lOMoARcPSD|36070675

Classical Physics

Physics for Scientists and Engineers (University of Western Australia)

Studocu is not sponsored or endorsed by any college or university

Downloaded by Oppa Fest ([email protected])
 lOMoARcPSD|36070675

The Breakdown of Classical Physics

and the rise of quantum physics

Lecture Notes

Darren T Grasso ➞ 2022

[email protected]

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Contents

1 Introductory remarks 1

2 Broad historical overview 2

3 Background 6

4 The failure of classical physics 9

5 The ultraviolet catastrophe 11

6 The photoelectric effect 17

7 The Compton effect 20

8 Atomic spectra and the Bohr atom 25

9 The de Broglie hypothesis and matter waves 28

10 Bohr’s atom and matter waves 30

11 X-ray and electron scattering from crystals 32

12 Probability in physics: the double slit experiment with electrons 35

ii

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

1 Introductory remarks

These notes are designed to accompany my lectures on the Breakdown of Classical Physics,
and summarise almost all of what is said during those lectures. The material which will
not be found here will appear in the ‘Supplementary Material’ column in the ‘BoCP
Lecture Material’ file in this course’s LMS unit (www.lms.uwa.edu.au).

To be absolutely clear on what is expected of you, you are only required to be able to solve
the types of problems you’ll find in the practice question sets and assignments. The test
and examination questions will consist of problems only of this kind. There are a lot of
dates, names and historical information in those notes, you are not expected to memorise
them. You are only expected to be able to solve problems using the central ideas, and
the equations found on the equation sheet.

Throughout these notes, any terms appearing in bold are being defined. You should pay
special attention to such terms.

Finally, these notes are new (written in Semester 1 2021), so if you notice any typos,
find anything unclear or ambiguous, or feel that there is something missing please let me
know.

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

2 Broad historical overview

Prior to the beginning of the 20th century – and persisting for some time into the early
1900’s – all physicists believed that the universe was governed by a set of classical laws,
where one event causes the next, and that everything unfolds in an intuitive and com-
pletely predictable fashion. These laws were Newton’s laws of motion and Maxwell’s equa-
tions of electromagnetism. However, by around 1930 something had drastically changed,
and many if not most members of the physics community had become convinced that at
a fundamental level the laws of the universe had an unpredictable and uncertain element
to them. Many physicists came to believe that certain events were not directly caused by
anything at all, and that there existed fundamental limits on what one can actually know
about a physical system. Starting from around the year 1900, it slowly dawned on physi-
cists that something was fundamentally wrong with their current ‘classical physics’ view
of the universe. Over the next few decades the old classical physics view of the universe
was swept aside, being replaced by the more fundamental and the highly non-intuitive
quantum physics.

What actually happened in the early 1900’s to trigger this radical shift on how physicists
view the world is the subject of these lectures. They are a very brief introduction to some
of the history of physics and how we came to be where we are now. Our main focus will
be on the experiments and evidence that showed that classical physics is inadequate at
the fundamental level, and on the key ingredients of quantum physics.

At first, some aspects of quantum physics appear truly bizarre. For example, under
some circumstances a single particle can behave like it’s a wave and somehow spread out
and interfere with itself. And under other circumstances a wave can exhibit particle-like
behaviour. By about 1930 the exact rules of quantum mechanics had been clearly written
down, as clearly as Newton’s three laws of motion. But these rules – which physics
students will study in a future physics unit – require some level of familiarity with the
history of physics if they are to make any kind of sense. Quantum mechanics is quite
mysterious and without some appreciation of history, one has little hope of understanding
it.

To better appreciate the chronology, for some broader historical context, and as an

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

overview of what we will be covering in these notes, below is a list of some of the most
significant events in the history of physics (those which are pertinent to our story). Note
that those items in the list which carry an asterisk will be discussed in greater detail later
on, so do not be concerned if you have not heard of them yet. This list will perhaps be
more useful to you on a second reading. Also note that some of the following dates are
approximate.

1610: Johannes Kepler publishes his three laws of planetary motion. This is a significant
step forward in the history of science.

1687: Isaac Newton publishes ‘Principia’ (Mathematical Principles of Natural Philos-

ophy). In Principia Newton formulates his three laws of motion and universal
gravitation. He mathematically derives Kepler’s laws of planetary motion.

1840: Michael Faraday introduces the concept of fields into physics.

1865: James C. Maxwell publishes A Dynamical Theory of the Electromagnetic Field.

This work includes the final form of Maxwell’s equations and completes the classical
theory of electromagnetism. Using his equations Maxwell shows that light is made
up of electromagnetic waves.

*1900: The cracks in classical physics begin to appear. The so-called blackbody spectrum
– the spectrum of electromagnetic radiation in thermal equilibrium with matter at
some given temperature – had been measured prior to 1900. All attempts to derive
the formula which describes the observed spectrum – attempts based on classical
physics – had failed. In 1900 Max Planck used guesswork to arrive at a formula
which precisely agreed with the measured spectrum. Although originally Planck did
not derive his formula, it is not long before he managed to do so. In his famous
derivation Planck simply postulates that in equilibrium, matter absorbs and emits
electromagnetic radiation in discrete units. His postulate is justified only by the
fact that it worked.

*1905: Einstein further suggests that light itself only exists as particles with specific
energy and momentum. This is a step further than Planck, for Planck applied his

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

postulate to the matter that was absorbing and emitting, rather than to the radia-
tion itself. Einstein uses his idea to explain the photoelectric effect. Many physicists
remained skeptical about Einstein’s idea that light comes in packets, a skepticism
that was only later extinguished by the scattering experiments of Compton (1923).

1911: Based on the results of his famous gold foil experiment, Ernest Rutherford deter-
mines the structure of the atom: negatively charged electrons orbiting a very dense
positively charged nucleus. This model, although well supported by the experimen-
tal data, gives rise to a new mystery that classical physics cannot explain: how
is the atom stable? The classical theory of electromagnetism predicts that accel-
erating charges emit electromagnetic radiation, so the electrons in orbit around a
nucleus should continuously emit radiation, lose energy, and ultimately spiral into
the nucleus. How is it that the electrons remain in orbit and do not continuously
emit radiation?

