Dark Sky Dark Matter

Dark Sky, Dark Matter
Series in Astronomy and Astrophysics

Series Editors: M Birkinshaw, University of Bristol, UK
M Elvis, Harvard–Smithsonian Center for Astrophysics, USA
J Silk, University of Oxford, UK
The Series in Astronomy and Astrophysics includes books on all aspects of

theoretical and experimental astronomy and astrophysics. Books in the series
range in level from textbooks and handbooks to more advanced expositions of
current research.
Other books in the series
Dust in the Galactic Environment, 2nd Edition

D C B Whittet
An Introduction to the Science of Cosmology

D J Raine and E G Thomas
The Origin and Evolution of the Solar System

M M Woolfson
The Physics of the Interstellar Medium

J E Dyson and D A Williams
Dust and Chemistry in Astronomy

T J Millar and D A Williams (eds)
Observational Astrophysics
R E White (ed)
Stellar Astrophysics
R J Tayler (ed)
Forthcoming titles
The Physics of Interstellar Dust

E Krügel
Very High Energy Gamma Ray Astronomy

T Weekes
Series in Astronomy and Astrophysics
Dark Sky, Dark Matter
J M Overduin
Universität Bonn, Bonn, Germany and
Waseda University, Tokyo, Japan
P S Wesson
University of Waterloo, Waterloo, Canada
Institute of Physics Publishing

Bristol and Philadelphia

c IOP Publishing Ltd 2003
All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without the prior permission
of the publisher. Multiple copying is permitted in accordance with the terms
of licences issued by the Copyright Licensing Agency under the terms of its
agreement with Universities UK (UUK).
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.
ISBN 0 7503 0684 X
Library of Congress Cataloging-in-Publication Data are available
Series Editors: M Birkinshaw, University of Bristol, UK

M Elvis, Harvard–Smithsonian Center for Astrophysics, USA
J Silk, University of Oxford, UK
Commissioning Editor: John Navas

Production Editor: Simon Laurenson
Production Control: Sarah Plenty
Cover Design: Victoria Le Billon
Marketing: Nicola Newey and Verity Cooke
Published by Institute of Physics Publishing, wholly owned by The Institute of

Physics, London
Institute of Physics Publishing, Dirac House, Temple Back, Bristol BS1 6BE, UK
US Office: Institute of Physics Publishing, The Public Ledger Building, Suite
929, 150 South Independence Mall West, Philadelphia, PA 19106, USA
Typeset in LATEX 2ε by Text 2 Text, Torquay, Devon

Printed in the UK by MPG Books Ltd, Bodmin, Cornwall
Contents
Preface ix
1 The dark night sky 1
1.1 Olbers’ paradox 1
1.2 A short history of Olbers’ paradox 2
1.3 The paradox now: stars, galaxies and Universe 5
1.4 The resolution: age versus expansion 9
1.5 The data: optical and otherwise 10
1.6 Conclusion 13
2 The modern resolution and energy 16
2.1 Big-bang cosmology 16
2.2 The bolometric background 16
2.3 From time to redshift 20
2.4 Matter and energy 22
2.5 The expansion rate 24
2.6 The static analogue 27
2.7 A quantitative resolution 29
2.8 Light at the end of the Universe? 35
3 The modern resolution and spectra 41
3.1 The spectral background 41
3.2 From bolometric to spectral intensity 42
3.3 The delta-function spectrum 44
3.4 Gaussian spectra 50
3.5 Blackbody spectra 51
3.6 Normal and starburst galaxies 54
3.7 Back to Olbers 61
4 The dark matter 66
4.1 From light to dark matter 66
4.2 The four elements of modern cosmology 67
4.3 Baryons 68
4.4 Cold dark matter 73
4.5 Neutrinos 76
vi Contents
4.6 Vacuum energy 78

4.7 The coincidental Universe 85
5 The vacuum 90
5.1 Vacuum decay 90
5.2 The variable cosmological ‘constant’ 91
5.3 Energy density 95
5.4 Source regions and luminosity 101
5.5 Bolometric intensity 105
5.6 Spectral energy distribution 106
5.7 The microwave background 108
6 Axions 113
6.1 Light axions 113
6.2 Rest mass 114
6.3 Axion halos 117
6.4 Intensity 119
6.5 The infrared and optical backgrounds 120
7 Neutrinos 127
7.1 The decaying-neutrino hypothesis 127
7.2 Bound neutrinos 128
7.3 Luminosity 130
7.4 Free-streaming neutrinos 133
7.5 Intergalactic absorption 135
7.6 The ultraviolet background 139
8 Supersymmetric weakly interacting particles 146
8.1 The lightest supersymmetric particle 146
8.2 Neutralinos 147
8.3 Pair annihilation 149
8.4 One-loop decays 154
8.5 Tree-level decays 157
8.6 Gravitinos 161
8.7 The x-ray and γ -ray backgrounds 163
9 Black holes 170
9.1 Primordial black holes 170
9.2 Initial mass distribution 171
9.3 Evolution and number density 173
9.4 Cosmological density 175
9.5 Spectral energy distribution 177
9.6 Luminosity 179
9.7 Bolometric intensity 179
9.8 Spectral intensity 183
9.9 Higher-dimensional ‘black holes’ 186
Contents vii
10 Conclusions 192
A Bolometric intensity integrals 196
A.1 Radiation-dominated models 196
A.2 Matter-dominated models 197
A.3 Vacuum-dominated models 199
B Dynamics with a decaying vacuum 202
B.1 Radiation-dominated regime 202
B.2 Matter-dominated regime 203
B.3 Vacuum-dominated regime 204
C Absorption by galactic hydrogen 206
Index 209
Preface
The darkness of the night sky awed our ancestors and fascinates modern
astronomers. It is a fundamental problem: why is the night sky dark, and just how
dark is it? This book examines these overlapping problems from the viewpoints
of both history and modern cosmology.
Olbers’ paradox is a vintage conundrum that was known to other thinkers
prior to its formulation by the Prussian astronomer in 1823. Given that the
Universe is unbounded, governed by the standard laws of physics, and populated
by light sources of constant intensity, the simple cube law of volumes and
numbers implies that the sky should be ablaze with light. Obviously this is not so.
However, the paradox does not lie in nature but in our understanding of physics.
A Universe with a finite age, such as follows from big-bang theory, necessarily
has galaxies of finite age. There is therefore a kind of imaginary spherical surface
around us which depends on the age and the speed of light. That is, we can only
see some of the galaxies in the Universe, and this is the main reason why the night
sky is dark. Just how dark can be calculated using the astrophysics of galaxies
and their stars, and the dynamics of relativistic cosmology.
We know from the dynamics of individual galaxies and clusters of galaxies
that the majority of the matter which exerts gravitational forces is not detectable
by conventional telescopes. This dark matter could, in principle, have many
forms, and candidates include various types of elementary particles as well as
vacuum fluctuations, black holes and others. Most of these candidates are unstable
to decay and produce photons. So dark matter does not only affect the dynamics
of the Universe, but the intensity of intergalactic radiation as well. Conversely, we
can use observations of background radiation to constrain the nature and density
of dark matter. Thus does Olbers’ problem gain new importance.
Modern cosmology is a cooperative endeavour, and we would like to
acknowledge some of the people who have helped to shape our ideas about
the Universe and what it may contain. It is a pleasure to thank H-J Fahr and
K-I Maeda for hospitality at the Institut für Astrophysik und Extraterrestrische
Forschung in Bonn and at Waseda University in Tokyo, where much of this
book was written with support from the Alexander von Humboldt Foundation
and the Japan Society for the Promotion of Science. Many of the results we
will present on astrophysics, general relativity and particle physics have been
ix
x Preface
derived in collaboration with colleagues over a number of years. For sharing their
expertise we are particularly indebted to S Bowyer, T Fukui, W Priester, S Seahra
and R Stabell. Finally, our motivation for carrying out these calculations is based
simply on a desire to understand the Universe. This motivation is infectious, and
in this regard we thank especially the late F Hoyle and his colleage D Clayton.
We hope, in turn, that this book will inspire and educate others who wonder about
the dark night sky.
J M Overduin and P S Wesson

10 June 2002
Chapter 1
The dark night sky
1.1 Olbers’ paradox
It is a fundamental observation, which almost anyone can make, that the night sky
is dark. But why?
It might be thought that to a person standing on the side of the Earth that
faces away from the Sun the sky must necessarily be dark. However, on even a
moonless night there is a faint but perceptible light from stars. Consider a situation
in which there are no stars in the line of sight; for example, the hypothetical case
of someone living on a planet at the edge of the Milky Way, who looks out into
intergalactic space. Even in this situation, the night sky would not be perfectly
dark because there are many other luminous galaxies in the Universe. It is at this
point that the fundamental nature of the problem of the dark night sky begins to
become apparent. For modern cosmology implies that the population of galaxies
is uniform in density (on average) and unbounded. Given these conditions,
and assuming that conventional geometry holds, the number of galaxies visible
increases as the cube of the distance. This means that along the line of sight, the
galaxies become increasingly crowded as the distance increases. In fact, under the
conditions formulated, any given line of sight must end on a remote galaxy. Thus,
the view in all directions must ultimately be blocked by overlapping galaxies;
and the light from these, propagating towards us from the distant reaches of the
Universe, should make the night sky bright.
Clearly this does not happen in reality, and the reason why has been a subject
of controversy among astronomers for centuries. In 1823, Olbers published a
paper that drew attention to this problem [1], which has since come to be known
as Olbers’ paradox. However, the concept of a paradox within modern physics
implies a failure in terms of one or more of reasoning, theory or observational
data. Below, we will see how a scrutiny of all three of these has recently enabled
us to resolve Olbers’ paradox.
1
2 The dark night sky
Figure 1.1. H W M Olbers (1758–1840).
1.2 A short history of Olbers’ paradox
H W M Olbers was a German astronomer who was born in 1758 and died in
1840. During his lifetime he was mainly notable for his research on the solar
system. Thus, a comet that returned in 1815 (and again in 1887) was called
Olbers’ comet. He was also involved in the search for a hypothetical missing
planet in the gap between the orbits of Mars and Jupiter. An object (Ceres) was
discovered on 1 January 1801 by G Piazzi, and its orbit was worked out by the
great mathematician K F Gauss and found to lie in the expected region. However,
an unexpected second object (Pallas) was discovered in March of 1802 by Olbers
with an orbit at almost the same distance from the Sun. This was followed in later
years by the finding of yet more objects, including a second (Vesta) by Olbers in
March of 1807. In this way Olbers was associated with the discovery of what later
came to be known as the asteroid belt.
The work for which Olbers is mainly remembered today was first published
in English in 1826 under the title ‘On the Transparency of Space’ [2]. It was then
apparently forgotten for a good many years, only to be recalled in more recent
A short history of Olbers’ paradox 3
times following the proposal of the steady-state cosmology by H Bondi, T Gold

