensity of the light source. However, if electrons are
emitted, then the magnitude of their current is proportional to the intensity of the
incident light. Finally, the energy of the photoelectrons varies linearly with the
frequency of the light.
The classical theory of electromagnetic radiation can explain some of these
characteristics but not all of them. Credit for solving this problem goes to Einstein
who, in 1905, refined and extended the ideas Planck used to explain the black body
radiation spectrum and assumed that ‘light consists of quanta of energy, called
photons’. In fact, Planck had introduced the concept of material resonators
possessing quanta of energy nhν, where n is an integer, while Einstein assumed
that each quantum of light possesses the energy hν. The absorption of a single

1-
 Quantum
Mechanics

photon by an electron increases the energy of the electron by hν. Part of this energy
is used to remove the electron from the metal. This is called the work function. The
remaining part of the energy imparted to the electron increases its velocity and
consequently its kinetic energy. Thus if hν, the energy of a photon incident on a
metal is greater than the energy E required to separate the electron from the metal,
and v is the velocity of the emitted electron, then the following relation must hold:
1
hν = E + mv2. (1.2)
2
All the characteristics of this effect are easily explained by the concept that light
consists of photons. The above formula shows that if the energy of the incident
photon is less than the work function, the electrons cannot be separated from the
surface of the metal and therefore will not be emitted. For a particular metal, the
work function E being constant, the relationship between the energy of the incident
photon and the kinetic energy of the emitted electron is linear. It is also clear that a
more intense source of light will cause photons to be emitted at a greater speed and
this will produce a stronger electron current. Thus Einstein was able to provide a
completely satisfactory picture of the photoelectric effect by using the concept of
the quantum nature of light.
In fact, the dual nature of light is brilliantly reflected by the very assumption
Einstein made about the energy of a photon. The frequency is determined by the
wave nature of light and is used to define the energy of the particles constituting the
light. It is interesting to note that, in 1921, Einstein was awarded the Nobel Prize
in
physics ‘for his services to Theoretical Physics and especially for his discovery of the
law of the photoelectric effect’ and not for propounding special relativity in 1905 and
general relativity in 1915. His extraordinarily remarkable work on relativity
changed
the complexion of the entire field of physics and ensured him a seat among the
immortals of the subject, but surprisingly this magnificent contribution to the pool
of knowledge was never considered specifically for that enviable prize!

1.3 The Compton effect

Compton was an American physicist who in 1923 performed a crucial experiment
which strongly confirmed the corpuscular nature of light. The results of this
experiment were explained by Compton himself and independently by Debye. The
experiment not only confirmed the law of the conservation of energy, which was
previously verified by the photoelectric effect, but also the law of conservation of
linear momentum. It was noticed that when electromagnetic radiation of high
frequency is incident on electrons of a light element in which the electrons are loosely
bound to the nucleus and can be treated as free, the scattered radiation is found to
have a smaller frequency than the radiation of the original frequency. This is
known as the Compton effect. The experiment exhibits that the change in the
frequency of incident radiation is independent of its initial frequency and depends
only upon the angle of scattering. This can be satisfactorily explained by the
quantum theory of light by making use of relativistic expressions for various
quantities.
1-
 Quantum
Mechanics

Figure 1.1. The Compton effect.

Consider the scattering of a photon of frequency ν falling on an electron of rest

mass m0, in a frame of reference in which the electron is at rest. Let ν′ be the
frequency of the scattered photon. Let p be the linear momentum of the electron
after its collision with the photon and α the angle between the final and initial
directions of the photon. This is shown in figure 1.1. Since the incident radiation is
of high frequency, the incoming photon, by virtue of the relation E = hν, is very
energetic. Therefore the electron which is lightly bound to its atom may be
considered, to a good approximation, as free. Thus the Compton effect may be
treated as a problem of the collision between a photon and a stationary free
electron. The initial energy of the photon is hν and its initial momentum is hν/c,
while its final energy and momentum are hν′ and hν′/c, respectively. According to
the laws of conservation of energy and momentum, we have

hν + m0c2 = hν′ + E (conservation of energy), (1.3)

hν hν′
a= b + p (conservation of momentum), (1.4)
c c
where a and b are unit vectors in the directions of the incident and scattered photons,
respectively, and E is the energy of the electron after the collision. Rearranging the
terms in (1.3), we have

hν − hν′ = E − m0c2.

Squaring both sides of the above equation, we obtain

h2ν2 + h2ν′2 − 2h2νν′ = E 2 + m 2 c4 − 2Em0c2. (1.5)

0

Rearranging the terms in (1.4) and multiplying through by c, we have

hν a − hν′ b = cp.

Squaring the two sides of the above equation, we obtain

h2ν2 + h2ν′2 − 2h2νν′ cos α = c2p2

(1.6)

1-
 Quantum
Mechanics

where α is the angle between a and b. Substituting the expression for c2p2 from the
equation E 2 = m 2 c4 + c2p2 into (1.6), we obtain
0

h2ν2 + h2ν′2 − 2h2νν′ cos α = E 2 − m 2 c4. (1.7)

0

Subtracting (1.5) from (1.7), we have

(
2h2νν′(1 − cos α) = 2m0c2 E − m0c2 ) = 2m c (hν − hν′)
0
2

where, in obtaining the expression on the extreme right, we have made use of (1.3).
The above equation can be written as
h ν − ν′ 1 1
(1 − cos α) = = – = λ′ λ
2
− ,
m0c νν′ ν′ ν c c
or
h
λ′ − λ = (1 − cos α). (1.8)
m0c
The quantity h/m0c is called the Compton wavelength and has the value 0.02426 ×
10−8 cm. Equation (1.8) has been found to be consistent with experiments. For his
contribution, Compton shared the Nobel Prize in physics in 1927.

1.4Problem
Heisenberg’s uncertainty principle
In classical physics,
1.1. Comment on itthe
is statement
tacitly assumed
that in that the operationeffect
the photoelectric of observation
the photondoes
not appreciably disturb a system and, at least in principle, the disturbance
transfers all of its energy while in the Compton effect only part of the caused
by the measurement process can be rectified
energy is transferred to the electron. exactly. It required the ingenuity of
Heisenberg, one of the most brilliant even among the Nobel laureates, to show that
wave–particle duality imposes restraints on simultaneous precise measurements of
position and momentum. The measurement process in general disturbs a system by
an amount which cannot be predicted. For instance, consider the hypothetical
experiment shown in figure 1.2, devised for the precise measurement of the
position of an electron. The apparatus includes a microscope with 2α as its
aperture. The electron beam is moving in the positive direction of the x-axis with a
well-defined momentum px. In order to measure the position of an electron, it has to
be observed. For that purpose, we shine a beam of light along the negative x-axis.
Now, it is necessary that at least one photon after falling on it should be scattered
into the microscope so that the observer sees it through the microscope. Due to the
particle

1-
 Quantum
Mechanics

Figure 1.2. Measurement of the position of an electron.

nature of light, as a photon strikes the electron, the latter is disturbed. The
momentum of recoil of the electron could be calculated if the initial and final
momenta of the electron were known. But because of the finite aperture of the
microscope, the photon can enter it along any direction on the illumination cone of
the observer. That is, the direction of the photon scattered into the microscope is
undetermined within the angle subtended by the aperture. The uncertainty in the
measurement of the momentum of the electron is

Δpx
hν
=2sin α. (1.9)
c
We also notice that, using standard optical theory, the resolving power of the
microscope is given by
λ
Δx = , (1.10)
sin α
where λ is the wavelength of light. This means that Δx gives the uncertainty in the
position of the electron. The uncertainty Δx can be made as small as we like by
making λ as small as we please and/or making the aperture 2α as large as we
desire. But this will enhance the uncertainty in the measurement of momentum. For
instance, if we decrease λ, i.e. increase the frequency of the incident light, it will
certainly decrease the uncertainty Δx in the measurement of the position of the
electron, but it will increase the uncertainty Δpx because the photon striking with
greater energy (hν) will disturb the electron to a greater extent. Multiplying (1.9)
and (1.10), we obtain

ΔxΔpx
hν
=2 λ= (1.11)
2h
c
as λ = c/ν. This means that the product of the uncertainties in the simultaneous
measurement of the x-components of position and momentum is of the order of
Planck’s constant. This is just one example. In fact, any experiment designed for a
simultaneous precise measurement of position and momentum will encounter the
1-
 Quantum
Mechanics

same constraint. This is not an error in experimental measurement. It is inherent in

nature in the sense that it is due to the unavoidable interaction between the observer
and the observed during the process of observation. The above analysis shows that
a
simultaneous precise measurement of position and momentum is impossible. This is
known as Heisenberg’s uncertainty principle or principle of indeterminacy. It can be
shown that similar uncertainty relations exist between energy and time and angular
momentum and angle. In other words, a simultaneous precise measurement of two
canonically conjugate variables is impossible.