*1913: Niels Bohr, faced with trying to explain the stability of the atom and the discrete
energy emissions observed in atomic spectra, guesses that the electron’s orbital
angular momentum might only be able to come in discrete units. This guess, along
with some classical physics reasoning, leads to the well known Bohr model. It gives
very close agreement with the observed hydrogen spectrum and provides some kind
of explanation as to why the atom is stable: the electrons are only ‘allowed’ to orbit
at very specific distances and cannot spiral inward like classical physics models
predict.

*1923: Arthur Compton further demonstrates that electromagnetic radiation can some-
times behave like particles when interacting with electrons. He gives an explanation
of billiard ball-like collisions – photons bouncing off electrons – for what is now
known as the Compton effect (or Compton scattering).

*1923: Louis de Broglie reasons as follows: since light is thought to be a wave, but clearly
can manifest particle-like behaviour like the photoelectric effect, particles may re-
ciprocally manifest wave-like behaviour. For the first time in history matter begins
to be described by wavefunctions, just as waves traditionally were. Interference-like
patterns from electron scattering by crystals later confirm that electrons can indeed

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

manifest wave-like properties (1927). Much later (in the 1960’s and 1970’s) dou-
ble slit experiments are directly performed using electrons, further confirming de
Broglie’s hypothesis. The idea that waves can behave like particles, and viceversa,
under some circumstances became known as wave-particle duality.

*1926: Born notices that treating particles as waves comes with a number of conceptual
difficulties. Thinking about particle detection in scattering experiments (where you
only ever detect a whole particle, never some fraction of a particle), Born reasons
that particles do not actually spread out like waves. He is thus led to postulate that
the square of the wavefunction describing a particle is related to the probability of
its detection, rather than an intensity as it would be for normal waves.

1926 onward: It is not long before quantum mechanics is soon distilled into a set of
clear rules and its mathematical structure examined in detail.

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

3 Background

In an attempt to avoid any potential confusion, before we begin let’s define and clarify a
few terms.

From a modern physicist’s point of view our theories and models of the universe fall into
one of the following broad categories: (i) classical physics; or (ii) quantum physics.

Roughly speaking, classical physics theories describe – or are useful for describing –
what happens in macroscopic systems. By ‘macroscopic’ we mean large or everyday
length scales. Our successful classical theories include all of the following:

❼ Newtonian or ‘classical’ mechanics.1

❼ Electromagnetism (i.e. Maxwell’s equations).

❼ Special relativity and relativistic mechanics. A classical physics account of

particles, taking into account the fact that the universe has a fundamental speed
limit, the speed of light c. It effectively replaces Newtonian mechanics. In the
limit where the speeds involved are much less that the speed of light, relativistic
mechanics reduces to Newtonian mechanics.

❼ General relativity (i.e. Einstein’s field equations). This is a classical field theory
description of gravitation, taking into account the fact that the universe has a
fundamental speed limit c. It describes gravity as the curvature of spacetime.

All classical physics theories and models have something in common: they are all de-
terministic. By a deterministic theory or model we mean that there is no element of
randomness in them, and that given the initial state of a system the future (and past) of
the system is completely fixed or determined. For example, in Newtonian mechanics, if
you are given all information about a system at some initial time – i.e. if at time t = 0
you know the positions, masses and velocities of all particles – then you can predict with
absolute certainty where those particles will be and how they will be moving at any time
1
When you see the word ‘mechanics’ you should think of particles. So Newtonian mechanics is about
particles and how they interact. In contrast, we also have ‘field theories’. Electromagnetism is an example
of a field theory, a theory about electric and magnetic fields and how they behave.

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

t in the past or future. In short, for classical theories, before we carry out an experiment,
if we have enough data, we can predict precisely what will happen.

In contrast, quantum physics theories describe what happens in systems on atomic or

microscopic length scales. This includes:

❼ Quantum mechanics. This is a theory which describes a system consisting of

some fixed number of particles and how it evolves according to the rules of the
universe at atomic length scales.

❼ Quantum field theory. This is the mathematical framework of the standard

model of particle physics: the quantum version of a classical field theory.

The technical difference between quantum theories and classical theories is somewhat
complicated, but in short all quantum physics theories are indeterministic. By an inde-
terministic theory we mean that there is an element of randomness in them, and that
given the initial state of a system, the future of the system is inherently unpredictable.
In quantum theories the very best we can do is predict the probabilities of the various
outcomes of experiments.

Another stark difference between classical and quantum theories is related to the fun-
damental rules and mathematics on which each are based. In a loose sense, in classical
physics theories everything is always perfectly smooth and continuous. For example, there
are no abrupt jumps or changes when a classical system evolves in time, and there are no
limitations on what energies and momenta particles can have (they are allowed to take
on a continuum of values). In contrast, in quantum physics, frequently there are limita-
tions on what energies and momenta particles can have: often they are only allowed to
have specific discrete values and nothing in between (like the allowed energy levels in the
hydrogen atom). Additionally, the evolution of quantum systems often involves abrupt
jumps and changes as the system leaps from one allowed energy level to another.

For the purposes of clearly understanding the story presented here, it is vitally impor-
tant to understand two central predictions of classical electromagnetism as described by
Maxwell’s equations:

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

1. Light is an electromagnetic wave, consisting of oscillating electric and magnetic

fields.

2. Accelerating charge particles emit electromagnetic waves.

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

4 The failure of classical physics

Around the year 1900 physicists started to realise classical mechanics and classical elec-
tromagnetism do not work on atomic length scales. At these scales a new theory, later
called quantum mechanics, is required for accurate predications. The key experiments
that exposed the inadequacy of classical physics were those which:

1. Demonstrate that electromagnetic radiation consists of particles (now called pho-

tons);

2. Demonstrate that matter particles can exhibit wave-like properties (like diffraction
and interference) under appropriate circumstances.