and F Hoyle in 1948. The steady-state theory has an infinite past, and so is very
different from the big-bang theory, which has become accepted as the standard
cosmology because observations support its basic premise of a finite past (see
later). These two theories actually involve quite different approaches to Olbers’
paradox, and it is now the general opinion that any cosmology that hopes to be
taken seriously must have a satisfactory solution to the problem of the dark night
sky. This attitude, and the reawakening of interest in Olbers’ paper and its central
problem, appears to date from a discussion in Bondi’s classic book Cosmology,
which first came out in 1952 [3].
Bondi’s discussion listed several conditions that result in the paradox of a
bright night sky. These included the assumptions that the average density and
luminosity of sources do not vary in space or time, that the sources do not have
large systematic motions, that space is Euclidean, and that the known laws of
physics apply. This list has been criticized by S L Jaki [4] and others, who have
objected that it is not equivalent to what Olbers wrote in his original article. This
may be true, and it is certainly the case that what is called Olbers’ paradox today
exists in many different versions. There is also confusion of other types: some
people call the author of the paradox Wilhelm Olbers and some Heinrich Olbers;
while some date the original formulation of the problem to 1826 rather than
1823. Also, it has been pointed out by various historians that the problem was,
in fact, appreciated in the two centuries preceding Olbers, notably by J Kepler,
N Hartsoeker, E Halley and P L de Chéseaux [5, 6]. But whatever confused path
it has taken through history, it has arrived in modern times as a list of conditions
that leads to a contradiction with observation and is associated primarily with the
name of Olbers.
Currently, there are several slightly different lists of assumptions in use that
lead to the paradox of a bright night sky. But they have in common an implicit
dependence on modern astrophysical knowledge. For example, there is seldom
much discussion today of a subject implied by the title of Olbers’ original article,
namely absorption of light in space. This is primarily because observations of
remote sources such as quasars place low limits on intergalactic matter. However,
it should also be pointed out that even if there were large amounts of absorbing
matter in space, the energy absorbed at one wavelength would eventually be re-
emitted at another wavelength of what we nowadays know to be a continuous
electromagnetic spectrum, so the problem would not be eliminated but merely
shifted. Also, today we know that the sources of light in the paradox should
not be taken to be single stars as assumed in historical times, but galaxies—
‘island universes’ as they were once termed—each containing billions of stars.
Nowadays it is rare to find any mention of avoiding the paradox by taking the
sources to be non-uniformly distributed, since modern observations imply that
the galaxies may be assumed to be uniformly distributed, at least if averaged over
large enough distances. Furthermore, while space over large distances may be
curved in accordance with Einstein’s theory of general relativity, the curvature
actually does not figure directly in the expression for the intensity of intergalactic
light in the case where the Universe is uniform; and even if this were not the case,
space could hardly be so far from Euclidean or flat as to offer a gravitational kind
of resolution of Olbers’ paradox. In the same vein, no compelling evidence has
emerged that the laws of physics are significantly different in remote parts of the
Universe from those nearby, so that way of solving the problem is hardly ever
mentioned now.
Thus, the most up-to-date version of Olbers’ paradox, slimmed down to
avoid mentioning what modern astrophysics tells us can be taken for granted, is
as follows. In a Universe consisting of galaxies that are of infinite age and static,
the accumulated light would be so intense as to make the night sky bright and not
dark as observed.
In this form, it is obvious that one or both of the remaining assumptions are
wrong. It is actually the case, as will be seen later, that both are wrong. The
galaxies have a finite age, which means that the amount of light which they have
emitted into space is limited. And the galaxies are receding from each other due to
cosmic expansion, which means that their light has been reduced in intensity. But
this is merely to scratch the surface of a solution which, to be understood in depth,
requires a critical look at several areas of modern astrophysics and cosmology. In
particular, it is superficial to know only that the two assumptions in this version
of the paradox are wrong. As in other scientific studies, this just leads to the
important question of quantification: which assumption is more wrong? Or more
precisely: what is the exact relative importance of the two assumptions for the
darkness of the night sky?
The person often associated with this question is E R Harrison. In 1964
he discussed the timescales associated with stars and the non-static nature of
the Universe, and concluded that the latter was not of major importance for the
resolution of Olbers’ paradox [7]. He published several more articles thereafter,
and in 1981 in a popular book he restated his view that, in a Universe consisting
of luminous sources, the fact that they have finite ages is far more important than
whether they are static or not [8]. But while Harrison’s view was supported by
a few notable cosmologists, it was not universally accepted. This was mainly
because, following Bondi’s earlier discussion, the motions of the galaxies had
somehow become accepted as the factor in resolving Olbers’ paradox. Also,
Harrison’s technical papers on the subject were not easy to follow, and this
delayed a proper evaluation of the age factor. Thus, while some headway was
made in the 1960s, 1970s and early 1980s towards an objective treatment of the
paradox, controversy and confusion continued to prevail. A survey of modern
astronomy textbooks carried out in 1987 showed how divided opinion was. About
30% agreed with Harrison in saying that the failure of the infinite-age assumption
was most important. While about 50% attributed the resolution of the paradox to
the failure of both the infinite-age assumption and the static assumption, though
without a proper assessment of which was the more important. And about 20% of
the textbooks surveyed mentioned only the failure of the assumption that galaxies
The paradox now: stars, galaxies and Universe 5
are static. This state of confusion prompted the appearance in 1987 of an article by
Wesson et al [9] that introduced a new method, to be discussed later, of tackling
the problem of the dark night sky. Using this, they gave an exact quantitative
assessment of the relative importance of the age and motion factors, and thereby
resolved Olbers’ paradox in a definitive fashion.
To complete this section, it is interesting to note that while Olbers’ paradox
has through history been a subject of mainly theoretical attention, in recent years
it has also entered an observational phase that is likely to become of increasing
importance in the future. Observational attempts to detect the faint intergalactic
light have in the past produced useful but non-definitive data. However, recent
advances in telescope technology have brought us to the threshold of detection
[10]. This prospect is very interesting, since actual measurement would provide
an opportunity to check theoretical estimates of the light’s intensity, and thereby
the underlying principles of astrophysics and cosmology.
1.3 The paradox now: stars, galaxies and Universe
We saw in the preceding section that Olbers’ paradox has evolved in content
and likely resolution since it was propounded by its author in 1823. In recent
years, attention has focused on two ways in which a bright night sky can be
avoided. First, the ages of galaxies may not be infinite as originally assumed
but finite. Second, galaxies may not be static but instead have large systematic
motions. Modern astrophysics tells us that both of these things are the case. But
to understand how they affect the darkness of the night sky we need to look more
closely at our current picture of the Universe.
Stars are the main radiation-producing bodies in the Universe. An average
star converts elements of low atomic number (primarily hydrogen) to elements
of higher atomic number by means of nuclear fusion reactions. The process
by which there is a gradual change in the chemical makeup of a star during its
life, from ‘light’ to ‘heavy’ elements, is called nucleosynthesis. Combining
theoretical data on nucleosynthesis with observational data on the element
abundances in a given star (gathered mainly from spectroscopy) allows its age
to be determined. A by-product of the nuclear reactions in a star is energy, which
is emitted largely in the form of optical radiation. The energy output of a typical
star is colossal by conventional standards due to its very large mass, but the rate of
energy production per unit mass is a parameter with a more reasonable value. As a
rough average value for the energy output per unit time per unit mass of luminous
matter in the Universe, we shall take 0.5 erg s−1 g−1 . This is about one-quarter of
the value for the matter in the Sun, which reflects the fact that most stars (as seen
for instance on a Hertzsprung–Russell diagram of nearby stars) are intrinsically
less luminous than ours. As we will show later, it is the relative smallness of this
number that is one of the reasons the night sky appears dark.
Galaxies are the main mass concentrations in the Universe, each consisting
primarily of a clump of numerous stars bound together by their own gravity. There
is a considerable range in the masses, luminosities and shapes of galaxies. The
shapes are used to classify galaxies into spirals, ellipticals and irregulars. The
masses and sizes are uncertain, because some galaxies may have halos of unseen
dark matter that are considerably larger and more massive than the parts of the
galaxies we can see. This is especially likely to be true of spirals like the Milky
Way. However, the study of the light due to galaxies involves primarily luminous
as opposed to dark matter, so we will defer discussion of dark matter to later
chapters. An average value for the density of luminous matter in the Universe
can be obtained if we assume that the galaxies are distributed uniformly in space.
Galaxies actually tend to clump together because of their mutual gravitational
interactions. But such clusters appear to have only limited sizes, with no sure
observational evidence for very large ones. So if we consider large distances
only, the distribution of galaxies can indeed be assumed to be uniform. Given
this, we can describe the distribution of luminous matter in the Universe in terms
of one parameter, namely the average density. Observational evidence indicates
that this quantity takes a value of about 4 × 10−32 g cm−3 . Like the stellar energy
output per unit mass, this is a very small number by everyday standards, and is
one of the main reasons the night sky appears dark.
Cosmology is the study of the Universe in the large; and because it has more
observational support than its rivals, we will presume the validity of the big-bang
theory of cosmology. This theory is based primarily on two assumptions. The
first, which was discussed earlier, is that the matter distribution over sufficiently
large distances can be taken to be uniform. It should be noted that the word
‘uniform’ implies that there is no identifiable centre or boundary to the matter
distribution, so models of the Universe that emerge from this theory must in some
sense be unbounded or infinite in extent. The second assumption on which the
big-bang theory is based is that the Universe is expanding. This means that the
galaxies are receding from each other (remember: no special or central one), and
in particular they are receding from the Milky Way. We infer that this is happening
because the light we receive from remote galaxies is redshifted, and the most
natural way to account for this is in terms of a Doppler shift from a receding
source. Thus, just as the sound waves from the whistle of a receding train are
received with longer wavelengths, producing a lower pitch, so are the light waves
from the stars of a receding galaxy received with longer wavelengths, producing
a redshift. (Redshift in the big-bang theory is really more subtle than this, and
involves expansion as distinct from velocity, but the two words are commonly
taken to imply each other and this will be done here also.) Redshifts and recession
speeds increase with distance, and it can be shown that the precise relationship—
called Hubble’s law—is consistent with the assumption of uniformity. Insofar as
both of the prime assumptions on which the big-bang theory is based are derived
from observations, the theory is fairly well-founded.
If these assumptions are plugged into Einstein’s theory of general relativity,
which is the appropriate theory of gravity and accelerations for large distances
The paradox now: stars, galaxies and Universe 7
and masses, then the evolution of the Universe can be worked out in detail. There
are actually several different kinds of evolution allowed by the observational input
data, and these are often called standard cosmological models. For each of these,
the distances between galaxies change somewhat differently with time. But all of
the standard models start in a big bang or state of infinite density at time zero.
Note that the big bang is not localized at a special point in space, a concept
that would violate the principle that the standard models are uniform with no
preferred spot. Rather, the big bang is like an explosion that occurs everywhere at
the same time. According to the theory, the real Universe just after the big bang
must have been very hot, and we find evidence of this today in the cooled-down
radiation of the 3 K cosmic microwave background or CMB (which should not
be confused with galactic radiation). Galaxies are thought to have condensed out
of the cooling matter not long after the big bang. Thus, the time that has elapsed
since the big bang and the age of the galaxies are expected to be approximately
equal. Insofar as the big-bang theory implies that galaxies condensed out at one
fairly well-defined epoch, it agrees with the observation that many galaxies have
nearly the same age. The age of the galaxies inferred indirectly (by matching
their motions to the standard models) agrees reasonably well with the age of the
galaxies inferred directly (by dating their oldest stars using nucelosynthesis). A
typical nucleosynthesis value is 16 × 109 years, or 16 Gyr [11]. This number
must clearly play a crucial role in the subject of Olbers’ paradox. But while it is
large, it will be seen later that it cannot offset the influence of the other two small
numbers introduced earlier.
The contents of the three preceding paragraphs may be summed up thus:
stars in galaxies produce light, but the galaxies have only been in existence for
a time of order 16 Gyr and are also expanding away from each other. How do
these last-mentioned factors, the finite age of the galaxies and the expansion of
the Universe, reduce what would otherwise be a blaze of intergalactic light?
The finite age of the galaxies implies directly that they have not had time to
populate the vast reaches of intergalactic space with enough photons to make it
bright. (We will assume there were negligible numbers of optical photons around
at the time galaxies and their stars formed, so the intensity of intergalactic light
started at essentially zero.) This is the simple way to understand this factor. But
there is another way which, while conceptually more difficult to appreciate, has
the advantage of allowing a geometrical argument against the paradox of the
bright night sky on a par with the one used originally by Olbers. The crux of
this argument is light-travel time. To appreciate the importance of this, consider
what happens when we look at a nearby galaxy, like the one in Andromeda. This
is ‘only’ about 2 million light-years away, but even so the finite speed of light
implies that it takes about 2 million years for photons from Andromeda to reach
us, which means that we see it as it was about 2 million years ago. For galaxies
further away, the light-travel time is even larger. Galaxies at very great distances
can be seen, in principle, as they were around the time of their formation about
16 Gyr ago. So, there is an imaginary spherical surface about us at origin, with
a radius of order 16 × 109 light-years (or 16 Gly for short), where galaxies can
in principle be seen as they were at formation. (In practice it is very difficult to
observe such galaxies because they are very remote and thus appear very dim.
Also the cosmological distance associated with a light-travel time of 16 Gyr is
actually larger than 16 Gly, because the Universe has expanded in the meantime.
These points do not alter the argument in a qualitative way.) The light of our
night sky comes only from the limited number of galaxies within this imaginary
surface, because if we could peer further out we would only see the presumably
non-luminous material from which they formed.
Thus, Olbers’ paradox is avoided because, contrary to what he thought, we
do not see arbitrarily large numbers of sources as we look further away. Instead,
we effectively run out of galaxies to see at a distance of order 16 Gly. This
argument is logical, but conceptually difficult to accept for many people because
it seems to imply that we are necessarily talking about a Universe that is finite in
extent. However, this is not so. If we could by some miracle take an instantaneous
snapshot of all of the Universe as it is ‘now’, without taking into account the
lag associated with the light-travel time, we would see many more galaxies; and
depending on the kind of Universe we are living in, these could be arbitrarily
large in number and stretch to arbitrarily great distances. Thus, if we could see
all of the Universe as it is ‘now’, it might well be infinite in extent and contain an
arbitrarily large number of galaxies, and it would certainly be unbounded. (It is
possible that it could contain a large but finite number of galaxies, because space
according to general relativity can curve around and close up on itself, but such a
Universe would still have no boundaries.) However, we do not see to arbitrarily
great distances because light travels with finite speed, and this causes our view of
the Universe to be of things as they were in the past. Galaxies can only be seen
out to a distance corresponding to the time when they formed. We can sum up the
finite-age factor by saying that it directly limits the number of photons that have
been pumped into space by galaxies, or it creates an imaginary surface that limits
the portion of the Universe from which we receive photons emitted by galaxies.
On either interpretation, it is clear that intergalactic space and the night sky will
not be arbitrarily bright.
The expansion of the Universe implies two things that both reduce the
intensity of intergalactic light and are relatively easy to understand. Firstly, as the
galaxies recede from each other and the volume of intergalactic space increases,
the number of photons per unit volume and therefore the intensity of the light
decreases. Secondly, as the galaxies recede the photons they emit are redshifted,
so by Planck’s law their intrinsic energies decrease, as does the intensity of the
light. These things combine to reduce the brightness of intergalactic space and
the night sky.
The age and expansion factors described here have been much discussed.
The first has often been misunderstood because of the conceptual difficulty
involved in understanding the light-travel-time effect. The second, perhaps
because of this, has often been quoted in the literature as if it was the major or
The resolution: age versus expansion 9
only factor in avoiding Olbers’ paradox. But before this problem can be laid to
rest, a quantitative answer has to be given to the following question: Which of the
two factors, the finite lifetime of the galaxies or the expansion of the Universe, is
the more important in accounting for the dark night sky?
1.4 The resolution: age versus expansion