1.5 The correspondence principle

Although classical mechanics breaks down when applied to determining the behaviour
of tiny objects such as electrons, protons, etc, it has been providing correct answers
to mechanical phenomena at the macroscopic level. Therefore, at this level,
quantum theory should be consistent with classical mechanics. This is known as
Bohr’s correspondence principle and is said to serve as a guide in discovering the
correct
quantum laws. In fact, under old conditions, a new theory should always yield the
same results as the old theory which it is replacing, because the original theory has
been explaining the experimental data in its own domain. In the case of quantum
mechanics, this correspondence may be specified by claiming that, for large
quantum numbers, quantum theory must be consistent with classical physics.
Moreover, if a quantum system has a classical analogue, then for the limit h → 0, it
must yield the corresponding classical results. Thus, in the uncertainty principle, as h
→ 0 in the classical limit, the
product ΔxΔpx → 0 and therefore a simultaneous precise measurement of position
and momentum at macroscopic level becomes permissible. The importance of the
correspondence principle lies not in stating that quantum theory should yield the
same results as classical mechanics at the macroscopic level, but in describing
the conditions under which it should happen.

1.6 The Schrödinger wave equation

There is no doubt that Planck’s quantum theory of black body radiation, Einstein’s
hypothesis of light quanta for the explanation of the photoelectric effect, Bohr’s
postulates regarding the interpretation of the spectrum of the hydrogen atom and de
Broglie’s hypothesis about the dual nature of matter were the milestones in the
progress of physics from 1900 to 1923. But physicists desired a differential
equation
which could govern the behaviour of mechanical phenomena and consequently
explain various experimental results. In 1926, Schrödinger set up such a differential
equation which is named after him and was supposed to replace Newton’s second
law of motion as the basic law of nature in mechanics. The mechanics based on this
differential equation was called wave mechanics and is now known as quantum
mechanics. Schrödinger’s differential equation undoubtedly outshone all the above-
mentioned postulates and started commanding immense attention in the physics
community immediately after its advent. We will now set up this differential
equation. It must be clearly stated that the Schrödinger wave equation cannot be
logically derived. The historical development may make its presence somewhat

1-
 Quantum
Mechanics

plausible. But we will establish it by adopting an operational technique which is

to-the-point and simple. We proceed as follows.
In terms of the kinetic energy T and the potential energy V of a particle, the non-
relativistic classical expression for its total energy E is given by
E=T+V,
where V is, in general, a function of space and time coordinates, V = V(r, t), but
1 p2
T= mv2 = ,
2 2m
where m is the mass, v is the velocity and p is the linear momentum of the particle.
This yields

p2
E = 2m + V. (1.12)
It is assumed that the transition from classical to quantum mechanics is made by
interpreting E, p and V as operators such that
∂
E → iℏ , (1.13)
∂t
h
where ,
2π
ℏ
p → −iℏ∇, (1.14)
and

V → V. (1.15)
The operator ∇ is given by

∂ ∂ ∂
∇= i+ j+ k,
∂x ∂y ∂z

where x, y, z are space coordinates and i, j, k are unit vectors along the x-, y- and z-
axes. The time coordinate, wherever it occurs, is denoted by t. Note that while E
and p are interpreted as differential operators, the potential energy is assumed to be
only a multiplication operator, its form remaining unchanged when moving from
classical to quantum mechanics. Equation (1.12) therefore yields
∂
i ℏ2 2
ℏ = − ∇ +V,
∂t 2m
where
∂2
∇2 ≡ ∂2
2 + .
∂x 2∂
+ ∂y2 ∂z2
1-
 Quantum
Mechanics

If a function Ψ(x, y, z, t), which for convenience we may also write as Ψ(r, t) or
merely as Ψ, represents the particle under consideration, then operating this
equation on Ψ(r, t) ≡ Ψ and interchanging the two sides of the equation thus
formed, we obtain ⎛ ⎞
ℏ2 2 ∂Ψ
⎜− ∇ + V ⎟ iℏ , (1.16a)
Ψ=
⎝ 2m ⎠ ∂t
or
∂Ψ
H Ψ = iℏ , (1.16b)
∂t
where
⎛ ℏ2 ⎞
2
H≡⎜ − ∇ +V (1.16c)
⎟ ⎝ 2m
⎠

is the Hamiltonian operator for the system. Equation (1.16a), equivalently (1.16b), is
a partial differential equation in four independent variables and is known as the
time- dependent Schrödinger wave equation. It is called a wave equation as it is
similar to a differential equation for waves. It is assumed to be the fundamental
differential
equation governing the behaviour of mechanical phenomena. It replaces Newton’s
second law of motion in mechanics. However, unlike Newton’s law of motion, it is
not garbed in words. This law of nature presents itself only as a differential
equation. The function Ψ(r, t), a solution of the time-dependent Schrödinger wave
equation for a system, is called a wave function and is necessarily a complex function
because of the
complex nature of the differential equation. It should not be considered to be a physical
wave. It is actually a mathematical function containing all the information that can
be obtained about the system it represents. The time-dependent Schrödinger wave
equation is a linear partial differential equation of the first order in the time
derivative and of the second order in the spatial derivative. This implies that if its
solution at a particular time t0 is known, it can be calculated at any time t. But just
as in the case of
Newton’s second law of motion, there is no logical derivation of the Schrödinger wave
equation. It was a brilliant guess of Schrödinger in the perspective of the dual
nature
of matter. The ultimate test of a theory comes from its confrontation with
experimental data. And of course, for non-relativistic mechanical phenomena, the
theory based on Schrödinger’s wave equation has emerged successful.
Whenever we have to solve a physical problem in quantum mechanics, we resort
to this equation, just as in classical mechanics we use the second law of motion. To
write the Schrödinger wave equation for a particular system, we have to find the
classical expression for the potential energy V of the system and substitute it into
equation (1.16a). This gives us the desired differential equation which may be
solved to obtain a complex solution Ψ(r, t) ≡ Ψ. Since the differential equation
contains only a first-order time derivative, the wave function is uniquely

1-
 Quantum
prescribed, once its Mechanics
value at a time t = t0 is known. But how is this function Ψ(r, t) interpreted so as to

1-
 Quantum
Mechanics

relate it to physically measurable characteristics of the system? Certain

prescriptions were proposed but after facing insurmountable difficulties, on a
suggestion by Born, a consensus was ultimately developed in the physics
community. The complex
function Ψ(r, t) representing the particle, being itself not directly observable, is
interpreted so that
Ψ*(r, t) Ψ(r, t) dxdydz
represents the probability of finding the particle in a small volume dxdydz (≡ dτ) about
the point r and at time t. It must be emphasised that this interpretation is only a
hypothesis. Its validity is established by the success of its predictions. No doubt,
the time evolution of the wave function Ψ(r, t) is inextricably connected to
probabilistic
concepts. It has been rightly stated, in flowery language, that Ψ*Ψ is the window
through which we can view the world of the atom. Schrödinger himself was
shocked when he was told about the statistical interpretation of quantum
mechanics. He once
told Bohr ‘If we are going to stick to this damned quantum jumping, then I regret
that I ever had anything to do with quantum theory’. Bohr quipped ‘But the rest
of us are thankful that you did’. To this interpretation, which demolishes the
determinism of classical mechanics, Schrödinger and Einstein could not reconcile
themselves, even to the last days of their lives. The total probability of finding the
particle in a volume in which it is confined is

∫ Ψ*(r, t) Ψ(r, t) dτ ( ≡ ∫ Ψ*(r, t) 2 dτ )

and must be 100%, i.e. 1, because the particle must be somewhere in that volume.
Hence we may write

∫ Ψ*(r, t) Ψ(r, t) dτ = 1. (1.17)

The wave function Ψ(r, t) is then said to be normalised to unity or simply normalised.
The above equation is said to express the normalisation condition. This equation
exists only if the integral