The clearest evidence for the existence of photons include the following experiments and
theoretical considerations. For each, classical physics failed to account for the experimen-
tal data:

1. Around 1900: The blackbody spectrum.

2. Around 1905: The photoelectric effect.

3. Around 1923: Compton scattering.

The following experiments and theoretical considerations are the clearest evidence that
under certain circumstances matter particles can be behave like waves:

1. In 1913 Niels Bohr developed a simple model of the hydrogen atom, but did not
provide a reason for why it worked. Later, based on the idea that electrons can
behave like waves, a more satisfying explanation of Bohr’s model was given.

2. From around 1923 to 1927 Clinton Davisson and Lester Germer carried out what
are now know as the Davisson–Germer experiments. In these experiments electrons
where scattered by the surface of a crystal of nickel metal and they clearly displayed
diffraction (interference) patterns.

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Collectively, the above experiments and evidence shows that all fundamental entities have
a ‘dual’ nature: they could sometimes behave as waves and sometimes behave as particles,
depending on the experiment. This came to be know as wave-particle duality and there
is nothing in classical physics that can account for this.

We will now examine each of the above experiments and the corresponding theoretical
considerations.

10

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

5 The ultraviolet catastrophe

In an earlier part of this unit – on the topic of Heat and Thermodynamics – we learned
that an object at any temperature will emit electromagnetic or thermal radiation from
its surface. The equation we encountered was the so-called Stefan-Boltzmann law:

Pem = eσAT 4 Stefan-Boltzmann law (5.1)

where: Pem is the total rate at which the object will emit energy via electromagnetic
radiation (in watts); A is the object’s surface area (in square metres); T is the object’s
temperature (in kelvin); σ = 5.67 × 10−8 W m−2 K−4 is the so-called Stefan-Boltzmann
constant; and e is the emissivity of the object’s surface. The emissivity is a dimension-
less number which has a value between 0 and 1 depending on the specific nature of the
surface of the object. A perfect emitter – a so-called a blackbody or backbody radia-
tor – has an emissivity of e = 1. When we encountered the Stefan-Boltzmann law in Heat
and Thermodynamics we glossed over the fact that hot bodies actually emit a continuous
distribution of wavelengths from all portions of the electromagnetic spectrum. The above
equation is just the total rate of energy emitted, it doesn’t say anything at all about how
much energy is emitted at a specific wavelength. It turns out that at room temperature
objects emit most of their thermal radiation in the infrared part of the spectrum, and it
is for this reason that we cannot see them in the dark. In contrast, a hotter object like
molten rock from a volcano emits a significant amount of its thermal radiation in the red
part of the visible spectrum, which is why lava glows red.

The perfect thermal emitters – the blackbody radiators – are idealised objects that do not
actually exist in nature, but they are an extremely useful approximation for theoretical
study. It turns out that their surface properties not only make them perfect emitters, but
also perfect absorbers of thermal radiation: that is, they will absorb all light that falls
upon them. Thus, a perfect absorber would appear black since it does not reflect any
light at all. It is for this reason that these idealised bodies are called blackbodies.

The exact characteristics of the light that is emitted by an object (i.e. exactly how much
radiation and at what frequency) depends on the specific nature of the surface. It is the
details of the surface of an object which determines its emissivity e.

11

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

An extremely good approximation of a idealised blackbody radiator with e = 1 is a tiny

hole which leads to the interior of a hollow object or cavity such as that shown in Figure
1.

Figure 1: A cavity with a small hole. This is a good approximation for a

perfect absorber since any light entering the cavity (a light ray is shown in
red) bounces around inside the cavity until it is completely absorbed.

It is not difficult to convince yourself that any light that is incident on a hole that leads
to the inside of a hollow object will not be reflected. The light that does enter the cavity
will bounce around inside. At each reflection from the inside surface the light is partially
absorbed by the walls of the cavity and partially reflected, and eventually all of the light
is absorbed by the cavity walls. Any light that does escape through the hole does not
actually depend on what material the inside walls of the cavity is made from, it only
depends on the temperature of the cavity walls. So in studying the actual electromagnetic
emission from such a cavity, we are effectively looking at the thermal radiation of a
blackbody.

During the late 1800’s many experiments were performed that mapped the exact char-
acteristics of radiation being emitted from cavities with small holes. In particular, the
intensity of light was measured at various wavelengths and different cavity temperatures.
And so, prior to 1900 the intensity versus wavelength plots of blackbody radiation at
different temperatures were well know. Recalling that intensity I is a measure of how
‘bright’ the light is – more precisely the rate of energy emitted per unit area – Figure
2 shows a few plots of the intensity I of blackbody radiation as a function of the wave-
length. In other words, Figure 2 shows the spectrum of blackbody radiation at three
different temperatures.

12

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Figure 2: The blackbody spectrum (intensity as a function of wavelength) at

three different temperatures. Credit: Physics for Scientist and Engineers 4th
ed. by Serway

Extensive study of the emission spectrum from these cavities showed that: (i) the total
power being emitted was given by the Stefan-Boltzmann law equation (5.1); and (ii) at a
given temperature T (in kelvin) the wavelength λmax (in metres) at which the maximum
intensity emissions are observed is given by the relationship

λmax T = 2.898 × 10−3 m K Wien’s displacement law (5.2)

which came to be known as Wien’s displacement law.

The number appearing on the right-hand-side of Wien’s displacement law was empirical,
meaning that it was deduced directly from experimental data. This law states that the
wavelength of maximum intensity is inversely proportional the temperature: the higher
the temperature the smaller the wavelength of maximum intensity (as can be seen from
Figure 2).

To a very good approximation the sun is a blackbody radiator with a temperature of

around 5800 K. Wien’s displacement law then tells us that the intensity of the sun’s
emissions is peaked around λmax = 500 nm which falls right in the middle of the visible
spectrum (which is about 380 nm to about 750 nm).2
2
Of course this is no accident: we have evolved to exploit the ‘brightest’ part of the sun’s spectrum.