Finding the relative sizes of these factors is not easy: a proper calculation involves
the heavy mathematics of Einstein’s theory of general relativity, wherein the
two factors are intertwined and deeply embedded. Later, a new method will be
outlined that can deal with this complexity. But for the present, to get some idea
of the kinds of numbers involved in the problem, let us do a simpler calculation.
From our discussion of stars, galaxies and cosmology, we are in possession
of three important numbers. These are: the average rate of energy production per
unit mass of luminous matter (0.5 erg s−1 g−1 ); the average density of luminous
matter (4 × 10−32 g cm−3 ); and the age of the galaxies (16 Gyr = 5 × 1017 s
approximately). The product of these three numbers is 1×10−14 erg cm−3 , which
is an energy density. If we multiply this by the speed of light (3 × 1010 cm s−1 )
we obtain 3 × 10−4 erg s−1 cm−2 , which is an intensity. Dimensional analysis
suggests that this number must be related to the intensity of intergalactic light.
It can be shown [12] that it is actually equal to the intensity of light in a static
Universe where relativistic effects are negligible.
The preceding calculation is useful because it shows that even if the Universe
were not expanding the intensity of intergalactic light would still be quite low, of
order 10−4 or 10−3 erg s−1 cm−2 . (This result is quasi-Newtonian, and inaccurate
mainly because it neglects the expansion.) In other words, since expansion can
only reduce the intensity, the preceding calculation provides us with an upper limit
for the brightness of intergalactic space. This upper limit can be better appreciated
via the following comparison. Consider a 100 W light bulb hanging in the centre
of an average-sized room whose walls, floor and ceiling have a summed area of
100 m2 (106 cm2 ). The intensity of illumination for this room, assuming all of
the 109 erg s−1 power of the bulb is converted to light, is 103 erg s−1 cm−2 . By
comparison with the level of illumination in the average domestic situation, the
light due to galaxies is at least a million times fainter.
From this comparison we can return to the main question concerning the
relative importance of the age and expansion factors. How can we find out which
is the more important in our Universe, in a way that is accurate and takes into
account the complexities of Einsteinian cosmology?
If we could stop the expansion of the Universe, while keeping the properties
of the galaxies (and especially their age) unchanged, we could work out the
intensity of intergalactic light in such a static Universe, knowing that the finite age
of the galaxies was the determining factor. Then we could allow the expansion
to resume, knowing that the intensity of intergalactic light was determined by
both the age factor and the expansion factor. If we thereafter formed the ratio of
intensities, with and without expansion, we would be able to state not only that it
is less than unity, but also its exact value. If the ratio was small compared to unity,
we would be able to conclude that expansion is the dominant factor. Whereas, if
the ratio was of order unity, we would be able to conclude that expansion is not
the dominant factor.
The procedure just outlined can be carried out by using models. If we were
sure of the parameters that describe the real Universe, it would only be necessary
to consider one model, and the calculation leading to the ratio of intensities with
and without expansion could be carried out by hand. Unfortunately, we are not
sure of the parameters that describe the real Universe, so it is necessary to consider
many models and leave the calculations of the ratios of intensities to a computer.
We will describe the detailed calculations—which have wide applicability—in the
next chapter. But they have a consensus, which is that the ratio of intensities with
and without expansion is approximately 0.5. In other words, the intensity of light
due to galaxies in an expanding Universe is reduced from that in an equivalent
static Universe by about a factor of two.
This conclusion effectively resolves Olbers’ paradox once and for all. We
previously saw that the intensity of intergalactic light in a static Universe,
where only the finite age of the galaxies is significant, is of the order 10−4 or
10−3 erg s−1 cm−2 . We now see that the intensity in an expanding Universe
with equivalent properties is merely reduced by half. If the order of magnitude
of the intensity is determined by the finite age of the galaxies, and only modified
by a number of order unity by the expansion of the Universe, then clearly we
are justified in saying that the former factor is the more important. This agrees
with previous arguments by a few respected cosmologists (notably E R Harrison),
but it disagrees with statements in several textbooks. The method just discussed
leaves no doubt about the true situation, however. The reason why the night sky
is dark has mainly to do with the finite age of the galaxies, not the expansion of
the Universe.
1.5 The data: optical and otherwise
Olbers’ paradox has traditionally been discussed in terms of light. But of course
this is only the optical, or human-eye-receiving, band of an infinitely wide
electromagnetic spectrum. Due to absorption by the Earth’s atmosphere, data
are easy or difficult to obtain by ground-based observations, depending on the
waveband.
In recent times, we have been able to obtain better data by using rocket-
borne instruments and artificial satellites. However, technology progresses more
slowly than insight. Thus we are in the position that observations provide mainly
constraints rather than proofs of what we have discussed earlier. Specifically,
we have good constraints in the optical band on intergalactic light. (This is
The data: optical and otherwise 11
Figure 1.2. The full spectrum of the diffuse background radiation reaching our galaxy
(figure courtesy of R C Henry [13]). This is a compilation of observational measurements
and upper limits in wavebands from 1 radio to 2 microwave, 34 infrared, 5 optical,
67 ultraviolet, 8 x-ray and 9 γ -ray regions. Immense amounts of work, both
theoretical and experimental, have gone into understanding each waveband. We will
return to these in the chapters that follow, focusing on specific parts of the spectrum and
comparing them to the predicted background intensities from galaxies and other sources of
radiation.
commonly referred to in the literature as the extragalactic background light

or EBL, meaning that we exclude primarily the light from nearby stars; and we
have recently obtained improved data on this from the Hubble Space Telescope.)
On the short-wavelength side of the optical, we have good data in the ultraviolet
(UV), x-ray and γ -ray bands. On the long-wavelength side, we have usable data in
the infrared (IR) band, and excellent data in the microwave band, beginning with
the Cosmic Background Explorer (COBE). All this information is summarized
graphically in figure 1.2. Here, we wish to briefly describe what techniques
are used to study the optical band, and note how sharply the Olbers problem is
constrained by data in other bands.
Several attempts have been made to detect the Olbers light or EBL using
large, ground-based telescopes. The idea behind such attempts is simple enough:
if a telescope is pointed towards an area of the sky where there are no nearby
sources, the light from the many remote galaxies in the field of view should
combine to make a considerable flux, even if individual galaxies cannot be
distinguished because their images are too faint. However, there is a major
problem to contend with: the light from remote galaxies is swamped by other
kinds of light that also enter the telescope. One of these is the light from the stars
of the Milky Way. This has to be discounted, as does light from other non-galactic
sources. The procedure to do this is complicated, but it is essential if one is to
isolate the light from external galaxies. What was probably the most thorough
ground-based search for the intergalactic light, both in terms of observations and
the correction procedure just mentioned, was made by R R Dube, W C Wickes and
D T Wilkinson, who used the Number 1 telescope at Kitt Peak in Arizona, USA
[14]. Unfortunately, after their observations had been corrected for light from
stars in the Milky Way and other places, they found that there was actually nothing
measurable left. (This means that the integrated light from many remote but
unresolved galaxies lying in the field of view of the telescope was too weak to be
identified by the detection equipment used.) Such a null result can still be useful,
though, because it provides an upper limit to the intensity of intergalactic light
that can be employed to constrain and check cosmological theory. Numerically,
the upper limit set by the three scientists is 3 × 10−4 erg s−1 cm−2 approximately.
It is doubtful if this can be improved on, or a positive identification made, using
ground-based optical telescopes. Other techniques, however, provide more hope.
One of these is to use the Hubble telescope in space, the data from which can be
combined with other ground-based observations. This technique has been used by
R Bernstein and co-workers [10]. As mentioned earlier, this method may provide
the first concrete detection of the Olbers light in the optical band, but it is still in
development.
We can also make observations from space in the infrared band. It is not
difficult to understand the logic behind this. The intergalactic radiation field
consists not only of optical photons from galaxies that are relatively nearby
and at intermediate distances, but also includes photons from galaxies far away
that have been redshifted to considerably longer infrared wavelengths by the
expansion of the Universe. Just what part of the total energy of the field is
at infrared wavelengths is hard to say, because it depends on various poorly
known cosmological factors such as the formation epoch of galaxies. But it is
probably significant. (It can be remarked in passing that theoretical estimates
of the intensity of the radiation field due to galaxies are often bolometric ones,
integrated over all wavelengths; whereas observations are made at one or a few
wavelengths.) In addition to photons that have been redshifted into the infrared,
there could also be photons of this type that were re-emitted from dust associated
with young distant galaxies. Thus, while we must certainly receive a lot of
optical photons from galaxies that are not too far away, we may also receive many
infrared photons from galaxies that are remote. This situation, wherein the flux
Conclusion 13
of photons we receive at the Earth is a combination due to various mechanisms,

is common to all wavebands.
The CMB is better understood than the infrared background. The
intergalactic radiation field at wavelengths of order 1 cm is known from both
ground and satellite observations to be highly isotropic and very cold. To be
specific, direction-dependent fluctuations are known from the COBE satellite to
be only of order 1 in 105, and the mean temperature is close to 2.7 K (or about
3 degrees above absolute zero). These data are commonly taken to mean that
this field was produced in the fireball that followed the big bang, and has been
cooled by the expansion of the Universe. Irrespective of its origin, the intensity
of the CMB provides a kind of cosmological baseline that severely restricts
contributions from other sources.
On the short-wavelength side of the optical, we can use the data available
to constrain the Olbers radiation and thus mechanisms that produce it. While we
have yet to delve into the murky waters of dark matter, there is ample evidence
from gravitational effects that we only see a minor fraction of the matter in the
Universe. A well-known candidate for the rest is neutrinos. These particles
are uncharged but ubiquitous, so even if they had a small mass they could, in
principle, dictate the dynamical evolution of the cosmos. However, the most-
discussed model for massive neutrinos, due to D W Sciama, predicts that they
should decay. This would produce an Olbers field at ultraviolet wavelengths. Data
on the intergalactic radiation field at wavelengths of order 1000 Å constrain this
model, and enable us to work backward and set stringent constraints on neutrinos.
The same argument, using data in other wavebands, applies to axions (neutral
particles that some theories of particle physics predict should have been produced
copiously in the big bang), WIMPs (weakly-interacting massive particles, for
which there is evidence from particle accelerators), black holes (which decay by
a process known as Hawking evaporation) and even the decay of the vacuum
itself (allowed by quantum field theory). In short, dark matter is not black: while
an individual particle may only produce an amount of energy that is small, the
summed contribution to the intergalactic radiation field may be significant in a
Universe that is vast.
1.6 Conclusion
The traditional paradox named after Olbers can be stated in modern language as
follows: ‘in a Universe consisting of galaxies that are of infinite age and static,
the accumulated light would be so intense as to make the night sky bright and
not dark as observed’. We now know that both of the assumptions remaining in
this formulation are wrong. However, the ways in which the age and expansion
factors lead to a resolution of the problem are different.
• The age of the galaxies is finite, so the amount of light they have pumped
into intergalactic space has been limited. This, in combination with the
relatively low rate of energy production and vast distances between galaxies,
has resulted in a low intensity of intergalactic light. An alternative, though
more subtle, way of regarding the age factor is as a geometrical limitation.
The fact that the galaxies formed a finite time ago means that we can only see
them out to a distance which is (roughly) of the order of their age multiplied
by the speed of light. This implies that there is an imaginary surface about us
which effectively delimits that part of the (unbounded) Universe from which
we presently receive galactic light. Since we only receive light from a finite
number of them, the light due to galaxies is limited in intensity.
• The Universe is expanding, which means that the galaxies are receding from
each other. This causes the volume of intergalactic space to increase, and
the energy density and intensity of intergalactic light to decrease. Also, the
light emitted by galaxies is redshifted by the cosmological version of the
Doppler effect. And since by Planck’s law redder photons have less energy,
the intensity of intergalactic light suffers a further decrease.
Of these two factors, the former is more important because it sets the order
of magnitude of intergalactic light, while the second only reduces it further by
about a half.
In later chapters, we will lay the calculational foundation for this conclusion,
and then proceed to see how modern data on the darkness of the night sky can be
applied to the problems of modern astrophysics. Of these problems, the largest
is that we do not see all of the matter in the Universe and do not know its
composition. However, the prime candidates involve a slow decay which feeds
photons into intergalactic space. There is an intimate connection between the
dark sky and dark matter.
References
[1] Olbers W 1823 Berliner Astronomisches Jahrbuch für das Jahr 1826 (Berlin: C F E
Späthen) p 110
[2] Olbers W 1826 Edinburgh New Phil. J. 1 141
[3] Bondi H 1952 Cosmology (Cambridge: Cambridge University Press)
[4] Jaki S L 1967 Am. J. Phys. 35 200
[5] Hoskin M 1997 Cambridge Illustrated History of Astronomy (Cambridge:
Cambridge University Press) pp 202–7
[6] Jaki S L 2001 The Paradox of Olbers’ Paradox 2nd edn (Pinckney, MI: Real View
Books) pp 1–94
[7] Harrison E R 1964 Nature 204 271
[8] Harrison E R 1981 Cosmology, the Science of the Universe (Cambridge: Cambridge
University Press)
[9] Wesson P S, Valle K and Stabell R 1987 Astrophys. J. 317 601
[10] Bernstein R A 1999 The Low Surface Brightness Universe (Astronomical Society of
the Pacific Conference Series, Volume 170) ed J I Davies, C Impey and S Phillipps
(San Francisco, CA: ASP) p 341
Conclusion 15
[11] Vandenberg D A, Bolte M and Stetson P B 1996 Ann. Rev. Astron. Astrophys. 34 461
[12] Wesson P S 1986 Sp. Sci. Rev. 44 169
[13] Henry R C 1999 Astrophys. J. 516 L49
[14] Dube R R, Wickes W C and Wilkinson D T 1979 Astrophys. J. 232 333
Chapter 2
The modern resolution and energy
2.1 Big-bang cosmology

The best theory of the Universe we have is based on the assumptions that the
galaxies are distributed approximately uniformly; that they are increasing their
separations from one another; and that matter came out of a singularity. The
best theory of gravity we have is Einstein’s general relativity. It is justifiable
to neglect the other interactions or forces known to modern physics in many
astronomical problems: electromagnetism is long-range (like gravity) but there
is no evidence of net electrical charges on astronomical bodies, so its effects are
localized; the weak interaction, whereby a neutron changes to a proton and an
electron, is the basis of nuclear fission but is not of astrophysical importance; the
strong interaction figures crucially in the process whereby the Sun produces its
energy by fusion, but the stars and galaxies can be treated as point sources of
radiation for many purposes in cosmology. Thus, of the four known interactions,
gravity rules in the Universe at large.
It would be philosophically pleasing, of course, to have a unified field theory
of all the ( 4) forces. Such theories have been reviewed from an astronomical
standpoint in a recent book [1], and we will have need in later chapters to ‘import’
the physics of particles into the physics of the Universe. Here, however, we will
work with gravity alone, and specifically with general relativity, which has been
amply verified by Solar System and other tests [2, 3]. Our aim in this chapter will
be to show how Olbers’ paradox is resolved in a modern context using Einstein’s
physics.
2.2 The bolometric background

