4 j INTRODUCTION AND ORIENTATION

Einstein showed that matter is quantized too. Quantization appears to be

universal. Neither was able to justify the form that quantization took (with
oscillators excitable in steps of hn), but that is a problem we shall solve later
in the text.

0.3 The photoelectric and Compton effects

In those enormously productive months of 1905–6, when Einstein formu-
lated not only his theory of heat capacities but also the special theory
of relativity, he found time to make another fundamental contribution
to modern physics. His achievement was to relate Planck’s quantum
hypothesis to the phenomenon of the photoelectric effect, the emission of
electrons from metals when they are exposed to ultraviolet radiation. The
puzzling features of the effect were that the emission was instantaneous when
the radiation was applied however low its intensity, but there was no emis-
sion, whatever the intensity of the radiation, unless its frequency exceeded a
threshold value typical of each element. It was also known that the kinetic
energy of the ejected electrons varied linearly with the frequency of the
incident radiation.
Einstein pointed out that all the observations fell into place if the elec-
tromagnetic field was quantized, and that it consisted of bundles of energy
of magnitude hn. These bundles were later named photons by G.N. Lewis,
and we shall use that term from now on. Einstein viewed the photoelectric
effect as the outcome of a collision between an incoming projectile, a
photon of energy hn, and an electron buried in the metal. This picture
accounts for the instantaneous character of the effect, because even one
photon can participate in one collision. It also accounted for the frequency
threshold, because a minimum energy (which is normally denoted F and
called the ‘work function’ for the metal, the analogue of the ionization
energy of an atom) must be supplied in a collision before photoejection can
occur; hence, only radiation for which hn > F can be successful. The linear
dependence of the kinetic energy, EK, of the photoelectron on the frequency
of the radiation is a simple consequence of the conservation of energy,
which implies that

EK ¼ hn  F ð0:6Þ

If photons do have a particle-like character, then they should possess a

linear momentum, p. The relativistic expression relating a particle’s energy to
its mass and momentum is

E2 ¼ m2 c4 þ p2 c2 ð0:7Þ

where c is the speed of light. In the case of a photon, E ¼ hn and m ¼ 0, so

hn h
p¼ ¼ ð0:8Þ
c l
 0.4 ATOMIC SPECTRA j 5

This linear momentum should be detectable if radiation falls on an electron,

for a partial transfer of momentum during the collision should appear as a
change in wavelength of the photons. In 1923, A.H. Compton performed the
experiment with X-rays scattered from the electrons in a graphite target, and
found the results fitted the following formula for the shift in wavelength,
dl ¼ lf  li, when the radiation was scattered through an angle y:

dl ¼ 2lC sin2 12 y ð0:9Þ

where lC ¼ h/mec is called the Compton wavelength of the electron

(lC ¼ 2.426 pm). This formula is derived on the supposition that a photon
does indeed have a linear momentum h/l and that the scattering event is like a
collision between two particles. There seems little doubt, therefore, that
electromagnetic radiation has properties that classically would have been
characteristic of particles.
The photon hypothesis seems to be a denial of the extensive accumulation
of data that apparently provided unequivocal support for the view that
electromagnetic radiation is wave-like. By following the implications of
experiments and quantum concepts, we have accounted quantitatively for
observations for which classical physics could not supply even a qualitative
explanation.

0.4 Atomic spectra

There was yet another body of data that classical physics could not elucidate
before the introduction of quantum theory. This puzzle was the observation
that the radiation emitted by atoms was not continuous but consisted of
discrete frequencies, or spectral lines. The spectrum of atomic hydrogen had a
very simple appearance, and by 1885 J. Balmer had already noticed that their
wavenumbers, ~n, where ~n ¼ n/c, fitted the expression
 
1 1
~n ¼ RH 2  2 ð0:10Þ
2 n

where RH has come to be known as the Rydberg constant for hydrogen

(RH ¼ 1.097  105 cm1) and n ¼ 3, 4, . . . . Rydberg’s name is commemorated
because he generalized this expression to accommodate all the transitions in
atomic hydrogen. Even more generally, the Ritz combination principle states
that the frequency of any spectral line could be expressed as the difference
between two quantities, or terms:

~n ¼ T1  T2 ð0:11Þ

This expression strongly suggests that the energy levels of atoms are confined
to discrete values, because a transition from one term of energy hcT1 to
another of energy hcT2 can be expected to release a photon of energy hc~n, or
hn, equal to the difference in energy between the two terms: this argument
6 j INTRODUCTION AND ORIENTATION

leads directly to the expression for the wavenumber of the spectroscopic

transitions.
But why should the energy of an atom be confined to discrete values? In
classical physics, all energies are permissible. The first attempt to weld
together Planck’s quantization hypothesis and a mechanical model of an atom
was made by Niels Bohr in 1913. By arbitrarily assuming that the angular
momentum of an electron around a central nucleus (the picture of an atom
that had emerged from Rutherford’s experiments in 1910) was confined to
certain values, he was able to deduce the following expression for the per-
mitted energy levels of an electron in a hydrogen atom:
me4 1
En ¼  n ¼ 1, 2, . . . ð0:12Þ
8h2 e20 n2
where 1/m ¼ 1/me þ 1/mp and e0 is the vacuum permittivity, a fundamental
constant. This formula marked the first appearance in quantum mechanics of
a quantum number, n, which identifies the state of the system and is used to
calculate its energy. Equation 0.12 is consistent with Balmer’s formula and
accounted with high precision for all the transitions of hydrogen that were
then known.
Bohr’s achievement was the union of theories of radiation and models of
mechanics. However, it was an arbitrary union, and we now know that it is
conceptually untenable (for instance, it is based on the view that an electron
travels in a circular path around the nucleus). Nevertheless, the fact that he
was able to account quantitatively for the appearance of the spectrum of
hydrogen indicated that quantum mechanics was central to any description of
atomic phenomena and properties.

0.5 The duality of matter

The grand synthesis of these ideas and the demonstration of the deep links
that exist between electromagnetic radiation and matter began with Louis de
Broglie, who proposed on the basis of relativistic considerations that with any
moving body there is ‘associated a wave’, and that the momentum of the body
and the wavelength are related by the de Broglie relation:
h
l¼ ð0:13Þ
p
We have seen this formula already (eqn 0.8), in connection with the prop-
erties of photons. De Broglie proposed that it is universally applicable.
The significance of the de Broglie relation is that it summarizes a fusion
of opposites: the momentum is a property of particles; the wavelength is
a property of waves. This duality, the possession of properties that in classical
physics are characteristic of both particles and waves, is a persistent theme
in the interpretation of quantum mechanics. It is probably best to regard
the terms ‘wave’ and ‘particle’ as remnants of a language based on a false
 0.5 THE DUALITY OF MATTER j 7

(classical) model of the universe, and the term ‘duality’ as a late attempt to
bring the language into line with a current (quantum mechanical) model.
The experimental results that confirmed de Broglie’s conjecture are the
observation of the diffraction of electrons by the ranks of atoms in a metal
crystal acting as a diffraction grating. Davisson and Germer, who performed
this experiment in 1925 using a crystal of nickel, found that the diffraction
pattern was consistent with the electrons having a wavelength given by
the de Broglie relation. Shortly afterwards, G.P. Thomson also succeeded
in demonstrating the diffraction of electrons by thin films of celluloid
and gold.2
If electrons—if all particles—have wave-like character, then we should
expect there to be observational consequences. In particular, just as a wave of
definite wavelength cannot be localized at a point, we should not expect
an electron in a state of definite linear momentum (and hence wavelength) to
be localized at a single point. It was pursuit of this idea that led Werner
Heisenberg to his celebrated uncertainty principle, that it is impossible to
specify the location and linear momentum of a particle simultaneously with
arbitrary precision. In other words, information about location is at the
expense of information about momentum, and vice versa. This com-
plementarity of certain pairs of observables, the mutual exclusion of the
specification of one property by the specification of another, is also a major
theme of quantum mechanics, and almost an icon of the difference between it
and classical mechanics, in which the specification of exact trajectories was a
central theme.
The consummation of all this faltering progress came in 1926 when Werner
Heisenberg and Erwin Schrödinger formulated their seemingly different but
equally successful versions of quantum mechanics. These days, we step
between the two formalisms as the fancy takes us, for they are mathematically
equivalent, and each one has particular advantages in different types of cal-
culation. Although Heisenberg’s formulation preceded Schrödinger’s by a few
months, it seemed more abstract and was expressed in the then unfamiliar
vocabulary of matrices. Still today it is more suited for the more formal
manipulations and deductions of the theory, and in the following pages we
shall employ it in that manner. Schrödinger’s formulation, which was in terms
of functions and differential equations, was more familiar in style but still
equally revolutionary in implication. It is more suited to elementary mani-
pulations and to the calculation of numerical results, and we shall employ it in
that manner.
‘Experiments’, said Planck, ‘are the only means of knowledge at our
disposal. The rest is poetry, imagination.’ It is time for that imagination
to unfold.