13

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

It is important to note that classical electromagnetism (i.e. Maxwell’s equations) tells

us that accelerating electric charges emit electromagnetic radiation. And so, from a
classical physics point of view, bodies emit thermal radiation simply because they consist
of accelerating charged particles (i.e. they are made of atoms – charged electrons and
nuclei – in random thermal motion). However, when we carefully try to use classical
electromagnetism to predict the intensity I(λ, T ) of the blackbody spectrum as a function
of temperature T and wavelength λ, we fail spectacularly, and arrive at the following
expression known as Rayleigh-Jeans law:

2πckT
I(λ, T ) = Rayleigh-Jeans law (5.3)
λ4

where c is the speed of light and k = 1.381 × 10−23 J K−1 is Boltzmann’s constant.
Rayleigh-Jeans law does not agree with the experimental measurements of the blackbody
spectrum. For a comparison see Figure 3 which shows both the experimental data of in-
tensity versus wavelength for a blackbody at a given temperature T , and Rayleigh-Jeans
law at that same temperature.

Figure 3: A plot of the experimental blackbody data (red curve) at some

temperature, and at the same temperature the classical physics prediction (the
blue curve) known as Rayleigh-Jeans law. For short “ultraviolet” wavelengths
the intensity of the classical prediction tends to infinity, indicating that infinite
energy is being emitted – a feature known as the ultraviolet catastrophe.
Credit: Physics for Scientist and Engineers 4th ed. by Serway

14

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Rayleigh-Jeans law only gives agreement with the actual blackbody spectrum data for
long wavelengths. It disagrees at shorter wavelengths. In fact, from the functional form
of Rayleigh-Jeans law we see that it predicts that the intensity approaches infinity as
the wavelength approaches zero. The total area under the intensity curve is related to
the total power emitted, and so Rayleigh-Jeans law predicts that an infinite amount
of energy should be emitted by any object. This is clearly false. The failure of classical
electromagnetism and Rayleigh-Jeans law to give an accurate description of the blackbody
spectrum became known as the ultraviolet catastrophe. A ‘catastrophe’ because it
makes the catastrophically bad prediction that an infinite amount energy will be emitted,
and ‘ultraviolet’ because ultraviolet wavelengths are short, and it is in the short wavelength
part of the spectrum that large disagreement between experiment and theory are observed.

In 1900 Max Planck, being very familiar with the experimental blackbody spectrum,
noticed that the data could be fit with the following formula:

2πhc2
I(λ, T ) = Planck’s law (5.4)
λ5 (ehc/(λkT ) − 1)

where h = 6.626×10−34 Js is known as Planck’s constant, and was simply the parameter
Planck needed for his formula to fit the experimental data. He arrived at the above formula
by mere guesswork, but it just happened to fit the data perfectly at all wavelengths and
all temperatures. It was not long – just months – before Planck had managed to work out
a way to mathematically derive the above formula from first principles. In his derivation
he assumed that the blackbody radiation came from atoms within the walls of the cavity
oscillating back an fourth – so-called oscillators. But in order to arrive at his formula
above he found he was forced to make two radical assumptions about these oscillators,
two assumptions that where fundamentally inconsistent with classical physics:

Assumption 1: The energy of an oscillator can only have certain discrete values En :

En = nhf (5.5)

where: n = 1, 2, 3, . . . (i.e. n is a positive integer); f is the frequency of the os-

cillation; and h is Planck’s constant (the constant he introduced earlier). From a
modern perspective we would say that the energy of the oscillator is quantised –
meaning that it can only come in discrete units of a fixed size – and the number n

15

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

is a example of a quantum number which tells you the quantum state of the
oscillator.

Assumption 2: An oscillator can only emit or absorb energy in units of

E = hf . (5.6)

In short, Planck’s derivation assumed that the matter in the walls of the cavity only
absorbs and emits electromagnetic radiation in discrete packets. This assumption was an
act of desperation, justified only by the fact that it worked. This was the first example of
a quantum model in the history of physics, and there was no physical grounds or previous
experience that suggested that this kind discrete behavior should occur in nature.

The important point to note here is that this assumption was simply inconsistent with
classical physics, and Planck’s successful derivation showed that classical physics was
simply not adequate for predicting the correct behaviour of this physical system. This
was the first time in the history of physics that energy quantisation appears, and it marks
the birth of quantum physics.

16

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

6 The photoelectric effect

The quantisation assumptions made by Planck in 1900 when he derived his famous law
applied only to the matter in the walls of the cavity which emit and absorb the radiation,
not the actual radiation itself. More specifically, Assumption 2 and equation (5.6) was
actually about atomic oscillators, not about light. Planck did not go so far as to assume
that the actual radiation was made up of distinct particles with discrete energies, he only
thought of the matter in the cavity walls as having the peculiar property of emitting and
absorbing discrete chunks of energy. It was later explained by the famous physicist George
Gamow that Planck thought that light was like butter: butter itself can be separated into
any quantity, but it is only ever bought and sold in multiples of one quarter pound.

It was finally Einstein in 1905 who took the further step and boldly suggested that the
radiation itself consists of particles only – now known as photons – and that light with
frequency f is just made up of many discrete particles each with energy

E = hf . Energy of a photon (6.1)

This suggestions is fully consistent with Planck’s assumption (5.6).

Einstein used this idea to account for the experimental observations of a phenomenon
known as the photoelectric effect, which had so far defeated any classical physics based
attempt to explain it.

In the later part of the nineteenth century, very careful experiments showed that when
light was incident on certain metallic surfaces, as shown in Figure 4 below, electrons
would be emitted from those surfaces. This is known as the photoelectric effect and
the ejected electrons are referred to as photoelectrons.

These experiments showed that the photoelectrons have the following key properties:

1. The kinetic energy K of the photoelectrons have a range of values, but it is always
less than some maximum value Kmax (which depends on the particular metal).

2. The maximum kinetic energy of the ejected electrons Kmax is independent of light
intensity. The intensity is proportional to the number of electrons emitted from the
surface.