At the heart of Einstein’s general relativity lie the field equations
Êµν − 21 Êgµν − gµν = − 8πc4G Ìµν . (2.1)
16
The bolometric background 17
The left-hand side of these equations describes the geometrical structure of

spacetime, while the right-hand side describes the matter and energy content
of the Universe. The metric tensor gµν relates distance and time via ds 2 =
−gµν dx µ dx ν . For the latter we will adopt spherical polar coordinates x µ =
(ct, r, θ, φ), where c is the speed of light, t is cosmic time and r is coordinate
Ê
distance. The Ricci tensor µν and curvature scalar Ê are functions of gµν
Ì
and its derivatives. µν is known as the energy–momentum tensor while G, c
and are constants. About the value of , in particular, we will say more in
chapters 4 and 5. Our sign conventions throughout this book are the same as
those of Weinberg [3].
Observations on the largest scales indicate that the Universe is both isotropic
(similar in all directions) and homogeneous (similar at all places) along slices of
constant time. If spatial isotropy and homogeneity are assumed, then the metric
tensor is of Robertson–Walker form:

dr 2
ds = c dt − R (t)
2 2 2 2
+ r (dθ + sin θ dφ ) .
2 2 2 2
(2.2)
(1 − kr 2 )
Here k(= ±1 or 0) is the curvature constant and the cosmological scale factor
R = R(t) measures the change in distance between comoving objects due
to expansion. Equation (2.2) tells us how to relate distances (integrals over
dr, dθ, dφ) and times (integrals over dt) in a Universe whose geometry fluctuates
in a manner which is driven by its matter and energy content, and described
mathematically by the function R(t).
Let us now consider the problem of adding up the light from all the galaxies
in the Universe in such a way as to arrive at their combined energy density as
received by us in the Milky Way. We begin by considering a single galaxy at
coordinate distance r whose luminosity, or rate of energy emission per unit time,
is given by L(t). When it reaches us, this energy has been spread over an area
π 2π
A = dA = [R0r dθ ][R0r sin θ dφ] = 4π R02r 2 (2.3)
θ=0 φ=0
where we use the metric (2.2) to obtain the area element dA at the present time
t = t0 (figure 2.1). The subscript ‘0’ here and elsewhere denotes quantities taken
at this time, so R0 ≡ R(t0 ) is the present value of the cosmological scale factor.
The intensity, or energy flux per unit area, reaching us from this galaxy is
given by

R(t) 2 L(t) R 2 (t)L(t)
dQ g = = . (2.4)
R0 A 4π R04r 2
Here the subscript ‘g’ denotes a single galaxy, and the two factors of R(t)/R0 are
included to take expansion into account: this stretches the wavelength of the light
in space (reducing its energy), and also spaces the photons more widely apart in
time. These are sometimes known as Hubble’s ‘energy’ and ‘number’ effects [2].
18 The modern resolution and energy
Figure 2.1. Surface area element dA and volume dV of a thin spherical shell, containing
all the light sources at coordinate distance r .
We now consider a multitude of galaxies. Suppose that these are distributed

through space with some physical number density n g (t). In the covariant language
of general relativity, this quantity should be couched in four-dimensional terms.
µ
For this purpose one can imagine a galaxy current Jg , analogous to the flow of
charges in electromagnetic theory. In a homogeneous and isotropic Universe, this
galaxy current may be shown [3] to have the simple form
Jgµ = n g U µ (2.5)
where U µ ≡ (1, 0, 0, 0) is the galaxy four-velocity. The vanishing space

components of this velocity indicate that galaxies in such a Universe are comoving
with the expansion. Let us assume that they are also conserved (i.e. their rates of
formation and destruction by merging or other processes are slow in comparison
µ
to the expansion rate of the Universe). Then Jg obeys a conservation equation
∇µ Jgµ = 0 (2.6)
where ∇µ denotes the covariant derivative. Using the metric (2.2), one can put
this into the form
1 d 3
(R n g ) = 0 (2.7)
R 3 dt
or
−3
R
ng = n . (2.8)
R0
The bolometric background 19
We will henceforth reserve the symbol n to denote comoving number density

(i.e. that measured in a frame which expands with the Universe). Under the
assumption of galaxy conservation, which we shall make for the most part, this
quantity is equal to its value at z = 0 (n = n 0 = constant). For cases where
mergers or other galaxy-non-conserving processes are significant, the comoving
number density becomes a function of redshift, n = n(z). We shall consider one
such case in chapter 3.
Let us now shift our origin so that we are located at the centre of the spherical
shell in figure 2.1, and consider those galaxies located inside the shell which
extends from radial coordinate distance r to r + dr . The volume of this shell
is given with the help of (2.2) by
π 2π
R dr 4π R 3r 2 dr
dV = √ [Rr dθ ][Rr sin θ dφ] = √ . (2.9)
θ=0 φ=0 1 − kr 2 1 − kr 2
We can simplify this by exploiting the fact that the only trajectories of interest are
those of light rays striking our detectors at origin. By definition, these are radial
(dθ = dφ = 0) null geodesics (ds 2 = 0), for which the metric (2.2) relates time t
and coordinate distance r via
R dr
c dt = √ . (2.10)
1 − kr 2
The volume of the shell can then be written
dV = 4π R 2r 2 c dt (2.11)
and the latter may now be thought of as extending between look-back times t0 −t
and t0 − (t + dt), rather than distances r and r + dr .
The total energy received at the origin from the galaxies in the shell is just
the product of their individual intensities (2.4), their number per unit volume (2.8)
and the volume of the shell (2.11):
dQ = dQ g n g dV = cn(t) R̃(t)L(t) dt. (2.12)
Here we have defined the relative scale factor by
R̃(t) ≡ R(t)/R0 . (2.13)
(We will use tildes throughout this book to denote dimensionless quantities taken
relative to their value at the present time t0 .) Integrating over all the spherical
shells between t0 and t0 − tf , where tf is the source formation time, we obtain the
result t0
Q = dQ = c n(t)L(t) R̃(t) dt. (2.14)
tf
Equation (2.14) defines the bolometric intensity of the extragalactic background
light (EBL). This is the energy received by us (over all wavelengths of the
electromagnetic spectrum) per unit area, per unit time, from all the galaxies which
have been shining since time tf . In principle, if we let tf → 0, we will encompass
the entire history of the Universe since the big bang. Although this sometimes
provides a useful mathematical shortcut, we will see in later sections that it is
physically more realistic to cut the integral off at a finite formation time. The
quantity Q is a measure of the amount of light in the Universe, and Olbers’
‘paradox’ is merely another way of asking why it is low.
2.3 From time to redshift

While the cosmic time t is a useful independent variable for theoretical purposes,
it is not directly observable. In studies aimed at making contact with eventual
observation it is better to work in terms of redshift z, which is the relative shift in
wavelength λ of a light between the time it is emitted and observed:
λ R0 − R(t)
z≡ = = R̃ −1 − 1. (2.15)
λ R(t)
Differentiation gives
R0 dR R0 Ṙ dt
dz = − =− (2.16)
R2 R2
where an overdot signifies the time derivative. Inverting, we obtain
R 2 dz dz
dt = − =− . (2.17)
R0 Ṙ (1 + z)H (z)
Here we have introduced a new quantity, Hubble’s parameter:
H ≡ Ṙ/R (2.18)
which is the expansion rate of the Universe. Its value at the present time is known
as Hubble’s constant (H0). The numerical size of this latter quantity continues
to be debated by observational cosmologists. For this reason it is usually written
in the form
H0 = 100h 0 km s−1 Mpc−1 = 0.102h 0 Gyr−1 . (2.19)
Here 1 Gyr ≡ 109 yr and the uncertainties have been absorbed into a
dimensionless parameter h 0 whose value is currently thought to be in the range
0.6 h 0 0.9. We will have more to say about the observational status of h 0 in
chapter 4.
Putting (2.17) into the bolometric EBL intensity integral (2.14), and using
(2.15) to replace R̃ with (1 + z)−1 , we obtain
zf
n(z)L(z) dz
Q=c . (2.20)
0 (1 + z) H (z)
2
From time to redshift 21
Here z f is the redshift of galaxy formation, and we have written the comoving
number density and luminosity as functions of redshift rather than time. One
may think of n(z) and L(z) as describing the physics of the sources, the two
factors of (1 + z) as representing the dilution and stretching of the light signals
and H (z) as governing the evolution of the background spacetime. In practical
terms, the transition from an integral over time (2.14) to one over redshift (2.20)
is immensely valuable and might be likened to arming a traffic officer with a radar
gun rather than a stopwatch.
For some problems, and for this chapter in particular, the physics of the
sources themselves are of secondary importance, and it is reasonable to take
L(z) = L 0 and n(z) = n 0 as constants over the range of redshifts of interest.
In this case the integral (2.20) reads simply

cn 0 L 0 zf dz
Q= . (2.21)
H0 0 (1 + z)2 H̃ (z)
Here we have defined the relative expansion rate by
H̃ (z) ≡ H (z)/H0. (2.22)
(This is merely Hubble’s parameter, normalized to its present value.) The form of
the function H̃ (z), which will be of central importance throughout this book, is
derived in the section that follows.
All the dimensional information in (2.21) is now contained in the constant
outside the integral. In the numerator of this constant we may identify the present
comoving luminosity density of the Universe,
0 ≡ n 0 L 0 . (2.23)
This can be measured experimentally by counting galaxies down to some

faint limiting apparent magnitude, and extrapolating to fainter ones based on
assumptions about the true distribution of absolute magnitudes. A recent
compilation of seven such studies over the past decade is that of Fukugita et al [5]
in the B-band:
0 = (2.0 ± 0.2) × 108 h 0 L Mpc−3

= (2.6 ± 0.3) × 10−32 h 0 erg s−1 cm−3 . (2.24)
(This waveband, centred near 4400 Å, is where galaxies emit most of their light,
and is also close to the wavelength of peak sensitivity of human eyesight.) We
will use this number throughout our book. Recent data from the Sloan Digital
Sky Survey (SDSS) suggest a somewhat higher value with larger uncertainty,
(2.41 ± 0.39) × 108h 0 L Mpc−3 [6]; while a preliminary result from the Two
Degree Field (2dF) team is slightly lower, 0 = (1.82 ± 0.17) × 108h 0 L Mpc−3
[7]. If the final result inferred from large-scale galaxy surveys of this kind proves
to be significantly different from that in (2.24), then our EBL intensities (which
are proportional to 0 ) would go up or down accordingly.
Numerically, the constant factor outside the integral (2.21) sets the order of
magnitude of the integral itself, so it is of interest to see what value this takes.
Denoting it by Q ∗ we find using (2.24) that
Q ∗ ≡ c0 /H0 = (2.5 ± 0.2) × 10−4 erg s−1 cm−2 . (2.25)
There are two important things to note about this quantity. First, because the
factors of h 0 attached to both 0 and H0 cancel each other out, it is independent
of the uncertainty in Hubble’s constant. This is not always appreciated but
was first emphasized by Felten [8]. Second, the value of Q ∗ is very small
by everyday standards: more than a million times fainter than the bolometric
intensity produced by a 100 W bulb in an average-sized room (section 1.4). The
smallness of this number already contains the essence of the resolution of Olbers’
paradox. But to put the latter to rest in a definitive fashion, we need to evaluate
the full integral (2.20). This, in turn, requires the Hubble expansion rate H̃ (z),
which we proceed to derive next.
2.4 Matter and energy

The relative expansion rate H̃ (z) can be obtained from the field equations of
general relativity, if the matter and energy content of the Universe are specified.
Under the assumptions of isotropy and homogeneity, the latter two quantities can
be described [3] by an energy–momentum tensor of perfect fluid form
Ìµν = (ρ + p/c2 )Uµ Uν + pgµν . (2.26)
Here ρ is the density of the cosmic fluid and p its pressure. These two quantities,
in turn, are related by an equation of state
p = (γ − 1)ρc2 . (2.27)
Three specific equations of state are of particular relevance in cosmology and will
make regular appearances in the chapters that follow. They all have γ = constant,
as follows:
• radiation (γr = 43 ) : pr = 13 ρr c2
• dustlike matter (γm = 1) : pm = 0 (2.28)
• vacuum energy (γv = 0) : pv = −ρv c2
The first of these is a good approximation to the early Universe, when conditions
were so hot and dense that matter and radiation existed in nearly perfect
thermodynamic equilibrium (the radiation era). The second has usually been
used to model the present Universe, since we know that the energy density of
Matter and energy 23
electromagnetic radiation now is far below that of dustlike (pressure-free) matter.