∫ Ψ*(r, t) Ψ(r, t) dτ
converges, i.e. it is finite; for instance, if the wave function goes to zero sufficiently
rapidly as r tends to infinity. The function Ψ(r, t) is then said to be square integrable or
to have an integrable square. Symbolically, in this case, Ψ(r, t) → 0 as ∣r∣ → ∞. This
is the boundary condition which is always satisfied when the state is bound. For
instance, an electron bound to the hydrogen nucleus, constitutes the hydrogen atom
— a bound state. For such a state, the particle can never go to infinity. The
integral

∫ Ψ*(r, t) Ψ(r, t) dτ ≡ ∫ Ψ(r, t) 2 dτ

exists and the function Ψ(r, t) can be normalised. In certain cases, for instance for a
free particle, the above integral may diverge. Then a somewhat different
formulation
1-
 Quantum
Mechanics

of the normalisation condition is to be given which will be considered at a later

stage. Ψ*(r, t) Ψ(r, t) is the probability density, i.e. the probability per unit volume.
Notice that the statement in the above form about the probability is valid only if
Ψ(r, t) is normalised. It may be emphasised that the above prescription
incorporates the
statistical interpretation in the basic differential equation, the Schrödinger wave
equation. Since this is a homogeneous, linear differential equation, if Ψ(r, t) is a
solution of this differential equation, then CΨ(r, t) is also a solution of the same
where C is a constant. In most cases, the value of C can be determined by using the
normalisation condition.
In one-dimensional space, Ψ*(x, t) Ψ(x, t) dx is interpreted as the probability of
finding the particle in the length dx between x and x + dx at time t.
It is important to state at this stage that we have only interpreted Ψ so as to
relate it to the probability distribution. Actually, we have to compute various
physical quantities, representing dynamical variables in classical mechanics, such
as position,
linear momentum, energy and components of angular momentum, so that we may
compare our theoretical results with experimental data. Another postulate is to be
proposed for this purpose. Since new concepts are involved, we will consider it
later in this chapter when these concepts have been introduced.

1.7 Constraints on solutions

Every solution of the Schrödinger wave equation is not physically acceptable. The
above interpretation of Ψ*(r, t) Ψ(r, t) dτ as the probability of finding the particle
in the small volume dτ about r and at time t imposes certain constraints on the
solution Ψ(r, t) of this second-order partial differential equation. In fact, in order
for the solution to be physically acceptable, it must be well-behaved, i.e. the wave
function should be finite, single valued and continuous. Moreover, its first
derivatives with respect to space co-ordinates must be continuous. This is analysed
below.
1. The function Ψ(r, t) must be finite for all values of x, y, z. In fact, Ψ(r, t)
should be such that it vanishes sufficiently rapidly as infinity is approached
so
as to give us no trouble; the function remains square integrable. This is so
because otherwise the probability of finding the particle in the small region
about a point where the function diverges will become infinite which is
physically unacceptable.
2. The function Ψ(r, t) must be single-valued, i.e. for each set of the values of
the variables it should have only one value. This is essential because
otherwise the probability of finding the particle at a particular point and
at a certain time will not be unique; it will have more than one value, each
value depending upon the choice of the multivalued function Ψ(r, t) such as
tan−1x. Strictly speaking, according to this argument, it is Ψ*(r, t) Ψ(r, t)
which should be single-valued. However, successful results for the character-
istics of some physical quantities, such as the z-component of orbital angular
momentum, require that the wave function be single-valued.
3. The function Ψ(r, t) and its first derivatives with respect to space co-
ordinates should be continuous in all parts of the region.

1-
 Quantum
Mechanics

Before considering the last characteristic in detail, let us recall a mathematical

theorem: a function continuous at a point is not necessarily differentiable there but
a function differentiable at a point is necessarily continuous there. Thus if a function f
is differentiable at a point, i.e. if df/dx exists at a point, then f must be continuous
there. Similarly, if d2f/dx2 exists, then df/dx must be continuous there. To make
things simple, consider the time-dependent Schrödinger wave equation in one-
dimension,
⎛ ℏ2 ∂ 2
⎜−
∂Ψ + V ⎟Ψ = iℏ
⎞ . (1.18)

⎝ 2m ∂x2 ⎠ ∂t

Rearranging the terms, we have

ℏ2 ∂2Ψ ∂Ψ
VΨ = + iℏ .
2m ∂x2 ∂t
We will consider the case when the potential energy of a physical system is finite,
whether continuous or with a number of finite discontinuities, because infinite
energies do not occur in nature. Then the left-hand side of the above equation is
finite. Consequently, both the terms on the right-hand side should be finite. Thus, as
∂2Ψ/∂x2 is finite, i.e. the differential coefficient of ∂Ψ/∂x exists, the function ∂Ψ/∂x
must be continuous. Moreover, as ∂Ψ/∂x exists, i.e. the differential coefficient of Ψ
exists, the function Ψ must be continuous. Hence, the condition that the wave
function and its first space derivative should be continuous is a requirement
imposed by the finiteness of Ψ and consistency of the Schrödinger wave equation. The
analysis can easily be extended to three dimensions.
It can be mentioned that if Ψ is assumed to be continuous, then we need not
assume that it is finite because every continuous function is finite.
Owing to the requirement that the function Ψ(r, t) must be well-behaved and its
first derivative with respect to the space coordinate should be continuous, all the
mathematical solutions of the Schrödinger wave equation are not physically
acceptable. This in turn means that for a physical system, only those energies will
be allowed which correspond to physically acceptable wave functions. We will find
that by the requirement of admissible wave functions, energy and some other
physical quantities are quantised.
The time-dependent Schrödinger wave equation is of first order in the time-
derivative. Therefore, if the wave function Ψ(r, t0) at any initial time t0 is known,
the wave function Ψ(r, t) at any time t can be calculated.

Problem
1.2. Comment on the remark that the assumptions that the wave function be
finite and single-valued at all points in configuration space may be
more rigorous than necessary.

1-
 Quantum
Mechanics

Remark
If Ψ is finite and V is continuous or has finite discontinuities, then the
consistency of the Schrödinger wave equation requires that Ψ and its first
space derivative should be continuous. It may be emphasised that it is not a
consequence of the probabilistic interpretation of Ψ. This is a characteristic
of certain types of second-order differential equations.

1.8 Eigenfunctions and eigenvalues

Before we proceed further, we will define a few terms and illustrate them with the
help of examples. We first define the eigenfunctions and eigenvalues of any
operator. Consider an operator A operating on a function ϕ such that
Aϕ = λϕ, (1.19)

that is, it reproduces the function ϕ multiplying it with a constant λ. Then the
function ϕ is called an eigenfunction (or characteristic function) of the operator A
with eigenvalue (or characteristic value) λ or corresponding to the eigenvalue λ. The
equation itself is called an eigenvalue equation (or characteristic equation). Thus,
in the eigenvalue equation
d2
(sin 3x) = −9(sin 3x),
dx2
−9 is the eigenvalue of the operator d2/dx2 corresponding to the eigenfunction
sin 3x.
An operator has several eigenfunctions with the corresponding eigenvalues. These
eigenvalues may be discrete or continuous. If the eigenvalue spectrum is discrete,
the values are written as, say, λ1, λ2, λ3, …. If the eigenvalue spectrum is continuous,
the
values are denoted by λ′, λ″, λ‴, …. The eigenvalue spectrum may be partially
discrete and partially continuous. For simplicity, in general analysis, as far as
possible, we will be considering only discrete eigenvalues.
A set of functions ϕ1, ϕ2, …, ϕn is said to be linearly independent if its linear
combination a1ϕ1 + a2ϕ2 + ⋯ + anϕn cannot be made equal to zero for all values
of the variables except by taking all a equal to zero. For instance, sin x and cos x are
two
linearly independent functions as their linear combination, a1 sin x + a2 cos x, cannot
be made equal to zero for all values of the variable x except by taking a1 = a2 = 0.
Suppose that ϕ1, ϕ2, …, ϕn are n linearly independent eigenfunctions of an
operator A corresponding to the same eigenvalue λ. Then we may write
Aϕ1 = λϕ1,
Aϕ2 = λϕ2,
⋯ (1.20)
Aϕn = λϕn.