17

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

ejected
electrons
photons

metal

Figure 4: The photoelectric effect. Photons of sufficiently high frequency are

incident on a metallic surface and eject electrons (known as photoelectrons).

3. The maximum kinetic energy Kmax of the ejected electrons is proportional to the
frequency f of the electromagnetic radiation, and below some cutoff frequency fc
no electrons are ejected no matter how intense the light. This is shown in Figure 5
below.

Kmax

f
fc
cutoff frequency

Figure 5: For a given metal, a plot of the maximum kinetic energy Kmax of
photoelectrons as a function of the frequency f of the incident light. Below
a certain ‘cutoff’ frequency fc , no electrons are ejected from the surface no
matter how intense the light source.

It was well known that the characteristics of the photoelectrons could not be explained
with a classical view of light as an electromagnetic wave. In particular, classical elec-
tromagnetism makes the following predictions about what should happen, predications

18

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

which are at complete odds with the experimental observations:

1. The kinetic energy of the photoelectrons should depend only on the intensity of light.
The more intense the light, the more kinetic energy the electrons should carry.

2. There should be no relationship between the frequency of the light and the kinetic
energy of the electrons. Electrons should be ejected at any frequency of incident
light, provided the intensity is sufficiently enough.

Einstein’s contribution – building on Planck’s ideas – was to assume the following:

1. Radiation consists of particles of light called photons, travelling at a speed c.

2. The energy E of a photon is given by E = h f where f is the frequency of the

incident light and h is Planck’s constant.

3. The intensity of light is proportional to the number of photons per unit volume.
The more photons, the more intense the light.

With these assumptions Einstein gave a very simple explanation of the photoelectric effect.
An electron in the metal will absorb an incoming photon with frequency f (and energy
E = hf ), and will then be ejected with a kinetic energy K given by

K = h f − W, Kinetic energy of photoelectrons

where W is the binding energy of the electron in the metal. The maximum kinetic
energy Kmax of the ejected electrons occurs for the most loosely bound of the electrons.
We define Φ as the binding energy of the most weakly bound electrons – also
known as the metal’s work function – and so

Kmax = h f − Φ . Maximum kinetic energy of photoelectrons

If then follows that if the frequency of the incident light is small enough that h f < Φ,
then there is insufficient energy to free any electrons from the surface. Thus, the cutoff
frequency fc is given by the condition that Kmax = 0, and
Φ
fc = . Cutoff frequency
h
Finally, increasing the intensity of the light just means that there will be more photons
incident on the metal, and therefore more electrons ejected.

19

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

7 The Compton effect

In the realm of classical physics, waves and particles are completely distinct kinds of
‘thing’, they are quite unlike one another. Waves are spread out and particles have definite
locations. As strange as it may seem, in quantum physics this distinction is blurred, and all
‘things’ have a dual nature: they can manifest either wave-like or particle-like properties,
depending on the circumstances.

Generally speaking, waves are characterised by their two central properties: their fre-
quency f and wavelength λ. In contrast, particles may be characterised by their two
properties of energy E and momentum p~. In quantum mechanics the wave properties and
particle properties of a ‘thing’ are deeply connected.

By around 1919 Einstein had concluded that although classical electromagnetic radiation
could behave as a wave with a frequency f and wavelength λ (related by c = f λ), it also
could behave as if it was made up of particles with energy
hc
E = hf = Energy of a photon (7.1)
λ
and a momentum of magnitude
E
p= . Momentum of a photon (7.2)
c
By inserting (7.1) into (7.2) one could readily also establish a relationship between the
wavelength of a photon and its momentum
h
p= . Momentum of a photon in terms of wavelength (7.3)
λ
The confirmation of this last relationship between the momentum of the photon and its
wavelength was experimentally demonstrated in 1923 in two independent experiments,
one by Peter Debye and another by Arthur Compton. The experiment came to be known
as Compton scattering.

The Compton scattering experiment is quite simple to describe (although, like most ex-
periments, technically difficult to perform). In this experiment light in the form of x-rays
with a wavelength λ0 is directed towards a stationary free electron, as shown in the Figure
6. The light that is scattered from the electron is then detected by an x-ray detector at
various scattering angles θ.

20

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

scattered radiation

incident radiation
θ
(wavelength λ0 )

electron detector

Figure 6: Compton scattering. X-rays with wavelength λ0 are incident on a

free stationary electron. The radiation scattered from the electron at scattering
angle θ is received by a detector.

When this experiment is carried out at various angles θ, x-rays arrive at the detector with
two distinct wavelengths:

(i) a wavelength λ0 – the same wavelength as the original incident radiation;

(ii) a wavelength λ′ which is longer than the original incident radiation, i.e λ′ > λ0 .

The results of the experiment are shown in Figure 7 which is a sequence of plots of
the intensity of detected radiation as a function of the wavelength measured at different
scattering angles . The data3 reveals the following relationship:

λ′ − λ0 ∝ 1 − cos θ . Experimental conclusion (7.4)

Classically speaking one should expect only one wavelength to arrive at the detector
at any scattering angle, a wavelength of λ0 . This is because, considering light as an
electromagnetic wave with electric and magnetic fields oscillating at a frequency f0 = c/λ0 ,
when it passes by the electron the oscillating electric field induces the electron to oscillate
at the frequency f0 . An oscillating charge is accelerating, and it will emit electromagnetic
3
Note that the ‘free stationary electrons’ in the actual experiment are the outer electrons of a carbon
atom and they are moving around a little (but are at rest on average). It is for this reason that the peaks
in the plots are spread out.

21

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Figure 7: Compton scattering data: plots of the intensity of received radiation

as a function of the received wavelength at 4 different scattering angles. Two
distinct wavelengths, λ0 and λ′ are detected at all non-zero scattering angles
θ. Credit: Physics for Scientist and Engineers 4th ed. by Serway

radiation at the frequency of its vibration. Thus, classically the electron should only emit
radiation at the same frequency f0 and wavelength λ0 as the original incident radiation.4
So classically speaking there is simply no explanation for the observation of scattered
radiation with a wavelength λ′ .