The third may be a good description of the future state of the Universe, if recent
measurements of the magnitudes of high-redshift of Type Ia supernovae are borne
out. These indicate that the vacuum energy may already be more important than
all other contributions to the density of the Universe, including those from any
unseen dark-matter component.
A vacuum-like equation of state has also often been discussed as a possible
description of conditions during an ‘inflationary’ phase in the early stages of (or
perhaps even preceding) the radiation era. Various values have been proposed
for the index γ in this context, from zero (de Sitter inflation) to 1/3 (domain
walls) and 2/3 (cosmic strings). Inflation, or non-negative acceleration of the
scale factor, occurs for any perfect fluid with 0 γ 2/3, and cosmological
matter with this property has come to be known as quintessence (a reference to
the substance making up the realm beyond the planetary spheres in Aristotelian
astronomy). Such a fluid might be realized physically, for instance, in an early
Universe dominated by a minimally-coupled scalar field whose equation of state
would have γϕ = ϕ̇ 2 /[ϕ̇ 2 /2 + V (ϕ)], where ϕ is the value of the scalar field
and V (ϕ) its potential energy. This is ‘quintessential’ (γϕ 2/3) whenever the
field’s potential energy exceeds its kinetic energy, V (ϕ) ϕ̇ 2 . A pure vacuum
(γϕ ≈ 0) is recovered in the ‘slow-roll’ limit, V (ϕ) ϕ̇ 2 . In the opposite limit,
V (ϕ) ≈ 0, one has γϕ ≈ 2 and the scalar fluid equation of state approximates that
of stiff matter, ps = ρs c2 . The index γ is not necessarily constant, especially
for situations involving multiple interacting fluids. For our purposes, however, it
will be sufficient to assume that a single component dominates the cosmic fluid
during any given epoch, so that the Universe can be described by one of the cases
in (2.28).
Assuming that energy and momentum are neither created nor destroyed, one
µ
can proceed exactly as with the galaxy current Jg . The conservation equation in
this case reads
∇ µ Ìµν = 0. (2.29)
This reduces to
1 d 3 2 dp
3
[R (ρc + p)] = (2.30)
R dt dt
which may be compared with (2.7) for galaxies. Equation (2.30) is solved with
the help of the equation of state (2.27) to yield
ρ(t) = ρ0 [R(t)/R0 ]−3γ . (2.31)
In particular, for the single-component fluids in (2.28):
ρr = ρr,0 (R/R0 )−4 (radiation)

ρm = ρm,0 (R/R0 )−3 (dustlike matter) (2.32)
ρv = ρv,0 = constant (vacuum energy)
These expressions will frequently prove useful in later chapters. They are also
applicable to cases in which several components are present, but only if these do
not exchange energy at significant rates (i.e. relative to the expansion rate), so that
each is in effect conserved separately.
2.5 The expansion rate

With geometry as described by the metric (2.2), and matter–energy as described
by the energy–momentum tensor (2.26), we are in a position to solve the Einstein
field equations (2.1) for cosmology. This results in two differential equations for
the scale factor R and its time-derivatives. The equation for the expansion rate H
in terms of R is
kc2 c2 8π G
H2 + 2 − = ρ. (2.33)
R 3 3
We will take the cosmological fluid to consist of both radiation and dustlike matter
components so that ρ = ρr + ρm . It then proves useful to define two new
quantities. One is the vacuum energy density associated with the cosmological
constant:
c4
ρ c2 ≡ = constant. (2.34)
8π G
The corresponding pressure is given by (2.28) as p = −ρ c2 . The second
useful quantity is known as the critical density, for reasons that will become
clear momentarily. It reads:
3H 2(t)
ρcrit (t) ≡ . (2.35)
8π G
In particular the present value of this quantity is
3H02
ρcrit,0 ≡ = (2.78 × 1011)h 20 M Mpc−3
8π G
= (1.88 × 10−29)h 20 g cm−3 . (2.36)
This works out to the equivalent of between 4.0 protons per cubic metre (if
h 0 = 0.6) and 9.1 protons per cubic metre (if h 0 = 0.9). Using (2.34) and
(2.36) we can evaluate (2.33) at the present time as follows:

kc2 2 ρr,0 + ρm,0 + ρ − ρcrit,0
= H0 . (2.37)
R02 ρcrit,0
The physical significance of ρcrit,0 is now apparent: its value determines the
spatial curvature of the Universe. If the sum of the densities ρr,0 , ρm,0 and ρ is
exactly equal to ρcrit,0 then k = 0. The Universe in this case is flat or Euclidean.
Such a Universe is unbounded and infinite in extent. Alternatively, if the sum
The expansion rate 25
Figure 2.2. The spatial geometry of the Universe (here represented as a two-dimensional
surface) is determined by its total density, expressed in units of the critical density ρcrit,0
by the parameter tot,0 . If tot,0 = 1 then the Universe is flat (‘Euclidean’) and the
curvature parameter k can be set to zero in the metric (2.2). If tot,0 > 1 then the Universe
is positively curved (spherical) and k can be set to +1. If tot,0 < 1 then it is negatively
curved (hyperbolic) and k = −1. (Figure courtesy E C Eekels.)
of ρr,0 , ρm,0 and ρ exceeds ρcrit,0 , then k > 0 and the Universe is positively
curved. This is often referred to as a ‘k = +1 model’ because the magnitude
of the curvature constant k can be normalized to unity by choice of units for R0 .
Spatial hypersurfaces in this model are spherical in shape, and the Universe is
unbounded, but closed and finite. Finally, if ρr,0 + ρm,0 + ρ < ρcrit,0 , then the
Universe has negative curvature (k = −1) and hyperbolic spatial sections. It is
open, unbounded and infinite in extent. (Strictly speaking one can also obtain
flat and hyperbolic solutions which are finite by adopting a non-trivial topology
and suitably ‘identifying’ pairs of points; we do not pursue this here.) All three
possibilities are depicted in figure 2.2.
To simplify the notation we now rescale all our physical densities, defining
three dimensionless density parameters:
ρr,0 ρm,0 ρ c2
r,0 ≡ m,0 ≡ ,0 ≡ = . (2.38)
ρcrit,0 ρcrit,0 ρcrit,0 3H02
These parameters occupy central roles throughout the book, and we will usually
refer to them for brevity as ‘densities’ (or ‘present densities’ where this is not
clear from the context). Defining tot,0 ≡ r,0 + m,0 + ,0 , we find from
(2.37) that the curvature constant is
k = (R0 H0 /c)2 (tot,0 − 1). (2.39)
Thus flat models are just those with tot,0 = 1, while closed and open models
have tot,0 > 1 and tot,0 < 1 respectively. Substituting (2.39) back into (2.33),
using the definitions (2.38) and recalling from (2.15) that z = R0 /R − 1, we
obtain the expansion rate of the Universe:
H̃ (z) = [r,0 (1 + z)4 + m,0 (1 + z)3 + ,0

− (tot,0 − 1)(1 + z)2 ]1/2. (2.40)
When r,0 = 0, this is sometimes referred to as the Friedmann equation after

the Russian mathematician who obtained the first cosmological solutions of
Einstein’s equations under the assumption of zero pressure. This restriction was
first lifted several years later by Lemaı̂tre, so we refer to (2.40) as the Friedmann–
Lemaı̂tre equation.
The form of this equation already reveals a great deal about the contents
and evolution of the Universe. For example, the term m,0 (1 + z)3 shows that
matter acts to increase the expansion rate H (z) as one goes to higher z—that is,
to slow down the expansion rate with time. This is the braking effect of matter’s
gravitational self-attraction, and (as we shall learn in chapter 4) it appears to be
generated almost entirely by matter which we cannot see.
The term r,0 (1 + z)4 shows that radiation has the same effect, but with
a stronger dependence on redshift (this is related to the fact that pressure, as
well as density, acts as a source of gravitation in general relativity). As one
moves backward in time, photons (and relativistic particles) therefore become
increasingly important compared to pressureless matter. In fact, the dynamics
of the early Universe (at redshifts above z 104 ) must have been completely
dominated by them. The total present radiation density r,0 is, however, several
orders of magnitude below that of non-relativistic matter. (This is inferred, not
only from measurements such as those of the COBE satellite, but also from
the fact that too much pressure would have slowed expansion so much that the
Universe could not have lived long enough to contain the oldest stars.) In later
chapters we will usually be interested only in the matter-dominated era (or later),
and we can safely neglect the radiation term in (2.40).
The term ,0 is independent of redshift, which means that its influence
is not diluted with time. Any Universe with > 0 will therefore eventually
be dominated by vacuum energy. In the limit t → ∞, in fact, the other terms
drop out of (2.40) altogether and the vacuum energy density reads simply ,0 =
(H∞ /H0)2 where H∞ is the limiting value of H (t) as t → ∞ (assuming that this
latter quantity exists; i.e. that the Universe does not recollapse). From (2.38) it
then follows that
c2 = 3H∞ 2
. (2.41)
The static analogue 27
This constitutes a link between (a constant of nature in Einstein’s theory) and

the asymptotic expansion rate H∞ (a dynamical parameter). If > 0, then we
will necessarily measure ,0 ∼ 1 at late times, regardless of the microphysical
origin of the vacuum energy.
The last term in equation (2.40), finally, shows that an excess of tot,0 over
one (i.e. a positive curvature) acts to offset the contribution of the first three terms
to the expansion rate, while a deficit (i.e. a negative curvature) enhances them.
Open models, in other words, expand more quickly at any given redshift z (and
therefore last longer) than closed ones. This curvature term, however, goes only
as (1 + z)2 , which means that its importance drops off relative to the matter and
radiation terms at early times, and relative to the vacuum term at late ones.
It became commonplace during the 1980s to work with a simplified version
of equation (2.40), in which not only the first (radiation) term on the right-hand
side was neglected, but the third (vacuum) and fourth (curvature) terms as well.
This Einstein–de Sitter (EdS) model appeared reasonable at the time, for four
principal reasons. First, the four terms in question differ sharply from each other
in their dependence on redshift z, and the probability that we should happen to
find ourselves in an era when they have similar values would seem a priori very
remote. By this argument, which goes back to Dicke [9], it was felt that only one
term ought to dominate at any given time. Second, the vacuum term was regarded
with particular suspicion for reasons to be discussed in chapter 4. Third, a period
of inflation was asserted to have driven tot (t) to unity. (This is still widely
believed, but depends on the initial conditions preceding inflation, and does not
necessarily hold in all plausible models [10].) And finally, the EdS model was
favoured on grounds of simplicity.
These arguments are no longer compelling today, and the determination of
r,0 , m,0 and ,0 (along with H0) has shifted largely back into the domain
of observation. We defer a fuller discussion of these issues to chapter 4, being
content here to assume merely that radiation and matter densities are positive and
not too large (0 r,0 1.5 and 0 m,0 1.5), and that vacuum energy
density is neither too large nor too negative (−0.5 ,0 1.5).
2.6 The static analogue

This is a good place to pause and take stock of our results so far. In the foregoing
sections we have obtained a simple integral for the bolometric intensity of the
Universe, equation (2.21):
zf
dz
Q = Q∗ . (2.42)
0 (1 + z)2 H̃ (z)
The value of Q ∗ is given by (2.25). We have also obtained a general expression

for the expansion rate H̃ (z) in terms of the present densities of radiation, matter
and vacuum energy, equation (2.40). Strictly speaking, (2.42) assumes constant
galaxy luminosity and comoving number density, as discussed in section 2.3. This
is adequate for our purposes, because we are not primarily concerned here with
the physics of the galaxies themselves. We wish to obtain a definitive answer to
the question posed in chapter 1: is it cosmic expansion, or the finite age of the
Universe, which is primarily responsible for the low value of Q?
The relative importance of these two factors continues to be a subject of
controversy and confusion (see [4] for a review). In particular there is a lingering
perception that general relativity ‘solves’ Olbers’ paradox chiefly because the
expansion of the Universe stretches and dims the light it contains.
There is a simple way to test this supposition using the formalism we have
already laid out, and that is to ‘turn off’ expansion by setting the scale factor of
the Universe equal to a constant value, R(t) = R0 . Then R̃(t) = 1 from (2.13),
and the bolometric intensity of the EBL is given by (2.14) as an integral over time:
t0
Q stat = Q ∗ H0 dt. (2.43)
tf
Here we have taken n = n 0 and L = L 0 for convenience, as before, and used

(2.25) for Q ∗ . The subscript ‘stat’ denotes the static analogue of Q; that is, the
bolometric EBL intensity that one would measure in a Universe which did not
expand. Equation (2.43) shows that this is nothing more than the length of time
for which the galaxies have been shining, measured in units of Hubble time (H0−1)
and scaled by Q ∗ .
Now, we wish to compare (2.42) in the expanding Universe with its static
analogue (2.43), while keeping all other factors the same. In particular, if the
comparison is to be meaningful, the lifetime of the galaxies should be identical.
This is just the integral over dt, which may—in an expanding Universe—be
converted to one over dz by means of (2.17):
t0 zf
dz
H0 dt = (2.44)
tf 0 (1 + z) H̃ (z)
where we have used the definition (2.22) of H̃ (z). In a static Universe, of course,
redshift does not carry its usual physical significance. But nothing prevents us
from retaining z as an integration variable. Substitution of (2.44) into (2.43) then
yields zf
dz
Q stat = Q ∗ . (2.45)
0 (1 + z) H̃ (z)
We emphasize that z and H̃ are to be seen here as algebraic parameters whose

usefulness lies in the fact that they ensure consistency in age between the static
and expanding pictures.
Equation (2.42) and its static analogue (2.45) allow us to make a meaningful
comparison of the bolometric intensity of the EBL in models which are alike
in all respects, except that one is expanding while the other stands still. The
A quantitative resolution 29
two equations are almost identical, differing only by an extra factor of (1 + z)−1
attached to the integrand in the expanding case. Since this is less than unity for
all z, we expect Q to be less than Q stat , although the magnitude of the difference
will depend on the shape of H̃ (z). The latter must have been fairly smooth over
the lifetime of the galaxies in any plausible model, so that the difference between
the two integrals cannot amount to much more than a few orders of magnitude. It
follows immediately that the intensity of the EBL in expanding models (as in their
static analogues) must be determined primarily by the lifetime of the galaxies, so
that the effects of expansion are secondary. We proceed in the next section to
make this conclusion quantitative.
2.7 A quantitative resolution