1-
 Quantum
Mechanics

The number λ is called an n-fold degenerate eigenvalue of the operator A,

corresponding to linearly independent eigenfunctions ϕ1, ϕ2, …, ϕn. These eigenfunc-
tions are called n-fold degenerate, corresponding to the same eigenvalue λ. The
number n is called the degree of degeneracy of the eigenfunctions.

1.9Problem
The principle of superposition
We 1.3.
haveShow −4time-dependent
seen that the is a two-fold degenerate eigenvalue
Schrödinger of the operator
wave equation can be written as
d2/dx2 corresponding to linearly independent
∂Ψ eigenfunctions sin 2x and
cos 2x. H Ψ = iℏ . (1.16b′)
∂t
For convenience, changing the two sides of the above equation, we obtain
∂Ψ
iℏ ∂t = H Ψ. (1.21)
Let Ψ1 and Ψ2 be two solutions (maybe belonging to different values of energy) of
this differential equation so that
∂Ψ1
iℏ =H (1.22)
Ψ1
and ∂t
∂Ψ2
iℏ ∂t = H Ψ2. (1.23)
It can easily be verified that a linear combination of these solutions, i.e. a1Ψ1 + a2Ψ2,
is also a solution of the Schrödinger wave equation. This is shown below:
∂ ∂Ψ1 ∂Ψ2
iℏ (a1Ψ1 + a2 Ψ2) = a1iℏ + a2 iℏ
∂t ∂t ∂t
= a1 H Ψ 1 + a2 H Ψ 2
= H (a1Ψ1 + a2 Ψ2). (1.24)
In fact, we could directly state that as Schrödinger’s second order time-dependent
partial differential equation is linear (because the function Ψ and its derivatives
occur only to the first degree and not as higher powers or products), every linear
combination of its solutions is also a solution of this differential equation. This is
known as the principle of superposition and plays a very important role in quantum
mechanics. It can be mentioned that this is a characteristic of every homogeneous
linear differential equation.

1-
 Quantum
Mechanics

This is the right moment to point out that in classical mechanics, knowledge
about the position and momentum of a particle at any time describes what is called
the state of the particle. If we know the position and momentum of a particle at
any time, then we can compute its position and momentum at any other time by
using the second law of motion. This implies that if the initial state of the system is
given, the values of all other variables can be determined exactly for all times. In
quantum mechanics, according to Heisenberg’s uncertainty principle, position and
momentum cannot be measured precisely at the same time. Therefore, the state of
the particle in the classical sense cannot be described. Hence, in quantum mechanics,
there is no concept of the trajectory of a particle as it is determined by a
simultaneous precise knowledge of position and momentum at every moment;
such a trajectory does not exist when quantum effects are important as it is then
impossible to keep track of the particle. However, all the accessible information
about the particle is contained in the wave function Ψ. Naturally, the question arises:
how do we define the state of a particle in quantum mechanics? In fact, we assume as
a fundamental principle of quantum mechanics that every physically acceptable
solution of the Schrödinger wave equation represents a state of the system. The
time- dependent Schrödinger wave equation is a linear partial differential equation
of first degree in the time variable. Therefore, if Ψ is known at a particular time t0,
it can be
determined at any time t. Suppose that Ψ1 and Ψ2 represent two states of a system
which correspond to definite energy values, say E1 and E2. These states are then
called eigenstates of the system. Since a linear combination of these solutions, a1Ψ1 +
a2Ψ2, where a1 and a2 are (complex) numbers, is itself a solution of this differential
equation, it also represents a state of the system. However, this state does not
correspond to a definite energy. Such a state is called a quantum state of the
system. It has no analogue in classical mechanics where each state, like an
eigenstate in quantum mechanics, corresponds to a definite energy. Now suppose
that the functions Ψ1(t0) and Ψ2(t0) are two solutions of the Schrödinger wave
equation at time t0. Let

Ψ(t0) = a1Ψ1(t0) + a2 Ψ2(t0)

be a linear superposition of Ψ1(t0) and Ψ2(t0) at time t0. Suppose that the functions
Ψ1(t0) and Ψ2(t0) develop with time into functions Ψ1(t) and Ψ2(t). Then, by virtue
of the fact that the Schrödinger wave equation is linear in Ψ, we must have

Ψ(t) = a1Ψ1(t) + a2 Ψ2(t),

i.e. at any time t, Ψ(t) is the same linear combination of the functions Ψ1(t) and
Ψ2(t). The fact that the first of these equations entails the second equation is a
consequence of the linearity of the time development of a state.
An overview of various disciplines of physics shows that the principle of
superposition is one of the most significant attributes of the wave concept. It is
common to all types of waves. We may transcend a step higher than this and take it
as a characteristic of every homogeneous linear differential equation: a linear

1-
 Quantum
Mechanics

superposition of the solutions of a homogeneous linear differential equation is also

a solution of this differential equation. Quantum mechanics which emphasises this
formal aspect was developed by Dirac, a British Nobel laureate. It describes the
possible states of a system by abstract quantities called state vectors which obey
the principle of superposition. Indeed, the essence of this principle as reflected
by state vectors is much more attractive than that predicted by wave functions.
A quantity which can be measured is called an observable. In classical physics,
observables are represented by ordinary variables. In quantum mechanics, however,
observables are, by assumption, represented by operators. It is customary in the
literature to use the same letter for the observable, variable and operator. We will
also adopt this convention.

1.10 Complementarity
The statistical interpretation of the wave function and Heisenberg’s principle of
uncertainty are the concepts with which some eminent physicists of the good old
days were finding it difficult to come to terms. Bohr therefore devoted his full
attention to detailed analyses of various new concepts which were leading to new
trends in scientific attitudes and its philosophical consequences. His analysis that
the wave and particle aspects of matter are opposing but complementary modes of
its realisation has been named the principle of complementarity. This is the essence
of his views on the conceptual basis of quantum mechanics but it does not help in
making calculations in this field. In fact, classical physics, based upon the knowl-
edge gained from every day experience, i.e. from the behaviour of macroscopic
objects, tells us that an object in nature can behave either as a particle or as a wave.
It cannot exhibit both characteristics. According to the principle of complemen-
tarity, analysis at a microscopic level reveals that an object can behave both as
a particle and a wave but the two modes cannot be realised at the same time.
A measurement which emphasises one of the wave–particle attributes does so at the
expense of the other. An experiment designed to exhibit the particle properties does
not give any information on the wave aspect and vice versa. For instance, the
observation of cloud chamber tracks does not involve wave aspect at all while
interference and diffraction experiments do not contain any information about
the particles’ demeanour. It is interesting to note that all possibly available
knowledge about the characteristics of a microscopic entity, say an electron, is
contained in the wave function. We will express it like this. A microscopic entity
under one situation shows those properties which at macroscopic level are
attributed to particles and we say that it is behaving as a particle. In some other
situation, the same entity exhibits characteristics which at macroscopic level are
assigned to waves. Then we say that the entity is behaving as a wave. Actually, the
behaviour is determined by the same wave function and it is different, as it must be,
under different circumstances. We are surprised simply because it is not what we
were expecting from our experience in everyday life. But nature does behave that
way and sometimes it produces results even against those expectations which are
based on very careful considerations and analysis!

1-
 Quantum
Mechanics

1.11 Schrödinger’s amplitude equation

We will now show that the computations in solving the Schrödinger wave equation
are simplified if the potential energy V of the system does not depend upon time
explicitly: V ≡V(x, y, z) ≡V(r). Let us assume that in this case the wave function Ψ(r,
t) representing the particle can be expressed as a product of two functions, one
depending only on space coordinates and the other depending on time alone. If we
denote these functions, respectively, by ψ(x, y, z) (also written as ψ(r)) and ϕ(t), then
we may write
Ψ(x, y, z, t) = ψ (x, y, z)ϕ(t). (1.25)
Differentiating twice with respect to x, we obtain
∂2 Ψ ∂2 ψ
= ϕ. (1.26)
∂x2 ∂x2
where Ψ ≡ Ψ(x, y, z, t) ≡ Ψ(r, t), ψ ≡ ψ(x, y, z) ≡ ψ(r) and ϕ ≡ ϕ(t). We can obtain two
similar expressions for the second-order derivatives with respect to space coordinates
y and z. Next, differentiating (1.25) with respect to t, we obtain
∂Ψ dϕ
. (1.27)
∂t = dt
Notice that for the function ϕ, we have used the ordinary derivative instead of the
partial derivative as this function depends upon one variable t only. Substituting the
expressions from (1.26), etc, and equation (1.27) in equation (1.16a), we obtain
2 ⎡ 2 2 2 dϕ
⎤ ℏ ⎢ ∂ ψ ϕ + ∂ ψ ϕ + ∂ ψ V(x, y, z)ψϕ = iℏψ
– ,
⎥
2m ⎣ ∂x2 ∂y2 ϕ +
∂z2 ⎦ dt
or
ℏ2 2 dϕ
– (∇ ψ )ϕ + Vψϕ + iℏψ .
2m dt
Dividing throughout by ψϕ, we obtain
ℏ2 1 2 1 dϕ
– (∇ ψ ) + V = iℏ . (1.28)
2m ψ ϕ dt
The right-hand side of this equation is a function of time only while, as V depends
explicitly only on space coordinates, the left-hand side depends upon space
coordinates alone. Therefore, a variation in space coordinates will not affect the
right-hand side while a variation in time will not affect the left-hand side. This is
possible only if each side is equal to the same constant. We denote this constant by
E. Then we may write
1 dϕ
iℏ =E
ϕ dt (1.29)