An incredibly simple particle explanation for this phenomena was given by Compton: he
treated the experiment as if it were a two-particle collision in which the total relativistic
momentum and energy are both conserved. One of the particles is the photon, the other
is the electron. This is just a billiard ball-like collision, see Figure 8 for a sketch.

Before the scattering event the electron is at rest and the incident x-rays have wavelength
λ0 and frequency f0 . So in terms of a particle collision (using Einstein’s equations (7.1)
and (7.2)) we have:

hc
Photon before collision: The energy of the photon is E0 = hf0 = λ0
and the magni-
h
tude of its momentum is p0 = λ0
.
4
We are overlooking a complication here. Classically the electromagnetic radiation will also have a
radiation pressure which will push the electron. We are ignoring this complication, ultimately it does not
change our final conclusions.

22

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Figure 8: Compton scattering: a billiard ball-like collision between a photon

and a free electron at rest. The photon has an initial wavelength λ0 and is
incident on the electron from the left. The photon is scattered at an angle
θ and leaves the collision with a longer wavelength λ′ . Credit: Physics for
Scientist and Engineers 4th ed. by Serway

Electron before collision: The electron is at rest and so has zero kinetic energy and
zero momentum.

During the collision the total relativistic energy of the combined electron-photon system
is conserved, and so the photon effectively gives some of its energy and momentum to the
electron, which recoils as shown is Figure 8.

After the collision, the scattered photon has a wavelength λ′ and frequency f ′ , and so:

Photon after collision: The photon is scattered at an angle θ as shown in Figure 8 and
hc h
has energy E ′ = hf ′ = λ′
and momentum with magnitude p′ = λ′
.

Electron after collision: The electron recoils, carrying away some energy and momen-
tum. It is not detected in the experiment.

Thinking about the scattering event as a two-particle collision, the photon must give some
momentum to the electron, and so we immediately expect that

h h
p ′ < p0 ⇒ < ⇒ λ′ > λ0 . (7.5)
λ′ λ0

23

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

An exact analysis using conservation of relativistic energy and momentum gives the follow-
ing equation relating the scattering angle θ, the final wavelength of the scattered photon
λ′ and the initial wavelength of the incident photon λ0 :

h
λ′ − λ 0 = (1 − cos θ) Compton shift equation
me c

where: c is the speed of light; me the mass of the electron; and h Planck’s constant. The
above expression is known as the Compton shift equation since it is an expression for
the change or shift in the wavelength of the incident light compared to the detected light.
This equation is consistent with the empirical observations (7.4) and fully accounts for
the Compton scattering data. Once again we see an example of how classical physics is
incapable of accounting for a phenomena at the atomic scale, and new assumptions are
needed.

24

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

8 Atomic spectra and the Bohr atom

Another mystery that physicists faced throughout the nineteenth and early twentieth
century was the observation that hot atomic gases emit and absorb light at only very
specific frequencies and wavelengths. The exact pattern of the wavelengths emitted and
absorbed was shown to be unique to the element under inspection, and became known as
its atomic spectrum. Since the atomic spectrum was unique to each element, it became
a very useful tool for chemical analysis (i.e. identifying which elements were present in a
sample). However, there was no way that physicists could explain these observation until
they knew something more about the precise structure of the atom.

The electron was first discover by Joseph John Thomson in 1897, and the structure of the
atom was elucidated by Ernest Rutherford in experiments carried out in 1909 to 1911.
Rutherford demonstrated that atoms consist of negatively charged electrons orbiting a
very dense positively charged nucleus. However, this gave rise to the following questions
that classical physics was incapable of answering:

❼ If electrons are orbiting the nucleus in an atom, then they are accelerating. Ac-
cording to classical electromagnetism the orbiting electrons should therefore con-
tinuously emit radiation, lose energy and spiral into the nucleus. What makes the
atom stable?

❼ Why does an element’s emission and absorption spectrum consist of clearly defined
discrete lines?

In 1913 Niels Bohr offered a semi-classical explanation which helped answer these ques-
tions. The term ‘semi-classical’ is used because Bohr started with some distinctively non-
classical assumptions, but after that used ordinary classical physics to obtain a model of
a single electron atom/ion.5 Bohr’s model contains only one electron orbiting a positively
charged nucleus and so if the nucleus contains just one proton, his model describes the
neutral hydrogen atom. If the nucleus contains two protons then his model described a
positively charged helium ion, and so on for more protons in the nucleus. His model,
which describes a single electron circularly orbiting a positively charged nucleus, is not
5
Planck’s explanation of the blackbody spectrum is also semi-classical.

25

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

the fully correct quantum mechanical model6 , but it was an extremely important step
forward. Bohr’s model was developed as follows.

In 1913 Bohr noted that Planck’s constant h had the same units as angular momentum
(i.e. the units of J s is the same as kg m2 s−1 ). So Bohr simply guessed that the angular
momentum of an electron in a circular orbit around the nucleus must be an integer multiple
of the constant7 ~ = h/2π. An electron of mass me moving in a circle of radius r at speed
v about the nucleus has an angular momentum of magnitude L about the nucleus given
by:

L = me vr . (8.1)

So Bohr’s hypothesis that L must come in units of ~ is the statement that

me vr = n~ n = 1, 2, 3, . . . . (8.2)

Bohr then combined the above non-classical assumption with the following two equations
which are entirely classical. The magnitude of the centripetal force on the electron must
be equal to the size of the electrostatic force between the electron and the nucleus that
holds the electron in orbit (the electron’s charge is −e and the nucleus is assumed to have
a charge of Ze where Z is the number of protons):
me v 2 Ze2
=k 2 (8.3)
r r
where k = 1/(4πǫ0 ), and the electron’s total energy is
me v 2 Ze2
E= −k (8.4)
2 r
which is the sum of its kinetic energy and the potential energy which arises due to its
proximity to the nucleus.