The most straightforward way to do this is to evaluate the ratio Q/Q stat for a
wide range of values of the cosmological parameters r,0 , m,0 and ,0 . If
we find that Q/Q stat 1 over much of this phase space, then expansion would
reduce Q quite significantly from what it would otherwise be in an equivalent
static Universe. Conversely, values of Q/Q stat ≈ 1 would confirm that expansion
has little effect.
This section, then, is a largely technical one, where we calculate Q/Q ∗ ,
Q stat /Q ∗ and Q/Q stat for the widest possible range of models and present the
results in analytic or graphical form. For readers who are interested primarily
in final results and prefer to skip the details of this exercise, we give here the
conclusion, which is that the ratio Q/Q stat turns out to lie in the range 0.6 ± 0.1
across nearly the entirety of the cosmological phase space. While expansion does
reduce the bolometric intensity of the EBL, in other words, the reduction is slight.
If we could freeze expansion without affecting any of the other relevant factors,
the total amount of light reaching the Milky Way would go up by a factor of less
than two in any plausible cosmological model.
Let us begin with the simplest examples, in which the Universe has one
critical-density component or contains nothing at all (table 2.1). Consider first the
radiation model with a critical density of radiation or ultrarelativistic particles
(r,0 = 1) but m,0 = ,0 = 0. Bolometric EBL intensity in the expanding
Table 2.1. Simple flat and empty models

Model name r,0 m,0 ,0 1 − tot,0
Radiation 1 0 0 0
Einstein–de Sitter 0 1 0 0
De Sitter 0 0 1 0
Milne (empty) 0 0 0 1
Universe is, from (2.42),

1+z f
Q dx 21/64 (z f = 3)
= = (2.46)
Q∗ 1 x4 1/3 (z f = ∞)
where x ≡ 1 + z. The corresponding result for a static model is given by (2.45)
as 1+zf
Q stat dx 15/32 (z f = 3)
= = (2.47)
Q∗ 1 x 3 1/2 (z f = ∞).
Here we have chosen illustrative lower and upper limits on the redshift of galaxy
formation (z f = 3 and ∞ respectively). The actual value of this parameter has
not yet been determined, although there are now indications that z f may be as
high as six. Arbitarily large values of z f have little physical meaning within the
context of calculations of the EBL, since these presume that enough time has
elapsed since the big bang for stars and galaxies to have formed, and this implies
that z f is bounded above at some reasonable value. In any case, it may be seen
from the previous results that the overall EBL intensity is rather insensitive to this
parameter. Increasing z f lengthens the period over which galaxies radiate, and
this increases both Q and Q stat . The ratio Q/Q stat , however, is given by

Q 7/10 (z f = 3)
= (2.48)
Q stat 2/3 (z f = ∞)
and this changes but little. We will find this to be true in general.
Consider next the Einstein–de Sitter model, which has a critical density of
dustlike matter (m,0 = 1) with r,0 = ,0 = 0. Bolometric EBL intensity in
the expanding Universe is, from (2.42),
1+zf
Q dx 31/80 (z f = 3)
= = (2.49)
Q∗ 1 x 7/2 2/5 (z f = ∞).
The corresponding static result is given by (2.45) as
1+zf
Q stat dx 7/12 (z f = 3)
= = (2.50)
Q∗ 1 x 5/2 2/3 (z f = ∞).
The ratio of EBL intensity in an expanding Einstein–de Sitter model to that in the
equivalent static model is thus

Q 93/140 (z f = 3)
= (2.51)
Q stat 3/5 (z f = ∞).
These numbers are only slightly below those of the radiation case.
A third simple case is the de Sitter model, which consists entirely of vacuum
energy (,0 = 1), with r,0 = m,0 = 0. Bolometric EBL intensity in the
expanding case is, from (2.42),
1+zf
Q dx 3/4 (z f = 3)
= = (2.52)
Q∗ 1 x 2 1 (z f = ∞).
Equation (2.45) gives for the equivalent static case

1+zf
Q stat dx ln 4 (z f = 3)
= = (2.53)
Q∗ 1 x ∞ (z f = ∞).
The ratio of EBL intensity in an expanding de Sitter model to that in the equivalent
static model is then

Q 3/(4 ln 4) (z f = 3)
= (2.54)
Q stat 0 (z f = ∞).
The de Sitter Universe is older than other models, which means it has more time
to fill up with light, so intensities are higher. In fact, Q stat (which is proportional
to the lifetime of the galaxies) goes to infinity as z f → ∞, driving Q/Q stat to
zero in this limit. (It is thus possible to ‘recover Olbers’ paradox’ in the de Sitter
model, as noted by White and Scott [11].) Such a limit is, however, unphysical
as noted earlier. For realistic values of z f one obtains values of Q/Q stat which are
only slightly lower than those in the radiation and matter cases.
Finally, we consider the Milne model, which is empty of all forms of matter
and energy (r,0 = m,0 = ,0 = 0), making it an idealization but one which
has nevertheless often proved useful as a bridge between the special and general
theories of relativity. Bolometric EBL intensity in the expanding case is given by
(2.42) as
1+zf
Q dx 15/32 (z f = 3)
= = (2.55)
Q∗ 1 x3 1/2 (z f = ∞)
which is identical to equation (2.47) for the static radiation model. The
corresponding static result is given by (2.45) as
1+zf
Q stat dx 3/4 (z f = 3)
= = (2.56)
Q∗ 1 x 2 1 (z f = ∞)
which is the same as equation (2.52) for the expanding de Sitter model. The ratio
of EBL intensity in an expanding Milne model to that in the equivalent static
model is then
Q 5/8 (z f = 3)
= (2.57)
Q stat 1/2 (z f = ∞).
This again lies close to previous results. In fact, we have found in every case
(except the z f → ∞ limit of the de Sitter model) that the ratio of bolometric EBL
intensities with and without expansion lies in the range 0.4 º Q/Q stat º 0.7.
It may, however, be that models whose total density is neither critical nor
zero have different properties. To check this we expand our investigation to the
wider class of open and closed models. Equations (2.42) and (2.45) can be solved
analytically for these cases, if they are dominated by a single component. We
collect the solutions for convenience in appendix A, and plot them in figures 2.3–
2.5 for radiation-, matter- and vacuum-dominated models respectively. In each
Figure 2.3. Ratios Q/Q ∗ (long-dashed lines), Q stat /Q ∗ (short-dashed lines) and Q/Q stat
(unbroken lines) as a function of radiation density r,0 . Bold lines are calculated for z f = 3
while light ones have z f = ∞.
figure, long-dashed lines correspond to EBL intensity in expanding models

(Q/Q ∗ ) while short-dashed ones show the equivalent static quantities (Q stat /Q ∗ ).
The ratio of these two quantities (Q/Q stat ) is indicated by unbroken lines. The
bold curves have z f = 3 while light ones are calculated for z f = ∞.
In the cases of the radiation- and matter-dominated Universes, figures 2.3

and 2.4 show that while the individual intensities Q/Q ∗ and Q stat /Q ∗ do vary
significantly with r,0 and m,0 , their ratio Q/Q stat remains nearly constant
across the whole of the phase space, staying inside the range 0.5 º Q/Q stat º 0.7
for both models.
The case of the vacuum-dominated Universe appears more complicated, but

shows the same trend (figure 2.5). Absolute EBL intensities Q/Q ∗ and Q stat /Q ∗
differ noticeably from those in the radiation- and matter-dominated models, but
their ratio (full lines) is again close to flat. The exception occurs as ,0 → 1 (the
de Sitter model), where Q/Q stat can dip well below 0.5 in the limit of large z f , as
discussed earlier. As ,0 rises above one, the big bang disappears altogether (in
models with r,0 = m,0 = 0), and one finds instead a big bounce (i.e. a positive
minimum scale factor at the beginning of the expansionary phase). Hence there
(unbroken lines) as a function of matter density m,0 . Bold lines are calculated for z f = 3
while light ones have z f = ∞.
is a maximum possible redshift z max given by

,0
1 + z max = . (2.58)
,0 − 1
While such models are rarely considered, it is interesting to note that the same
pattern persists here. In light of (2.58), one can no longer integrate out to
arbitrarily high formation redshift z f . If one wants to integrate to at least z f , then
one is limited to vacuum densities less than ,0 < (1 + z f )2 /[(1 + z f )2 − 1].
In the case z f = 3 (shown with heavy lines), this corresponds to an upper limit
of ,0 < 16/15 (faint dotted line). More generally, for ,0 > 1 the limiting
value of EBL intensity (shown with light lines) is reached as z f → z max rather
than z f → ∞ for both expanding and static models. Over the entire parameter
space −0.5 ,0 1.5 (except in the immediate vicinity of ,0 = 1),
figure 2.5 shows that 0.4 º Q/Q stat º 0.7 as before.
When more than one component of matter is present, analytic expressions
for the bolometric intensity can be found in only a few special cases, and the
ratios Q/Q ∗ and Q stat /Q ∗ must, in general, be computed numerically. We show
the results in figure 2.6 for the case which is of most physical interest: a Universe
containing both dustlike matter (m,0, horizontal axis) and vacuum energy (,0 ,
(unbroken lines) as a function of vacuum energy density ,0 . Bold lines are calculated
for z f = 3 while light ones have z f = ∞ for ,0 1 and z f = z max for ,0 > 1.
The dotted vertical line marks the maximum value of ,0 for which one can integrate to
z f = 3.
vertical axis), with r,0 = 0. This is a contour plot, with five bundles of equal-
EBL intensity contours for the expanding Universe (labelled Q/Q ∗ = 0.37, 0.45,
0.53, 0.61 and 0.69). The bold (unbroken) lines are calculated for z f = 5, while
medium (long-dashed) lines assume z f = 10 and light (short-dashed) lines have
z f = 50. Also shown is the boundary between big bang and bounce models
(bold line in top left corner), and the boundary between open and closed models
(diagonal dashed line). Similar plots could be produced for the r,0 –m,0 and
,0 –r,0 planes.
Figure 2.6 shows that the bolometric intensity of the EBL is only modestly
sensitive to the cosmological parameters m,0 and ,0 . Moving from the lower
right-hand corner of the phase space (Q/Q ∗ = 0.37) to the upper left-hand one
(Q/Q ∗ = 0.69) changes the value of this quantity by less than a factor of two.
Increasing the redshift of galaxy formation from z f = 5 to 10 has little effect, and
increasing it again to z f = 50 even less. This means that essentially all of the light
reaching us from outside the Milky Way comes from galaxies at z < 5, regardless
of the redshift at which these objects actually formed.
While figure 2.6 confirms that the night sky is dark in any reasonable
Light at the end of the Universe? 35
Figure 2.6. The ratio Q/Q ∗ in an expanding Universe, plotted as a function

of matter-density parameter m,0 and vacuum-density parameter ,0 (with the
radiation-density parameter r,0 set to zero). Unbroken lines correspond to z f = 5, while
long-dashed lines assume z f = 10 and short-dashed ones have z f = 50.
cosmological model, figure 2.7 shows why. It is a contour plot of Q/Q stat , the
value of which varies so little across the phase space that we have had to restrict
the range of z f -values in order to keep the diagram from being too cluttered. The
bold (unbroken) lines are calculated for z f = 4.5, the medium (long-dashed) lines
for z f = 5, and light (short-dashed) lines for z f = 5.5. The spread in contour
values is extremely narrow, from Q/Q stat = 0.56 in the upper left-hand corner
to 0.64 in the lower right-hand one—a difference of less than 15%. Figure 2.6
confirms our previous analytical results and leaves no doubt about the resolution
of Olbers’ paradox: the brightness of the night sky is determined to order of
magnitude by the lifetime of the galaxies, and is reduced by no more than a factor
of 0.6 ± 0.1 due to the expansion of the Universe.
2.8 Light at the end of the Universe?

We have obtained a general integral (2.42) and a number of exact expressions for
the bolometric intensity Q of the EBL in an expanding Universe. These results
are phrased in terms of the densities r,0 , m,0 and ,0 of radiation, matter and
vacuum energy, as well as the redshift z f at which the galaxies formed. All four of
Figure 2.7. The ratio Q/Q stat of EBL intensity in an expanding Universe to that in a static
Universe with the same values of the matter-density parameter m,0 and vacuum-density
parameter ,0 (with the radiation-density parameter r,0 set to zero). Unbroken lines
correspond to z f = 4.5, while long-dashed lines assume z = 5 and short-dashed ones have
z = 5.5.
these parameters are ones that we can, in principle, measure with our instruments.
However, it is occasionally useful to shift attention from redshift z back to time t
in order to ask questions which, while not directly connected with experiment, are
of conceptual interest. In this section we ask: do the foregoing results mean that
the brightness of the night sky is changing with time? If so, is it getting brighter
or darker? How quickly?
To investigate these questions one would like the EBL intensity expressed as
a function of time, as in (2.14). Evaluation of this integral requires a knowledge
of the scale factor R(t), which is not well constrained by observation. An
approximate expression can, however, be obtained if the Universe has a density
very close to the critical one, as suggested by observations of the power spectrum
of the CMB (see chapter 4). In this case k = 0 and (2.33) simplifies to
2
Ṙ 8π G
= (ρr + ρm + ρ ) (2.59)
R 3
where we have set ρ = ρr + ρm and used (2.34) for ρ . If we now assume that
only one of these three components is dominant at a given time, then we can make
use of (2.32) to obtain


2 −4
Ṙ 8π G  ρr,0 (R/R0 ) −3 (radiation)
= × ρm,0 (R/R0 ) (matter) (2.60)
R 3 
ρ (vacuum).
These differential equations are separable and have solutions