1-
 Quantum
Mechanics

and
1
2
(∇2 ψ ) + V = E,
ℏ
–
2m ψ
or ⎛ ℏ2 2
⎞⎜ − ∇ +V Eψ. (1.30)
⎟

⎝ 2m ψ=
⎠
This is called the Schrödinger amplitude equation. The operator in brackets in the
above equation is the Hamiltonian operator H of the particle. The amplitude
equation therefore can be written as
Hψ = Eψ. (1.31)

It is called the Hamiltonian form of the Schrödinger amplitude equation. This is an

eigenvalue equation. The constant E is the eigenvalue of the Hamiltonian operator
H corresponding to the eigenfunction ψ. The Schrödinger amplitude equation thus
takes the form of an eigenvalue equation for the Hamiltonian H and this simplifies
the analysis of the problem.
Let us first solve the differential equation (1.29) involving only time as an
independent variable. Transposing iℏ to the right-hand side of this differential
equation and integrating with respect to time t, we obtain
iEt
ln ϕ(t) = −
ℏ + K.
Choosing the initial condition that makes K = 0, we obtain
iEt
ln ϕ(t) = − ,
ℏ
or
⎛ iEt ⎞
ϕ(t) = exp⎜ − ⎟. (1.32)
⎝ ℏ⎠

The expression in brackets on the right-hand side shows that E has the dimensions
of energy. We will assume here but will prove later on that E is the total energy of
the system.
Let us next consider the differential equation (1.30) that can be written as
2m
∇2 ψ + (E − V )ψ = 0. (1.33)
2
ℏ
This is the time-independent Schrödinger wave equation or the amplitude equation
or the steady-state Schrödinger equation. This differential equation for a system
can be solved only if the expression for the potential energy of the system is known.
1-
 Quantum
Mechanics

The solution of this differential equation is denoted by ψ(r). Equation (1.25) can now
be written as
⎛ iEt ⎞
Ψ(r, t) = ψ (r) exp ⎜ − ⎟,
⎝ ℏ⎠

where for convenience we have written r for x, y, z. If we are interested in finding

the characteristics of a physical system whose potential energy does not depend
explicitly on time, then instead of using the time-dependent Schrödinger equation,
which is relatively much more difficult to solve, we can use the time-independent
Schrödinger equation which is easier to solve. We obtain the solution ψ(r) of this
differential
equation and multiply it by ϕ(t) (≡ exp(−iEt/ℏ)) so as to obtain Ψ(r, t) which
represents the system. Then a wave function of the system corresponding to a
definite energy can be written as
⎛ iEt ⎞
Ψ(r, t) = ψ (r)ϕ(t) = ψ (r) exp ⎜ − ⎟. (1.34)
⎝ ℏ⎠

Thus

Ψ*(r, t) Ψ(r, t) = Ψ*(r) ψ (r). (1.35)

The above analysis shows that the probability density for a particle whose potential
energy does not depend upon time explicitly is constant in time. For this reason, a
wave function of the form (1.34) is said to represent a stationary state or an energy
eigenstate of the particle. The energy in a stationary state is said to be sharp or
well- defined. It can be said that although the wave function Ψ(r, t)
representing the
particle is time-dependent, the probability density is independent of time. Therefore,
the system would remain in that state indefinitely. Every measurement will always
give the same value of energy. This interpretation is consistent with the uncertainty
relation
ΔEΔt ∼ ℏ,
as this means that if the system is in an eigenstate with definite energy so that ΔE = 0,
then unlimited time should be available to make that measurement. If the energy
spectrum is discrete, the lowest energy state is called the ground state of the system.
The higher energy states are called excited states of the system.
For a stationary state, the normalisation condition,

∫ Ψ*(r, t) Ψ(r, t) dτ = 1,
is simplified to

∫ Ψ*(r)e iEt/ℏ
ψ (r)e−iEt/ℏ dτ = ∫ ψ *(r) ψ (r) dτ = 1. (1.36)

1-
 Quantum
Mechanics

In one dimension, the time-independent Schrödinger wave equation reduces to the

following form:

d2ψ 2m
2 + (E − V )ψ = 0, (1.37)
dx
2
ℏ
where ψ ≡ ψ(x). It is important to note that the amplitude function ψ(x), a solution
of the Schrödinger amplitude equation and related to the wave function Ψ(x, t) by
⎛ iEt ⎞
Ψ(x, t) = ψ (x)ϕ(t) = ψ (x) exp ⎜ − ⎟, (1.33')
⎝ ℏ⎠

should also be well-behaved. This is because if ψ(x) is not finite, then Ψ(x, t) will also
be not finite and if ψ(x) is not single-valued, then Ψ(x, t) will also not be single-
valued. This is due to the fact that the time function ϕ(t) is always finite and does
not
depend upon space coordinates. As far as the continuity of the function ψ(x) and its
first derivative dψ/dx is concerned, we note from (1.37) that as ψ(x) is always finite
and infinite energies cannot be achieved in nature, the first term on the left-hand
side of this equation, i.e. the second derivative of ψ(x), should also be finite.
Therefore, dψ/dx must be continuous. Moreover, as dψ/dx exists, the function ψ(x)
should be continuous. Thus, the very consistency of the Schrödinger amplitude
equation requires that both ψ(x) and dψ(x)/dx should be continuous.
Infinite energies do not occur in nature. But let us see what will happen if at
some point the potential jumps from a finite to an infinite value. Equation ( 1.37)
shows that in this case d2ψ/dx2 will be infinite. Therefore, dψ/dx may or may not be
continuous. That is, the condition of continuity of the space derivative of the wave
function at the discontinuity of the potential cannot be imposed. However, as dψ/dx
exists because otherwise the differential equation will not hold, the function ψ(x) will
be continuous. In the solution of the eigenvalue equation Hψ = Eψ for a physical
system, we will find that the energy eigenvalues very much depend upon the
conditions imposed on the solutions to the eigenvalue equation. We take this
opportunity to point out that it is not necessary that the integration should always
be
over the entire space extending from −∞ to +∞.

1.12 The orthonormal set of functions

Consider a set of functions ϕ1, ϕ2, ϕ3, … which are individually normalised, i.e.

∫ ϕ *ϕ i dτ =
i = 1, 2, 3, … (1.38)
1, i

and mutually orthogonal, i.e.

∫ ϕ i*ϕ i dτ = 0, j ≠ i, i, j = 1, 2, 3, … (1.39)

1-
 Quantum
Mechanics

They are said to form an orthonormal set of functions. These two conditions for an
orthonormal set can be expressed as

∫ ϕ i* ϕ j dτ = δij, (1.40)

where the Kronecker delta is defined as

δij = 1, for j = I,

δij = 0, for j ≠ i.