Using the three equations (8.2), (8.3) and (8.4), you should be able to show that the
allowed speeds vn , the radii of orbits rn and energy En of the electron are quantised
(come in discrete units) as follows:
kZe2 n 2 ~2 k 2 Z 2 e 4 me
vn = , rn = , En = − n = 1, 2, 3, . . . . (8.5)
n~ kZme e2 2n2 ~2
6
The correct version we know today was worked out by Erwin Schrödinger in 1926.
7
To save time we are simplifying history a little here. We will not go into further detail about why it
was integer multiples of ~ and not just h. But it should be emphasised that Bohr just guessed!

26

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Introducing the quantity

~2
a0 = = 0.0529 nm Bohr radius (8.6)
kme e2

which is a constant called the Bohr radius (and corresponds to the smallest radius orbit
of the hydrogen atom, i.e. the second equation in expression (8.5) with Z = 1 and n = 1),
we find the energy of the electron can be written as

ke2 Z 2
 
En = − n = 1, 2, 3, . . . (8.7)
2a0 n2

The number n is another example of a quantum number and it labels the allowed
quantum states in which the electron can exist. The quantum state with n = 1 is
known as the ground state. The quantum state with n = 2 is known as the first
excited state, the quantum state with n = 3 is known as the second excited state,
and so on. If we insert numerical values into the above expression we find – in units of
eV (where 1 eV = 1.602 × 10−19 J) – that it becomes
 2
Z
En = −13.606 eV n = 1, 2, 3, . . . . (8.8)
n2

When an electron makes a transition from an initial higher energy state with a quantum
number ni and energy Eni , to a final lower energy state with quantum number nf and
energy Enf , then the atom/ion will emit a photon with frequency f given by:
!
E ni − E nf kZ 2 e2 1 1
f= = − 2 . (8.9)
h 2a0 h n2f ni

Additionally, only photons of these frequencies can be absorbed by the atoms/ions. These
frequencies are in very close agreement with the measured atomic spectra of single electron
atoms/ions.

It was known, even at the time that Bohr constructed his model, that this could not be the
full story. His construction lacked any convincing theoretical basis, it could not account
for the more subtle or fine features of the hydrogen spectrum, and it was incapable of
describing the spectra of more complicated atoms/ions which had more than just one
electron. A more complete model would have to wait until 1926 when quantum physics
was better understood.

27

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

9 The de Broglie hypothesis and matter waves

As described earlier, by around 1919 Einstein had concluded that light with a frequency
f and wavelength λ can behave either as a wave, or as if it was made up of particles each
with energy

hc
E = hf = Energy of a photon (9.1)
λ

and a momentum of magnitude

E
p= . Momentum of a photon (9.2)
c

Together these lead to a relationship between the wavelength of a photon and its momen-
tum

h
p= . Momentum of a photon in terms of wavelength (9.3)
λ

In 1923 Louis de Broglie reasons as follows: since light is classically thought to be a wave,
but it clearly can manifest particle-like behaviour, as in the photoelectric effect, perhaps
particles may reciprocally manifest wave-like behaviour. This idea is unifying, for if true
it suggested that you may think of everything as having both particle and wave-like
properties. This reasoning led de Broglie to postulate (i.e. guess) that matter particles
– for example electrons – with a momentum of magnitude p could have an associated
wavelength λ given by

h
λ= . de Broglie wavelength (9.4)
p

This wavelength is now known as the de Broglie wavelength of a particle. Notice that
this last expression is identical to equation (9.3), and so it has the unifying property that
it applies to all particles: matter particles and photons.

With this suggestion, for the first time in history matter begins to be described by wave-
functions. You encountered the idea of a wavefunction earlier in this course in the topic
of Waves and Optics. For example, a one-dimensional sinusoidal wave is described by the
wavefunction y(x, t) = A cos(kx + ωt), and you can add a number of such wavefunctions
together to describe interference. Thanks to de Broglie, matter particles like electrons

28

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

now begin to be described by wavefunctions just as waves traditionally were, and these
so-called matter waves can also be added together to interfere.

Again, de Broglie simply guessed that particles like electrons could behave as waves. We
will now discuss some of the evidence for this hypothesis.

29

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

10 Bohr’s atom and matter waves

You should recall that Bohr’s model of the atom begins with the assumption that a single
electron of mass me moving in a circle of radius r at speed v about the nucleus has an
angular momentum of magnitude L about the nucleus which comes in units of ~:

L = me vr = n~ n = 1, 2, 3, . . . . (10.1)

As a starting point, this is not a very satisfying explanation. Why does the angular
momentum have this strange property?

It turns out that the de Broglie postulate – that a particle with momentum of size p can
behave as a wave with wavelength λ = h/p – could readily be used to give an explanation
for Bohr’s assumption.

Earlier in the Waves and Optics part of this course you learned that a string with both
ends fixed can support standing waves only of certain wavelengths. Likewise, if you
assume that the electron behaves as a wave propagating around the nucleus, and that
only ‘standing waves orbits’ are allowed, then the orbits must contain an integral number
of de Broglie wavelengths as shown in Figure 9.

n=3 n=4 n=5 n=6

Figure 9: Electrons as standing waves. Only orbits, like those shown, where the
electron wave interferes constructively with itself in a standing wave pattern
are allowed.

The requirement that the electrons are only allowed to orbit on a circle of radius r which
is exactly an integral number of de Broglie wavelengths is just:

2π r = n λ n = 1, 2, 3, . . . . (10.2)

30

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Inserting de Broglie’s postulate λ = h/p into this expression we find

nh
2π r = n = 1, 2, 3, . . . (10.3)
p

which rearranges to give

nh
pr = = n~ n = 1, 2, 3, . . . . (10.4)
2π

Now using p = me v with me the mass of the electron and v its speed, we find:

me vr = n~ n = 1, 2, 3, . . . (10.5)

which is precisely Bohr’s initial quantisation assumption (10.1). Although this is not
direct confirmation that electrons can behave like waves, it is highly suggestive and was
an important step forward in the history of physics. The matter wave hypothesis helped
explain Bohr’s wild guess, even though it too was just a wild guess.