R(t)  (t/t0 )2/3
1/2 (radiation)
= (t/t0 ) (matter) (2.61)
R0 
exp[H0(t − t0 )] (vacuum).
We emphasize that these results assume (i) spatial flatness and (ii) a single-
component cosmic fluid which must have the critical density.
Putting (2.61) into (2.14), we can solve for the bolometric intensity under
the assumption that the luminosity of the galaxies is constant over their lifetimes,
L(t) = L 0 : 
Q(t)  (1/3)(t/t0 )5/3
3/2 (radiation)
= (2/5)(t/t0 ) (matter) (2.62)
Q∗ 
exp[H0t0 (t/t0 − 1)] (vacuum)
where we have used (2.25) and assumed that tf t0 and tf t.
The intensity of the light reaching us from intergalactic space climbs as t 3/2
in a radiation-filled Universe, t 5/3 in a matter-dominated one, and exp(H0t) in
one which contains only vacuum energy. This happens because the horizon of
the Universe expands to encompass more and more galaxies, and hence more
photons. Clearly it does so at a rate which more than compensates for the dilution
and redshifting of existing photons due to expansion. Suppose, for argument’s
sake, that this state of affairs could continue indefinitely. How long would it take
for the night sky to become as bright as a well-lit room (Q ∼ 1000 erg cm−2 s−1 )?
The required increase of Q(t) over Q ∗ (= 2.5 × 10−4 erg cm−2 s−1 ) is
4 million times. Equation (2.62) then implies that
780 000 Gyr (radiation)
t≈ 240 000 Gyr (matter) (2.63)
250 Gyr (vacuum)
where we have taken H0 t0 ≈ 1 and t0 ≈ 16 Gyr as suggested by observational
data (chapter 4).
The last of the numbers in (2.63) is particularly intriguing. In a vacuum-
dominated model in which the luminosity of the galaxies could be kept constant
indefinitely, the sky would fill up with light over timescales of the same order
as the theoretical hydrogen-burning lifetimes of the longest-lived stars, those with
M º 0.3M [12]. Of course, the luminosity of galaxies cannot stay constant over
these timescales, because most of their light comes from much more massive stars
which burn themselves out after tens of Gyr or less. Still, the closeness of these
numbers prompts us to ask: is this steep increase in theoretical EBL intensity with
time a feature only of the pure de Sitter model; or does it also arise in models
containing matter along with vacuum energy?
To answer this, we would like to find a simple expression for R(t) in models
with both dustlike matter and vacuum energy. An analytic solution does exist
for such models, if they are flat (,0 = 1 − m,0 ). Its usefulness goes well
beyond the particular problem at hand and, since we have found it derived only in
German [13], it is worth reproducing here. Taking ρm (R) and ρ from (2.32) and
using the definitions (2.38), we find that (2.59) leads to the following differential
equation in place of (2.60):
2 −3
Ṙ R
= H0 m,0
2
+ ,0 . (2.64)
R R0
This is solved by making a change of variables from R to u ≡ R/R∗ , where R∗

is the value of the scale factor at the inflection point, R∗ /R0 = (m,0 /2,0 )1/3 .
It is then found that
u̇ 2 = H02,0 (u 2 + 2/u). (2.65)
This may be inverted for the age of the Universe:

du 1 du
t= = . (2.66)
u̇ H0 ,0 u + 2/u
2
One further change of variable to v ≡ u 3 + 1 puts this integral into elementary

form:
1 dv
t= √ . (2.67)
3H0 ,0 v2 − 1
Solving for t = t (R) and reinverting to get R̃(t) = R(t)/R0 , we find

m,0 2/3
R̃(t) = sinh 2 1 − m,0 H0t
3
(2.68)
1 − m,0
where we have replaced ,0 with 1 − m,0 . This elegant formula has many
uses and deserves to be known more widely, given the importance of vacuum-
dominated models in modern cosmology. Differentiation with respect to time
gives the Hubble expansion rate:

H̃ (t) = 1 − m,0 coth 32 1 − m,0 H0t . (2.69)

This goes over to ,0 as t → ∞, a result which (as noted in section 2.5) holds
quite generally for models with > 0. Alternatively, setting R̃ = (1 + z)−1 in
(2.68) gives the age of the Universe at a redshift z:

2 −1 1 − m,0
t= sinh . (2.70)
3H0 1 − m,0 m,0 (1 + z)3
Figure 2.8. Plots of Q(t)/Q ∗ in flat models containing both dustlike matter (m,0 ) and
vacuum energy (,0 ). The unbroken line is the Einstein–de Sitter model, while the
short-dashed line is pure de Sitter, and long-dashed lines represent intermediate models.
The curves do not meet at t = t0 because Q(t0 ) differs from model to model, and
Q(t0 ) = Q ∗ only for the pure de Sitter case.
In particular, this equation with z = 0 gives the present age of the Universe
(t0 ). Thus a model with (say) m,0 = 0.3 and ,0 = 0.7 has an age of
t0 = 0.96H0−1 or, using (2.19), t0 = 9.5h −1 0 Gyr. Alternatively, in the limit
m,0 → 1, equation (2.70) gives back the standard result for the age of an EdS
Universe, t0 = 2/(3H0) = 6.5h −1 0 Gyr.
Putting (2.68) into the bolometric intensity integral (2.14), holding L(t) =
L 0 = constant as usual, and integrating over time, we obtain the plots of EBL
intensity Q(t)/Q ∗ shown in figure 2.8. This diagram shows that the rapidly-
brightening Universe is not a figment only of the pure de Sitter model (short-
dashed line). In a less vacuum-dominated model with m,0 = 0.3 and ,0 =
0.7, for instance, the ‘reading-room’ intensity of Q ∼ 1000 erg cm−2 s−1 is still
attained after t ∼ 290 Gyr.
In theory then, it might be thought that our remote descendants could live
under skies perpetually ablaze with the light of distant galaxies, in which the
rising and setting of their home suns would barely be noticed. That this cannot
happen in practice, as we have said, is due to the fact that galaxy luminosities
necessarily change with time as their brightest stars burn out. The finite lifetime
of the galaxies, in other words, is critical, not only in the sense of a finite
past, but a finite future. A proper assessment of this requires that we move
from considerations of background cosmology to the astrophysics of the sources

themselves. Real galaxies emit light with characteristic spectra which depend
strongly on wavelength, and evolve in time as well. In the next chapter, we will
adapt our bolometric formalism to take such features into account.
References
[1] Wesson P S 1999 Space–Time–Matter (Singapore: World Scientific)
[2] McVittie G C 1965 General Relativity and Cosmology (Urbana, IL: University of
Illinois Press) p 164
[3] Weinberg S 1972 Gravitation and Cosmology (New York: Wiley)
[5] Fukugita M, Hogan C J and Peebles P J E 1998 Astrophys. J. 503 518
[6] Yasuda N et al 2001 Astron. J. 122 1104
[7] Norberg P et al 2002 Preprint astro-ph/0111011
[8] Felten J E 1966 Astrophys. J. 144 241
[9] Dicke R H 1970 Gravitation and the Universe (Philadelphia, PA: American
Philosophical Society) p 62
[10] Ellis G F R 1988 Class. Quantum Grav. 5 891
[11] White M and Scott D 1996 Astrophys. J. 459 415
[12] Alexander D R et al 1997 Astron. Astrophys. 317 90
[13] Blome H-J, Hoell J and Priester W 1997 Bergmann-Schaefer: Lehrbuch der
Experimentalphysik vol 8 (Berlin: de Gruyter) pp 311–427
Chapter 3
The modern resolution and spectra
3.1 The spectral background

The previous chapter has convinced us that the bolometric intensity (or luminosity
per unit area at all wavelengths) of the light reaching us is low mostly because the
galaxies have not had enough time to fill up the Universe with light. Cosmic
expansion does reduce this further, but only by a factor of about one-half.
However, galaxies do not emit their light equally in all parts of the spectrum,
and it could be that expansion plays a relatively larger role at some wavelengths
than others.
This is perhaps best appreciated in the microwave region (at wavelengths
from about 1 mm to 10 cm) where we know from decades of radio astronomy
that the ‘night sky’ is brighter than its optical counterpart. The majority of this
microwave background radiation is thought to come, not from the redshifted light
of distant galaxies, but from the fading glow of the big bang itself—the ‘ashes
and smoke’ of creation in Lemaı̂tre’s words. Since its nature and suspected origin
are different from those of the EBL, this part of the extragalactic background has
its own name, the cosmic microwave background (CMB). Here expansion is of
paramount importance, as emphasized by Peacock [1], since the source radiation
in this case was emitted at more or less a single instant in cosmological history
(so that the ‘lifetime of the sources’ is negligible). Another way to see this is to
take expansion out of the picture, as we did in chapter 2: the CMB intensity we
would observe in this ‘equivalent static model’ would be that of the primordial
fireball and would roast us alive.
While Olbers’ paradox involves the EBL, not the CMB, this example is
still instructive because it prompts us to consider whether similar (though less
pronounced) effects could have been operative in the EBL as well. If, for instance,
galaxies emitted most of their light in a relatively brief burst of star formation
at very early times, this would be a galactic approximation to the picture just
described, and could conceivably boost the importance of expansion relative
to lifetime, at least in some wavebands. To check on this, we need a way to
41
42 The modern resolution and spectra
calculate EBL intensity as a function of wavelength. This is motivated by other

considerations as well. Olbers’ paradox has historically been concerned primarily
with the optical waveband (from approximately 4000 to 8000 Å), and this is still
what most people mean when they refer to the ‘brightness of the night sky’. And
from a practical standpoint, we would like to compare our theoretical predictions
with observational data, and these are necessarily taken using detectors which are
optimized for finite portions of the electromagnetic spectrum. Our task in this
chapter, then, is to convert the bolometric formalism of the previous chapter into
a spectral one.
3.2 From bolometric to spectral intensity

Rather than considering the entire galaxy luminosity L(t), we focus on the energy
emitted per unit time between wavelengths λ and λ + dλ. Let us write this in
the form dL λ ≡ F(λ, t) dλ, where F(λ, t) is the spectral energy distribution
(SED) of the galaxy, with dimensions of energy per unit time per unit wavelength.
Luminosity is recovered by integrating the SED over all wavelengths:
∞ ∞
L(t) = dL λ = F(λ, t) dλ. (3.1)
0 0
We then return to (2.12), the bolometric intensity of the spherical shell of galaxies
depicted in figure 2.1. Replacing L(t) with dL λ in this equation gives the intensity
of light emitted between λ and λ + dλ:
dQ λ,e = cn(t) R̃(t)[F(λ, t) dλ] dt. (3.2)
This light reaches us at the redshifted wavelength λ0 = λ/ R̃(t). Redshift also

stretches the wavelength interval by the same factor, dλ0 = dλ/ R̃(t). So the
intensity of light observed by us between λ0 and λ0 + dλ0 is
dQ λ,0 = cn(t) R̃ 2 (t)F[ R̃(t)λ0 , t] dλ0 dt. (3.3)
The intensity of the shell per unit wavelength, as observed at wavelength λ0 , is

then given simply by
dQ λ,0
4π dIλ (λ0 ) ≡ = cn(t) R̃ 2 (t)F[ R̃(t)λ0 , t] dλ0 dt (3.4)
dλ0
where the factor 4π converts from an all-sky intensity to one measured per
steradian. (This is merely a convention, but has become standard.) Integrating
over all the spherical shells corresponding to cosmic times tf and t0 (as before)
we obtain the spectral analogue of our earlier bolometric result, equation (2.14).
It reads: t0
c
Iλ (λ0 ) = n(t)F[ R̃(t)λ0 , t] R̃ 2 (t) dt. (3.5)
4π tf
From bolometric to spectral intensity 43
This is the integrated light from many galaxies, which has been emitted at various
wavelengths and redshifted by various amounts, but which is all in the waveband
centred on λ0 when it arrives at us. We refer to this as the spectral intensity of
the EBL at λ0 . Equation (3.5), or ones like it, have been considered from the
theoretical side principally by McVittie and Wyatt [2], Whitrow and Yallop [3, 4]
and Wesson et al [5, 6].
We may convert this integral over time t to one over redshift z using (2.17)
as before. This gives
zf
c n(z)F[λ0 /(1 + z), z] dz
Iλ (λ0 ) = (3.6)
4π H0 0 (1 + z)3 H̃ (z)
which is the spectral analogue of (2.20). It may be checked using (3.1) that
bolometric intensity is just the integral of spectral intensity over all observed
∞
wavelengths, Q = 0 I (λ0 ) dλ0 . In subsequent chapters we will apply
equations (3.5) and (3.6) to calculate the intensity of the EBL, not only from
galaxies but from many other sources of radiation as well.
The static analogue (i.e. the equivalent spectral EBL intensity in a Universe
without expansion, but with the properties of galaxies unchanged) is also readily
obtained in exactly the same way as before (section 2.6). Setting R̃(t) = 1 in
(3.5), we obtain t0
c
Iλ,stat (λ0 ) = n(t)F(λ0 , t) dt. (3.7)
4π tf
Just as in the bolometric case, we may convert this to an integral over z if we
choose. The latter parameter no longer represents physical redshift (since this has
been eliminated by hypothesis), but is now merely an algebraic way of expressing
the age of the galaxies. This is convenient because it puts (3.7) into a form which
may be directly compared with its counterpart (3.6) in the expanding Universe:
zf
c n(z)F(λ0 , z) dz
Iλ,stat (λ0 ) = . (3.8)
4π H0 0 (1 + z) H̃ (z)
If the same values are adopted for H0 and z f , and the same functional forms
are used for n(z), F(λ, z) and H̃ (z), then equations (3.6) and (3.8) allow us
to compare model universes which are alike in every way, except that one is
expanding while the other stands still.
Some simplification of these expressions is obtained as before in situations
where the comoving source number density can be taken as constant, n(z) = n 0 .
However, it is not possible to go farther and pull all the dimensional content out
of these integrals, as was done in the bolometric case, until a specific form is
postulated for the source SED F(λ, z).
In the remainder of chapter 3 we take up this problem, modelling galaxy
spectra with several different SEDs and using (3.6) to calculate the resulting
optical EBL intensity. Our goals in this exercise are threefold. First, we wish
to build up experience with the simpler kinds of SEDs. These will provide a
check of our main results in this chapter, and allow us to model a variety of less
conventional radiating sources in later chapters. Second, we would like to get
some idea of upper and lower limits on the spectral intensity of the EBL itself, and
compare them to the observational data. Third, we return to our original question
and divide Iλ (λ0 ) by its static analogue Iλ,stat (λ0 ), as given by (3.8), in order to
obtain a quantitative estimate of the importance of expansion in the spectral EBL
and the resolution of Olbers’ paradox.
3.3 The delta-function spectrum