1.13 The equation of continuity

We know that any solution Ψ(r, t) of the time-dependent Schrödinger wave
equation is such that

ρ = Ψ*(r, t) Ψ(r, t) (1.41)

is interpreted as the probability density, i.e. the probability of finding the particle in
unit volume about the point r and at time t. Differentiating with respect to time,
we obtain
∂ρ ∂ ∂Ψ* ∂Ψ
= (Ψ*Ψ) = Ψ + Ψ* . (1.42)
∂t ∂t ∂t ∂t
But by virtue of equation (1.16a), i.e.
∂Ψ ⎛ ℏ2 2 ⎞
−iℏ = ⎜− ∇ +V ⎟
ψ
∂t ⎝ 2m ⎠
and its complex conjugate
∂Ψ* ⎛ ℏ2 2 ⎞
*
−iℏ = ⎜− ∇ +V ⎟ ,
ψ
∂t ⎝ 2m ⎠
(1.42) yields
∂ρ ∂Ψ* ∂Ψ
−iℏ = −iℏ Ψ + (−iℏ)Ψ*
∂t ∂t ∂t
⎛ ℏ2 2 * ⎞ ⎛ ℏ2 ⎞
= ⎜− ∇ ψ + Vψ ⎟ ⎜− ∇ ψ + Vψ ⎟
* 2
⎝ 2m ⎠ Ψ − ⎝ 2m ⎠
ℏ2
=− (Ψ∇2 ψ * − Ψ*∇2 ψ ),
2

1-
 Quantum
Mechanics

or
∂ρ iℏ
=− (
Ψ∇2 ψ * − Ψ*∇2 ψ )
∂t 2m
iℏ
=− (
∇ Ψ∇ψ * − Ψ*∇ψ . ) (1.43)
2
If we write

iℏ
– ( Ψ ∇Ψ − Ψ∇Ψ ) = j,
* *
(1.44)
2
then (1.43) can be expressed as
∂ρ
+ ∇j = 0,
∂t

or
∂ρ
+ div j = 0. (1.45)
∂t

This is the well-known equation of continuity. This equation also arises in electro-
magnetic theory and expresses the conservation of charge. If ρ represents the
charge density, i.e. the charge per unit volume, and j = ρv is the current density, i.e.
the charge passing unit area normal to its direction of motion in one second, then
the equation of continuity expresses the law of conservation of charge. The charge
which is decreasing with time in a bounded volume is accounted for by the charge
which is crossing the surface of the bounded volume. In quantum mechanics, ρ is
interpreted as probability density. Therefore, if j is interpreted as the probability
current density, then the equation of continuity guarantees the conservation of
probability. Thus, if the probability of finding a particle in a certain bounded region
decreases with time, it should correspond to the increase in probability of finding it
outside that region. The probability current density is also called the probability
flux. It is given by
iℏ
j=− (
Ψ*∇Ψ − Ψ∇Ψ* =
Im Ψ*∇Ψ .
) ( ) (1.46)
ℏ 2m m

1-
 Quantum
Mechanics
Problem
1.4. Prove that

(
j = ℏ Im Ψ*∇Ψ . ) (1.47)
m

1-
 Quantum
Mechanics

We will now prove that the total probability of finding the particle in space is
independent of time. For one-dimensional space, we have
ρ(x, t) = Ψ*(x, t) Ψ(x, t). (1.41')
Integrating with respect to x and then differentiating with respect to t, we obtain
d ∞ d ∞
∫ ρ(x, t) dx = ∫
Ψ*(x, t) Ψ(x, t) dx. (1.48)
dt −∞ dt −∞
But, in one-dimensional space, we have
ℏ2 ∂ 2 ∂Ψ(x , t )
– + V (x)⎟ (x, t) = ih . (1.16a′)
⎞
Ψ
⎝ 2m ∂x2 ⎠ ∂t
Taking the complex conjugate of both sides of equation (1.16a′), we obtain
ℏ2 ∂ 2 * ∂Ψ*(x , t )
– + V (x)⎟ (x, t) = −ih . (1.49)
⎞
Ψ
⎝ 2m ∂x2 ⎠ ∂t
*
Substituting these expressions for ∂Ψ (x, t)/∂t and ∂Ψ(x, t)/∂t in (1.48) and
simplifying, we obtain
d ∞ iℏ
∫ ρ(x, t) dx = − ∞
∂ ⎛ ∂Ψ*(x, t) ∂Ψ(x , t ) ⎞
∫ ⎜Ψ(x, t) + Ψ*(x, t) ⎟ dx
dt −∞ 2m −∞ ∂x ⎝ ∂t ∂t ⎠
⎡
iℏ ⎛ ∂Ψ
* (x , t ) ∂Ψ(x , t )∞
=− ⎜Ψ(x, t) + Ψ*(x, t) ⎞ ⎟⎤ = 0. (1.50)
⎢
⎥
2m ⎣ ⎝ ∂t ∂t ⎠ ⎦
−∞

The last result has been obtained because the wave function Ψ(x, t) → 0 as x → ∞ so
that the integral
∞

∫−∞ Ψ (x, t) Ψ(x, t) dx

*

may converge.
Equation (1.50) shows that the total probability is conserved. This also ensures
that the normalisation is preserved: if the wave function is normalised, it will
remain normalised.

1.14 Complete sets of functions

A set of functions ψ1(x), ψ2(x), … in a variable x is said to form a complete set if an
arbitrary square integrable function ϕ(x) can be expanded in terms of them:

ϕ(x) = ∑ai ψi (x), (1.51)

i

1-
 Quantum
Mechanics

where ai are called expansion coefficients. The values of ai can be obtained as

follows. Multiplying equation (1.51) by ψ *(x) and integrating with respect to x, we
j
obtain

∫ ψ *(x) ϕ(x) dx = ∑ia ∫ ψ *(x) ψ (x) dx.

i i (1.52)
j j
If the functions ψ1(x), ψ2(x), … form an orthonormal set, then

∫ ψ *j(x) ψ (x) dx = δ
i ji .

Therefore, (1.52) reduces to

∫ ψ *j (x) ϕ(x) dx = ∑ia δ i ij (x).

Summing over i, and finally changing j to i throughout the equation, we obtain

ai = ∫ ψ i*(x) ϕ(x) dx. (1.53)

This equation determines all the expansion coefficients.

1.15 The quantum theory of measurement

We will now analyse the process of measurement in quantum mechanics in detail.
We will concentrate on how to compute the physical quantities which are to be
compared with the experiment. In fact, whenever we want to make an accurate
measurement of any quantity, we measure that quantity a large number of times
and take the arithmetic mean of the measured values. This is called the average or
mean value of the variable. The average value of a variable x is denoted by x¯:
x1 + x2 + ⋯ + xn
x¯ = . (1.54)
n
In order to have an idea about the precision of the various values we must know the
scattering or dispersion of these individual values about their average. The individual
deviations from the mean are
x1 − x¯, x2 − x¯, …, xn − x¯.
But the average of these deviations is equal to
(x1 − x¯ ) + (x2 − x¯ ) + ⋯ + (xn − x1 + x2 + ⋯ + xn nx¯
x¯) = − = x¯ − x¯ = 0
n n n
This shows that whatever the deviations of x from its mean value, the average of
these deviations is always zero. The average of these deviations is therefore not
useful as a standard for measuring dispersion. Its value, being always zero, cannot
tell us whether the individual values are close to or far away from the average.

1-
 Quantum
Mechanics

Perhaps a better idea about the dispersion of the values of a variable about their
mean can be obtained if we consider the average of the square of deviations of the
variable x from its mean value x¯. This is called the variance of x and is denoted by σ.
Thus, we may write:

(Variance of x) = σ
(x1 − x¯)2 + (x2 − x¯)2 + ⋯ + (xn − x¯)2
= n
x2 +x2 +⋯+x2 nx¯ 2 x1 + x2 + ⋯ + xn
= + − ¯
1 2 n
2x
n n n

= x¯ 2 + x¯ 2 − 2x¯x¯
= x¯ 2 + x¯ 2 − 2x¯ 2
= x¯ 2 − x¯ 2 . (1.55)

The positive square root of the variance of a variable x is called the standard
deviation of x or the root mean square deviation or the uncertainty in the value of x
and is denoted by Δx. Thus
Δx = + x¯ 2 − x¯ 2 . (1.56)

1.16 Observables and expectation values

As already mentioned, in quantum mechanics, each physical quantity (or attribute
or property) which can be measured experimentally is called an observable and is
represented by an operator. For instance, the energy of a system is an observable
and is represented by the Hamiltonian operator H. Linear momentum is another
observable and is represented by the linear momentum operator −ih∇. The state of
a
physical system is a collection of observables and is specified by the wave function
representing the system. It is therefore common in the literature to use the term
eigenstate to indicate a state represented by an eigenfunction. The average value,
usually called the expectation value, of a sequence of measurements of an
observable represented by an operator A(r, −iℏ∇, t) on a system in a normalised
state described
by Ψ is, by definition, given by