More direct and compelling experimental evidence would have to wait until 1927.

31

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

11 X-ray and electron scattering from crystals

In 1912 Max von Laue was credited with the discovery that when x-rays are shined on
crystals they are diffracted and an interference pattern is observed: a wave phenomena.
Later that same year Lawrence Bragg came up with what is now known as Bragg’s law,
which not only gave a solid explanation for this phenomenon, but also demonstrated that
x-rays could be used to determine the structure of the target crystals.

In greater detail, in the early 1910’s it had been established that when monochromatic
beams of x-rays were directed towards samples of crystals, very specific interference pat-
terns were detected in the scattered radiation. A schematic picture of the experimental
setup is shown in Figure 10, and Figure 11 shows an example of an x-ray photograph
from such an experiment. These experiments came to be known as x-ray scattering or
x-ray diffraction experiments.

Figure 10: The experimental set-up for x-ray (and electron) scattering from
crystal targets. The photographic plate reveals a series of concentric rings:
regions of constructive interference. Credit: Quantum Physics of Atoms,
Molecules, Solids, Nuclei, and Particles 2nd ed. Eisberg and Resnick

Lawrence Bragg’s explanation for this phenomenon involved assuming that crystals were
made of parallel sheets of evenly spaced atoms and that incoming x-rays are reflected
off the different sheets. Upon reaching the detector, different path lengths of the x-rays
gives rise to interference. See Figure 12 which shows the geometry of x-rays incident at

32

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

Figure 11: An x-ray diffraction pattern. This is an image of the photographic

plates from an x-ray scattering experiment carried out on zirconium oxide
crystals. Credit: Quantum Physics of Atoms, Molecules, Solids, Nuclei, and
Particles 2nd ed. by Eisberg and Resnick

an angle8 of θ on two consecutive atomic layers separated by a distance d.

wavefronts

θ θ
θ θ
d layers of atoms

d sin θ

Figure 12: The x-rays are incident from the top left and are reflected off two
layers of atoms with a separation d. The path difference between the upper
and lower rays is 2d sin θ.

From Figure 12 we can see that constructive interference will occur if the path difference
8
Warning: It is conventional in crystallography to denote the angle of incidence as that angle between
the plane and the incoming ray, not the angle between the normal to the plane and the incoming ray as
is usually done in classical optics.

33

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

between the two rays is an integral number of wavelengths λ:

path difference = n λ

⇒ 2d sin θ = n λ Bragg’s law

This last expression is known as Bragg’s law and can be used to infer information about
the structure of the target crystal.

From 1923 to 1927 a series of x-ray-like scattering experiments were carried out on a
crystal of nickel metal, but instead of using x-rays it was electrons that were being fired
at the target crystal (the experimental set-up is that of Figure 10). It was observed
that electrons scattered from crystals display interference-like patterns, thus confirming
that electrons can indeed manifest wave-like properties. Additionally, the wavelengths
observed were those predicted by de Broglie. A side-by-side comparison of an x-ray
scattering interference pattern and electron scattering interference pattern is given in
Figure 13.

Figure 13: A comparison of x-ray and electron diffraction patterns. Left:

the photographic plates from an x-ray scattering experiment carried out on
zirconium oxide crystals. Right: the photographic plates from an electron
scattering experiment carried out on gold crystals. This is clear evidence that
electrons can behave as waves. Credit: Quantum Physics of Atoms, Molecules,
Solids, Nuclei, and Particles 2nd ed. by Eisberg and Resnick

34

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

12 Probability in physics: the double slit experiment

with electrons

In more modern times, there is mountains of evidence for matter waves. For example, the
double slit experiment – which was traditionally carried out with monochromatic light –
has been directly performed by shooting mono-energetic electrons (i.e. electrons all with
the same momentum and therefore the same de Broglie wavelength), one by one at a
screen through a pair of slits. The electrons arrive one at a time at the screen, but as
time passes the familiar wave interference pattern, slowly, spot by spot, builds up on the
screen. An animation of this will be shown in the lectures (it can also be found on LMS).
At the beginning of the experiment (or the animation) the electrons just appear to be
hitting the screen at random places, but as you accumulate more data it becomes very
clear it is not entirely random, the electrons are more likely to be found in some places
rather than others, and the interference pattern starts to become clear.

Notice that when performing the double slit experiment with electrons, although the
electrons are exhibiting wave-like behaviour, when you actually detect the electrons at
the screen they always arrive as discrete particles. Detected electrons are not spread out
in a wave-like way, but rather, upon detection you only ever observe a whole particle, never
some fraction of a particle. In 1926 Max Born started thinking about this issue and was led
to conclude that the amplitude squared of the wavefunction describing a particle cannot
possibly be related to intensity – as is the case for normal classical waves – because the
‘intensity of a particle’ simply makes no sense. Instead Born was led to postulate that the
amplitude squared of a particle’s wavefunction is related to the probability of detecting
that particle. Using Born’s probability idea, we should expect that an electron in the
double slit experiment is more likely to land in regions where the amplitude squared of
its wavefunction is greatest, but you cannot ever be certain about exactly where it will
land. This idea gives us precisely what we observe – the interference pattern.

In the case of a double slit experiment carried out with light, we directly see the inter-
ference pattern on the screen as light intensity. In the case of a double slit experiment
carried out with electrons, the interference pattern is effectively a visualisation of the
probability that an electron will land there. In both cases it is the amplitude squared of

35

Downloaded by Oppa Fest ([email protected])

 lOMoARcPSD|36070675

the wavefunction.

The idea that the amplitude squared of a particle’s wavefunction is related to the proba-
bility of its detection is known as Born’s rule and it has become a fundamental axiom
of modern quantum physics. Born’s rule is a starting point for quantum physics, just as
Newton’s laws are starting points for classical mechanics. It was in 1926 with Born’s rule
that probability first entered physics, and soon most physicists came to accept that at a
fundamental level the universe was not deterministic.

36

Downloaded by Oppa Fest ([email protected])