.......................................................................................................
2. It has been pointed out by M. Jammer that J.J. Thomson was awarded the Nobel Prize for
showing that the electron is a particle, and G.P. Thomson, his son, was awarded the Prize for
showing that the electron is a wave. (See The conceptual development of quantum mechanics,
McGraw-Hill, New York (1966), p. 254.)
8 j INTRODUCTION AND ORIENTATION

PROBLEMS

0.1 Calculate the size of the quanta involved in the 0.13 At what wavelength of incident radiation do the
excitation of (a) an electronic motion of period 1.0 fs, relativistic and non-relativistic expressions for the ejection
(b) a molecular vibration of period 10 fs, and (c) a pendulum of electrons from potassium differ by 10 per cent? That is,
of period 1.0 s. find l such that the non-relativistic and relativistic linear
momenta of the photoelectron differ by 10 per cent. Use
0.2 Find the wavelength corresponding to the maximum in
F ¼ 2.3 eV.
the Planck distribution for a given temperature, and show
that the expression reduces to the Wien displacement law at 0.14 Deduce eqn 0.9 for the Compton effect on the basis of
short wavelengths. Determine an expression for the constant the conservation of energy and linear momentum. Hint. Use
in the law in terms of fundamental constants. (This constant the relativistic expressions. Initially the electron is at rest
is called the second radiation constant, c2.) with energy mec2. When it is travelling with momentum p its
0.3 Use the Planck distribution to confirm the energy is ðp2 c2 þ m2e c4 Þ1/2. The photon, with initial
Stefan–Boltzmann law and to derive an expression for momentum h/li and energy hni, strikes the stationary
the Stefan–Boltzmann constant s. electron, is deflected through an angle y, and emerges with
momentum h/lf and energy hnf. The electron is initially
0.4 The peak in the Sun’s emitted energy occurs at about stationary (p ¼ 0) but moves off with an angle y 0 to the
480 nm. Estimate the temperature of its surface on the basis incident photon. Conserve energy and both components of
of it being regarded as a black-body emitter. linear momentum. Eliminate y 0 , then p, and so arrive at an
0.5 Derive the Einstein formula for the heat capacity of a expression for dl.
collection of harmonic oscillators. To do so, use the 0.15 The first few lines of the visible (Balmer) series in the
quantum mechanical result that the energy of a harmonic spectrum of atomic hydrogen lie at l/nm ¼ 656.46, 486.27,
oscillator of force constant k and mass m is one of the values 434.17, 410.29, . . . . Find a value of RH, the Rydberg
(v þ 12)hv, with v ¼ (1/2p)(k/m)1/2 and v ¼ 0, 1, 2, . . . . Hint. constant for hydrogen. The ionization energy, I, is the
Calculate the mean energy, E, of a collection of oscillators minimum energy required to remove the electron. Find it
by substituting these energies into the Boltzmann from the data and express its value in electron volts. How is
distribution, and then evaluate C ¼ dE/dT. I related to RH? Hint. The ionization limit corresponds to
0.6 Find the (a) low temperature, (b) high temperature n ! 1 for the final state of the electron.
forms of the Einstein heat capacity function. 0.16 Calculate the de Broglie wavelength of (a) a mass of
0.7 Show that the Debye expression for the heat capacity is 1.0 g travelling at 1.0 cm s1, (b) the same at 95 per cent of
proportional to T3 as T ! 0. the speed of light, (c) a hydrogen atom at room temperature
(300 K); estimate the mean speed from the equipartition
0.8 Estimate the molar heat capacities of metallic sodium principle, which implies that the mean kinetic energy of an
(yD ¼ 150 K) and diamond (yD ¼ 1860 K) at room atom is equal to 32kT, where k is Boltzmann’s constant, (d)
temperature (300 K). an electron accelerated from rest through a potential
0.9 Calculate the molar entropy Rof an Einstein solid at difference of (i) 1.0 V, (ii) 10 kV. Hint. For the momentum
T
T ¼ yE. Hint. The entropy is S ¼ 0 ðCV =TÞdT. Evaluate the in (b) use p ¼ mv/(l  v2/c2)1/2 and for the speed in (d) use
1 2
integral numerically. 2mev ¼ eV, where V is the potential difference.

0.10 How many photons would be emitted per second by a 0.17 Derive eqn 0.12 for the permitted energy levels for the
sodium lamp rated at 100 W which radiated all its energy electron in a hydrogen atom. To do so, use the following
with 100 per cent efficiency as yellow light of wavelength (incorrect) postulates of Bohr: (a) the electron moves in a
589 nm? circular orbit of radius r around the nucleus and (b) the
angular momentum of the electron is an integral multiple of
0.11 Calculate the speed of an electron emitted from a clean
h, that is me vr ¼ n
 h. Hint. Mechanical stability of the
potassium surface (F ¼ 2.3 eV) by light of wavelength (a)
orbital motion requires that the Coulombic force of
300 nm, (b) 600 nm.
attraction between the electron and nucleus equals the
0.12 When light of wavelength 195 nm strikes a certain metal centrifugal force due to the circular motion. The energy of
surface, electrons are ejected with a speed of 1.23  106 m s1. the electron is the sum of the kinetic energy and potential
Calculate the speed of electrons ejected from the same metal (Coulombic) energy. For simplicity, use me rather than the
surface by light of wavelength 255 nm. reduced mass m.