A ≡ A¯ = ∫ Ψ*(r, t)A − Ψ(r, t) dτ. (1.57)

Frequently, the expectation value of a physical quantity represented by an operator

A is denoted by ⟨A⟩, i.e. the angle brackets are usually used for expectation values.
Moreover, it is customary to use the same symbol for a dynamical variable as well
as the operator which represents it. It may again be remarked that in quantum
mechanics, a physical system is represented by a wave function such that our entire
theoretical knowledge about it is contained in the wave function. Since the

1-
 Quantum
Mechanics

interpretation of the wave function is statistical in nature, laws of physics can only
make probabilistic predictions. They cannot predict the precise behaviour of a
system. In other words, if a physical quantity is measured a large number of times
by repeating an experiment under identical conditions or by performing a large
number of identical experiments, its average value can be predicted by the above
relation. It should be noted that, in general, the result of a single measurement will
not be given by ⟨A⟩. It is only the average value of a large number of
measurements, made in the manner suggested above, which is to be compared with
the theoretical value predicted by equation (1.57). It may be emphasised that laws
of physics are statistical in nature even when we are dealing with a single particle.
If the wave function is not normalised, then the expectation value ⟨A⟩ is defined
by the equation

∫ Ψ*(r, t)A Ψ(r, t) dτ

A
∫ Ψ*(r, t) Ψ(r, t) dτ
which reduces to equation (1.57) for a normalised wave function.
Let us now calculate the uncertainty in the measurement of a physical quantity
represented by an operator A when the system is represented by a normalised
eigenfunction Ψn of the operator A corresponding to the eigenvalue λn. Then
AΨn = λnΨn (1.58)
2
and the expectation values of A and A are given by

A = ∫ Ψ*(r, t)A Ψ(r, t) dτ = ∫ Ψ*(r, t)λ Ψ(r, t) dτ n

= λn ∫ Ψ*(r, t) Ψ(r, t) dτ = λ n

and

A2 = ∫ Ψ *n A Ψ 2
n dτ = λ 2 ,
n
where we have used (1.57) and (1.58) and the normalisation condition for Ψn.

Problem
1.5. The variance in the measurement of a physical quantity represented by an operator A is defined b
(A − A )2 .
Show that it is equal to

A2 – A 2 .

1-
 Quantum
Mechanics

In the measurement of a physical quantity represented by A, the uncertainty ΔA,

defined as the positive square root of variance, is given by

ΔA = +A¯ 2 − A¯ 2

= +λ¯ 2 − λ¯2
= 0.
This result shows that if a system is represented by an eigenfunction Ψn of the
operator A, the uncertainty in a measurement of the physical quantity represented
by A is zero and it should yield the eigenvalue λn of A in an individual measurement.
We may point out that the Schrödinger wave equation written in its Hamiltonian
form

Hψ (x) = Eψ (x)

shows that the constant E which is the eigenvalue of the Hamiltonian operator H
must be the energy of the system. This was assumed previously and is proved now
when we have become familiar with the basic concepts and assumptions of
quantum mechanics. The set of all the eigenvalues of the Hamiltonian operator H is
called the energy spectrum of H.
Let us next see what values of A would be observed if the system happened to be
in a quantum state Ψ which is not an eigenfunction of A. We will assume that the
eigenfunctions of any operator representing a physical quantity form a complete set
of functions. We express Ψ as a linear combination of the complete orthonormal
set of eigenfunctions Ψn of the operator A:

Ψ= ∑anΨn. (1.59)
n

Since Ψn are eigenfunctions of the operator A, we must have

AΨn = λnΨn. (1.58')

Then

A = ∫ Ψ*ΑΨ dτ = ∫ ∑am* Ψ*mA∑anΨn dτ

m n

= ∑∑am* an ∫ Ψ*mAΨn dτ
m n

= ∑∑am* anλn ∫ Ψ*mΨn dτ

m n

= ∑∑a m* a n λ n δmn.
m n

1-
 Quantum
Mechanics

Summing over m, we obtain

A = an*anλn
∑
n

= a1*a1λ1 + a2*a2 λ2 + ⋯ (1.60)

Now according to the normalisation condition, we have

∫ Ψ*(r, t) Ψ(r, t)τ dτ = 1,

or

∫ ∑am* Ψ*m ∑an* Ψ*n dτ = 1,

m n

or

∑∑am* an ∫ Ψ*m Ψn dτ = 1,
m n

or

∑∑am* anδmn = 1.
m n

Summing over m, we obtain

∑an*an = a1*a1 + a2*a2 + ⋯ = 1. (1.61)

n

Equations (1.60) and (1.61) suggest that if we measure the physical quantity,
represented by A, on a system in the normalised quantum state Ψ, then an* an may
be interpreted as the probability that a measurement will yield the eigenvalue λn. Of
course, the total probability that measurement will yield any one of these
eigenvalues is unity. When an eigenvalue, say λr, is p-fold degenerate, the probability
of finding the value λr is par* ar. The coefficient an is sometimes called a probability
amplitude. Since (1.60) involves only the eigenvalues of the operator A, the
probability that a measurement on the system will yield a value which is not an
eigenvalue of A is zero. We conclude that the measurement of an observable on a
system should always yield one of the eigenvalues of the operator representing the
observable. Since the observables are always real, the operators representing
the observables should be such that their eigenvalues are always real. Such
operators are called Hermitian operators. Thus in quantum mechanics, observables
are represented by Hermitian operators. It can be emphasised that when the
system is in a quantum state, the result of a single measurement on the system is
unpredict- able. It may also be stated that since (1.61) does not involve time, a
wave function which has been normalised will always remain normalised.
The constants an

1-
 Quantum
Mechanics

occurring in this equation can be evaluated as follows. Multiplying equation (1.59)

on the left by Ψm* and integrating with respect to τ, we obtain

∫ Ψ*m Ψ dτ = ∑an* ∫ Ψ*m Ψn dτ = ∑an* δmn = am. (1.62)

n n

To summarise: quantum mechanically, the system can be either in an eigenstate or

in a quantum state. If it happens to be in an eigenstate, i.e. in a state with a definite
eigenvalue, then a measurement will yield that particular value. On the other hand,
if the system happens to be in a quantum state, i.e. in a state which in general is a
linear combination of various eigenstates, i.e. it is in a state given by
a1Ψ1 + a2 Ψ2 + ⋯ + anΨn,
where Ψ1, Ψ2, …, Ψn represent the eigenstates of the system, then the measurement
must yield one of the eigenvalues of the corresponding operator. It may be
emphasised that there is no classical analogue of a quantum state.
It may be remarked that if before the measurement of a physical quantity
represented by an operator A the system is in a quantum state Ψ, then a measure-
ment can yield any one of the eigenvalues of A. However, when the measurement is
made on the system, a definite eigenvalue, say λn, is obtained. This means that the
very act of measurement has so disturbed the system that it has been carried from a
quantum state Ψ into an eigenstate Ψn.

ItProblem
must have been evident from the above analysis that it is not essential that a
system should always be in a quantum state in which a given quantity does not
1.6. What
possess will happen
a definite value. Itif can
a measurement on the system
be in an eigenstate is made to a definite
corresponding
eigenvalueimmediately
but this isafter that?
not the most general situation.

1.17 Phases and relative phases

Suppose that Ψ(r, t) is a solution of the time-dependent Schrödinger wave equation
corresponding to a definite value of energy. Since the differential equation is linear,
cΨ will also be its solution, where c is a complex number. The normalisation
condition for this function gives

c*c ∫ Ψ*Ψ dτ = 1.
Now the complex number c may be written as
c = c eiα,

1-
 Quantum
Mechanics

where the real α is called the phase of c while eiα is called its phase factor. Then
c*c = ∣c∣e−iα ∣c∣eiα = ∣c∣2. That is, the product c*c is independent of the phase. Since
the probability and expectation values of observables always involve Ψ*Ψ and
therefore the factor c*c, the corresponding expression is independent of the phase
value. We can therefore give it any value we like. It will not affect the physical result.
For convenience, we take a phase equal to zero. Then c may be considered as real.
This will not affect the physical results.
The situation takes a turn when we consider a system in a quantum state Ψ. Then
in terms of eigenstates which correspond to definite energies and are represented by
the wave functions Ψi, the function Ψ can be represented by
Ψ(r, t) = ∑ai Ψi(r, t).
i

To make the things simple, suppose that Ψ is a linear combination of two eigenstates
only:
Ψ(r, t) = a1Ψ1(r, t) + a2 Ψ2(r, t). (1.63)
Since each complex function can be written as Reiα, where R and α are real, we write
the functions a1Ψ1, a2Ψ2 as
a1Ψ1 = R1e iα1,
a 2 Ψ2 = R 2 e iα 2 .
Substituting these expressions in equation (1.63), we obtain
Ψ(r, t) = R1e iα1 + R 2 e iα 2 .
Therefore:

( )( )
Ψ*Ψ = R1e−iα1 + R 2 e−iα 2 R1e iα1 + R 2 e iα 2 .
= R 2 + R 2 + 2R1R2 cos(α1 − α2 ).
1 2
The above analysis shows that although the overall phase in Ψ can be ignored, the
relative phase is important.

1.18 Postulates of quantum mechanics

For our convenience, let us now collect together the various postulates which have
been proposed while relating solutions of the Schrödinger wave equation for a
physical system with measurable quantities:
1. Every physically acceptable solution of the Schrödinger wave equation for a
system represents a state of the system. Since a linear combination of the
solutions of the Schrödinger wave equation is a solution of the differential
equation, it should also represent a state of the system. This is known as the
principle of superposition.
2. To every observable, there corresponds an operator A. In particular, −ih∇
and ih∂/∂t are energy and momentum operators.

1-
 Quantum
Mechanics

3. The nature of the operator should be such that its eigenvalues are always
real. Such an operator is called a Hermitian operator.
4. The measurement of a physical quantity represented by a Hermitian
operator A must yield one of the eigenvalues of the operator A.
5. The measurement of a physical quantity, represented by an operator A, on a
system in an eigenstate always gives the eigenvalue of A corresponding to
that eigenstate.
6. The average value of a large number of measurements of a physical quantity
on a system in an arbitrary state Ψ, a solution of the Schrödinger wave
equation, yields the average or expectation value ⟨A⟩ given by

A= ∫ Ψ*ΑΨ dτ
provided that Ψ is normalised and there exist suitable boundary conditions.
7. The probability of finding a particle represented by the wave function Ψ(r,
t) in a small volume dτ about the point r and at time t is Ψ*(r, t) Ψ(r, t)
dτ.

1.19 The Schrödinger wave equation under space reflection,

space inversion and time reversal
We will now briefly examine the behaviour of the Schrödinger wave equation
under space reflection, space inversion and time reversal.

1.19.1 Invariance under space reflection

Consider the time-independent Schrödinger wave equation in one-dimensional
space:
d2ψ (x) 2m
+ (E − V(x))ψ (x) = 0. (1.64)
dx2 ℏ2
Changing x to −x throughout the above equation, we obtain
d2ψ (−x) 2m
+ (E − V(−x))ψ (−x) = 0. (1.65)
2
dx 2
ℏ
If V(x), the potential energy of the system, is an even function of x, i.e. the function
does not change by changing the sign of the coordinate x, then V(−x) = V(x) and
the above equation reduces to
d2ψ (−x) 2m
+ (E − V(x))ψ (−x) = 0. (1.66)
dx2 ℏ2
Equations (1.64) and (1.66) show that if V(x) is an even function of x, then ψ(x)
and ψ(−x) are both solutions of the same amplitude equation. But, as shown in figure
1.3, a change in the sign of x is equivalent to a reflection of x in a plane mirror
passing through the origin and placed normal to x. Thus, ψ(−x) is the mirror image
of ψ(x)

1-
 Quantum
Mechanics

Figure 1.3. Mirror image.

and both are solutions of the same amplitude equation, provided V(−x) = V(x).
Hence, the amplitude equation with potential energy as an even function of the
space coordinate is invariant under reflection.

1.19.2 Invariance under space inversion

The analysis can easily be extended to three-dimensional space. Then it is called
invariance under space inversion.

1.19.3 Invariance under time reversal

Let us next consider the time-dependent Schrödinger wave equation
∂Ψ(r, t)
iℏ = H Ψ(r, t), (1.16b′)
∂t
where H = T + V(r) such that the potential energy V(r) is a real function of space
coordinates alone. Taking the complex conjugate of equation (1.16b′), we obtain
∂Ψ*(r, t)
−iℏ = H Ψ*(r, t). (1.67)
∂t
Changing t to −t throughout the equation, we obtain
∂Ψ*(r, −t)
iℏ = H Ψ*(r, −t). (1.68)
∂t
Equations (1.16b′) and (1.68) show that, provided the potential energy, and
consequently the Hamiltonian, is real and does not depend upon time explicitly,

1-
 Quantum
Mechanics

the wave functions Ψ(r, t) and Ψ*(r, −t) are both solutions of the same
time- dependent Schrödinger wave equation. The wave function Ψ*(r, −t) is often
called the time-reversed solution of the time-dependent Schrödinger wave
equation with respect to Ψ(r, t). The time-dependent Schrödinger wave equation
(1.16b′) is said to be invariant under time reversal. This means that if Ψ(r, t) is a
solution of the time- dependent Schrödinger wave equation, then Ψ*(r, −t) is also a
solution of the same. If the system is in a stationary state with definite energy E,
then we may write
⎛ iEt ⎞
Ψ(r, t) = ψ (r) exp ⎜ − ⎟. (1.69)
⎝ ℏ⎠

Taking the complex conjugate of the above equation, we obtain

⎛ iEt ⎞
Ψ*(r, t) = ψ *(r) exp⎜ ⎟.
⎝ℏ ⎠

Changing t to −t throughout the equation, we obtain

⎛ iEt ⎞
Ψ*(r, −t) = ψ *(r) exp⎜ − ⎟. (1.70)
⎝ ℏ⎠

Thus invariance under time reversal implies that, for a stationary state, if ψ(r) is a
solution of the amplitude equation, then ψ *(r) is also a solution of the same.

1.20 Concluding remarks

We conclude the first chapter by emphasising that the Schrödinger wave equation
is not a model for the explanation of some experimental results. It is a law of nature
and has replaced Newton’s second law of motion for non-relativistic mechanics. It
therefore provides the foundation for the explanation of mechanical phenomena.
However, it is in the form of a differential equation and is not garbed in words. Its
predictions based on a probabilistic interpretation of its solution are consistent with
experiments although sometimes they appear to be at variance with common sense.
But what is common sense? It is the knowledge that we gain from our everyday
experience with macroscopic objects. And why do we expect that the microscopic
world would exhibit the same characteristics and behave the same way? If we
define common sense as the knowledge gained from the behaviour manifested by
micro- scopic entities, quantum mechanics becomes fully consistent with common
sense! The theory is beautiful and elegant and its achievements are spectacular and
fantastic. It has introduced new concepts and changed the pattern of philosophical
interpretations. We have established the Schrödinger wave equation and interpreted
its solutions so as to relate them to physically measurable quantities. Our next
objective is to apply this theory to solve physical problems in mechanics and
during this process enhance our understanding of various characteristics of
mechanical phenomena.

1-
 Quantum
Mechanics

Additional problems
1.7. Why in the Compton effect is an x-ray photon scattered by an electron
with a change in wavelength while in the photoelectric effect an optical
photon transfers all its energy to the photoelectron?
1.8. The classical expression for the potential of a linear harmonic oscillator
is 1/2kx2, where k is a constant and x is the displacement of the
oscillator from its mean position. Can the Schrödinger amplitude
equation be used for solving the problem of a linear harmonic
oscillator?
1.9. A particle is somewhere on a line of about 7 m in length. Calculate the
uncertainty in the measurement of its linear momentum. What con-
clusion can be drawn from this result?
1.10. What is the Compton effect? Give its quantum theory and spell out its
significance.
1.11. What is the energy–time uncertainty relation? If an electron in an atom
takes about 10−8 s to emit radiation in falling from an excited state to a
lower energy state, calculate the energy uncertainty of such an excited
state.
1.12. State Heisenberg’s uncertainty relation for position and momentum.
Describe an idealised experiment to illustrate the uncertainty relation.
1.13. Show that the probability current density can be expressed as
ℏ
j=− (
Re iΨ*∇Ψ . )
m
1.14. Show that if Ψ is real, then the vector j(r, t) vanishes.

1-