The simplest possible source spectrum is one in which all the energy is emitted at
a single peak wavelength λp at each redshift z, thus

λ
F(λ, z) = Fp (z)δ −1 . (3.9)
λp
The function Fp (z) is obtained in terms of the total source luminosity L(z) by
normalizing over all observed wavelengths
∞
L(z) ≡ F(λ, z) dλ = Fp (z)λp (3.10)
0
so that Fp (z) = L(z)/λp . This SED is well suited to sources of electromagnetic

radiation such as elementary particle decays, which are characterized by specific
decay energies and may occur in the dark-matter halos surrounding galaxies. The
use of a δ-function is not a very good approximation for the spectra of galaxies,
but we will apply it here in this context to lay the foundation for later sections.
Since galaxies shine by starlight, a logical choice for the characteristic
wavelength λp would be the peak wavelength of a blackbody of ‘typical’ stellar
temperature. Taking the Sun as typical (T = T = 5770 K), this would be
λp = (0.290 cm K)/T = 5020 Å from Wiens’ law. Distant galaxies are seen
chiefly during periods of intense starburst activity when many stars are much
hotter than the Sun, and this would suggest a shift toward shorter wavelengths.
But any such effect must be largely offset by the fact that most of the short-
wavelength light produced in large starbursting galaxies (as much as 99% in the
most massive cases) is absorbed within these galaxies by dust and reradiated in
the infrared and microwave regions (λ 10 000 Å). It is also important to keep in
mind that while distant starburst galaxies are hotter and more luminous than local
spirals and ellipticals, the latter are likely to dominate the observed spectrum of
the EBL by virtue of their numbers, especially at low redshift. The best that one
can do with a single characteristic wavelength is to locate it somewhere within
the B-band (3600–5500 Å). For the purposes of this exercise we associate λp with
the nominal centre of this band, λp = 4400 Å, corresponding to a blackbody
temperature of 6590 K.
The delta-function spectrum 45
Strong evolution
9.5 Moderate evolution
Weak evolution
S97
log L ( h0 Lsun Mpc-3 )
F98
8.5
0 1 2 3 4
z
Figure 3.1. The comoving luminosity density of the Universe, as measured at z = 0

(F98 [8]) and extrapolated to higher redshifts using galaxies in the Hubble Deep Field
(S97 [9]). The unbroken curve is a least-squares fit to the data, while the dashed lines
represent upper and lower limits.
Substituting the galaxy SED (3.9) into the spectral intensity integral (3.6)
leads to
zf
c (z) λ0
Iλ (λ0 ) = δ − 1 dz (3.11)
4π H0λp 0 (1 + z)3 H̃ (z) λp (1 + z)
where we have introduced a new shorthand for the comoving luminosity density
of sources:
(z) ≡ n(z)L(z). (3.12)
For galaxies at redshift z = 0 this takes the value 0 as given by (2.24). Numerous
studies have shown that the product of n(z) and L(z) is approximately conserved
with redshift, even when the two quantities themselves appear to be evolving
markedly. So it would be reasonable to take (z) = 0 = constant. However, the
latest analyses have been able to benefit from new observational work at deeper
redshifts, and a new consensus is emerging that (z) does rise slowly but steadily
with z, peaking in the range 2 º z º 3, and falling away sharply thereafter [7].
This would be consistent with a picture in which the first generation of massive
galaxy formation occurred near z ∼ 3, being followed at lower redshifts by
galaxies whose evolution proceeded more passively.
Figure 3.1 shows the value of 0 from (2.24) at z = 0 (as taken from
Fukugita et al [8]), together with the extrapolation of (z) to five higher redshifts
(z = 0.35, 0.75, 1.5, 2.5 and 3.5) as inferred from analysis of photometric
galaxy redshifts in the Hubble Deep Field (HDF) by Sawicki et al [9]. (Most
of the galaxies in this extremely deep sample are too faint for their Doppler
or spectroscopic redshifts to be identified.) Let us define a relative comoving
luminosity density ˜ (z) by
˜(z) ≡ (z)/0 (3.13)
and fit this to the data with a cubic [log ˜ (z) = αz + βz 2 + γ z 3 ]. The resulting
least-squares best fit is shown in figure 3.1 together with plausible upper and lower
limits. We will refer to these cases as the ‘moderate’, ‘strong’ and ‘weak’ galaxy
evolution scenarios respectively.
Inserting (3.13) into (3.11) makes the latter read:

zf ˜(z) λ0
Iλ (λ0 ) = Iδ δ − 1 dz. (3.14)
0 (1 + z)3 H̃ (z) λp (1 + z)
The dimensional content of this integral has been concentrated into a prefactor Iδ ,
defined by
−1
c0 −1 λp
Iδ = = 4.4 × 10−9 erg s−1 cm−2 Å ster−1 . (3.15)
4π H0λp 4400 Å
This constant shares two important properties of its bolometric counterpart Q ∗
(section 2.3). First, it is explicitly independent of the uncertainty h 0 in Hubble’s
constant. Secondly, it is low by terrestrial standards. For example, it is well
below the intensity of the faint glow known as zodiacal light, which is caused
by the scattering of sunlight by dust in the plane of the Solar System. This is
important, since the value of Iδ sets the scale of the integral (3.14). Indeed,
existing observational bounds on Iλ (λ0 ) at λ0 ≈ 4400 Å are of the same
order as Iδ . Toller, for example, set an upper limit of Iλ (4400 Å) < 4.5 ×
−1
10−9 erg s−1 cm−2 Å ster−1 using data from the Pioneer 10 photopolarimeter
[10].
Dividing Iδ of (3.15) by the photon energy E 0 = hc/λ0 (where hc =
1.986 × 10−8 erg Å) puts the EBL intensity integral (3.14) into new units,
sometimes referred to as continuum units (CUs):

0 λ0 λ0
Iδ = Iδ (λ0 ) = = 970 CUs (3.16)
4πh H0 λp λp
where 1 CU ≡ 1 photon s−1 cm−2 Å−1 ster−1 . While both kinds of units
−1
(CUs and erg s−1 cm−2 Å ster−1 ) are in common use for reporting spectral
intensity at near-optical wavelengths, CUs appear most frequently. They are also
preferable from a theoretical point of view, because they most faithfully reflect
the energy content of a spectrum (as emphasized by Henry [11] ). A third type
Table 3.1. Cosmological test models

EdS/SCDM OCDM CDM BDM
m,0 1 0.3 0.3 0.03
,0 0 0 0.7 1
k 0 −1 0 +1
of intensity unit, the S10 (loosely, the equivalent of one tenth-magnitude star per
square degree) is also occasionally encountered, but will be avoided in this book
as it is wavelength-dependent and involves other subtleties which differ between
workers.
If we let the redshift of formation z f → ∞ for simplicity, then
equation (3.14) reduces to
 −2
 λ0 ˜(λ0 /λp − 1)
Iδ (if λ0 λp )
Iλ (λ0 ) = λp H̃ (λ0 /λp − 1) (3.17)

0 (if λ0 < λp ).
The comoving luminosity density ˜ (λ0 /λp − 1) which appears here is given
by the fit (3.13) to the HDF data. The Hubble parameter is given by (2.40) as
H̃ (λ0 /λp − 1) = [m,0(λ0 /λp )3 + ,0 − (m,0 + ,0 − 1)(λ0 /λp )2 ]1/2 for a
Universe containing dustlike matter and vacuum energy with density parameters
m,0 and ,0 respectively.
‘Turning off’ the luminosity density evolution (so that ˜ = 1 = constant),
one can obtain three trivial special cases with

 (λ0 /λp )−7/2 (Einstein–de Sitter)
Iλ (λ0 ) = Iδ × (λ0 /λp )−2 (de Sitter) (3.18)

(λ0 /λp )−3 (Milne).
This is taken at λ0 λp , where (m,0 , ,0 ) = (1, 0), (0, 1) and (0, 0)
respectively for the three models cited (section 2.7). The first of these is
the ‘ 72 -law’ which appears frequently in the particle-physics literature as an
approximation to the spectrum of EBL contributions from decaying particles. But
the second (de Sitter) probably provides a better approximation, given current
thinking regarding the values of m,0 and ,0 .
To evaluate the spectral EBL intensity (3.14) and other quantities in a general
situation, it will be helpful to define a suite of cosmological test models which
span the widest range possible in the parameter space defined by m,0 and ,0
(table 3.1). Detailed discussion of these is left to chapter 4, but we can summarize
the main rationale for each briefly as follows. The Einstein–de Sitter (EdS)
model has long been favoured on grounds of simplicity and was alternatively
known for some time as the ‘standard cold dark matter’ or SCDM model. It
has come under increasing pressure, however, as evidence mounts for levels
of m,0 º 0.5 and, most recently, from observations of Type Ia supernovae
(SNIa) which indicate that ,0 > m,0 . The open cold dark matter (OCDM)
model is more consistent with data on m,0 and holds appeal for those who have
been reluctant to accept the possibility of a non-zero vacuum energy. It faces
the considerable challenge, however, of explaining new data on the spectrum
of CMB fluctuations, which imply that m,0 + ,0 ≈ 1. The + cold dark
matter (CDM) model has rapidly become the new standard in cosmology
because it agrees best with both the SNIa and CMB observations. However, this
model suffers from a ‘coincidence problem’, in that m (t) and (t) evolve
so differently with time that the probability of finding ourselves at a moment
in cosmic history when they are even of the same order of magnitude appears
unrealistically small. This is addressed to some extent in the last model, where
we push m,0 and ,0 to their lowest and highest limits, respectively. In the case
of m,0 these limits are set by big-bang nucleosynthesis, which requires a density
of at least m,0 ≈ 0.03 in baryons (hence the + baryonic dark matter or
BDM model). Upper limits on ,0 come from various arguments, such as the
observed frequency of gravitational lenses and the requirement that the Universe
began in a big-bang singularity. Within the context of isotropic and homogeneous
cosmology, these four models cover the full range of what would be considered
plausible by most workers.
Figure 3.2 shows the solution of the full integral (3.14) for all four test
models. The short-wavelength cut-off in these plots is an artefact of the δ-
function SED, but the behaviour of Iλ (λ0 ) at wavelengths above λp = 4400 Å
is quite revealing, even in a model as simple as this one. In the EdS case (a),
the rapid fall-off in intensity with λ0 indicates that nearby (low-redshift) galaxies
dominate. There is a secondary hump at λ0 ≈ 10 000 Å, which is an ‘echo’ of the
peak in galaxy formation, redshifted into the near infrared. This hump becomes
successively enlarged relative to the optical peak at 4400 Å as the value of m,0
drops relative to ,0 . Eventually one has the situation in the de Sitter-like model
(d), where the galaxy-formation peak entirely dominates the observed EBL signal,
despite the fact that it comes from distant galaxies at z ≈ 3. This is because a
large ,0 -term (especially one which is large relative to m,0 ) inflates comoving
volume at high redshifts. Since the comoving number density of galaxies is fixed
by the fit to ˜ (z) (figure 3.1), the number of galaxies at these redshifts has no
choice but to go up, pushing up the infrared part of the spectrum. Although the
δ-function spectrum is an unrealistic one, we will see in subsequent sections that
this trend persists in more sophisticated models, providing a clear link between
observations of the EBL and the cosmological parameters m,0 and ,0 .
Figure 3.2 is plotted over a broad range of wavelengths from the near
ultraviolet (NUV; 2000–4000 Å) to the near infrared (NIR; 8000–40 000 Å). We
have included a number of reported observational constraints on EBL intensity.
These require a bit of comment as regards their origin. Upper limits (full symbols
(a) (b)
Strong evolution Strong evolution

Moderate evolution Moderate evolution
Weak evolution Weak evolution
LW76 LW76
SS78 SS78
D79 D79
Iλ ( CUs )
Iλ ( CUs )
1000 T83 1000 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
(c) (d)

LW76 LW76
SS78 SS78
D79 D79
Iλ ( CUs )
Iλ ( CUs )
1000 T83 1000 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
Figure 3.2. The spectral EBL intensity of galaxies whose radiation is modelled by
δ-functions at a rest frame wavelength of 4400 Å, calculated for four different cosmological
models: (a) EdS, (b) OCDM, (c) CDM and (d) BDM (table 3.1). Also shown are
observational upper limits (full symbols and bold lines) and reported detections (empty
symbols) over the waveband 2000–40 000 Å.
and bold lines) have come from analyses of OAO-2 satellite data (LW76 [12]),
ground-based telescopes (SS78 [13], D79 [14], BK86 [15]), the aforementioned
Pioneer 10 instrument (T83 [10]), sounding rockets (J84 [16], T88 [17]), the
Hopkins UVX experiment aboard the Space Shuttle (M90 [18]) and—most
recently, in the near infrared—the DIRBE instrument on the COBE satellite
(H98 [19]). The past two or three years have also seen the first widely-accepted
detections of the EBL (figure 3.2, open symbols). In the NIR these have come
from continued analysis of DIRBE data in the K-band (22 000 Å) and L-band
(35 000 Å; WR00 [20]), as well as the J-band (12 500 Å; C01 [21]). Reported
detections in the optical using a combination of Hubble Space Telescope (HST)
and Las Campanas telescope observations (B98 [22, 23]) are still preliminary but
very important, and we have included them as well.
Figure 3.2 shows that EBL intensities based on the simple δ-function
spectrum are in rough agreement with the data. Predicted intensities come in at
or just below the optical limits in the low-,0 cases (a, b) and remain consistent
with the infrared limits even in the high-,0 cases (c, d). Vacuum-dominated
models with even higher ratios of ,0 to m,0 would, however, run afoul of
DIRBE limits in the J-band. We will find a similar trend in models with more
realistic galaxy spectra.
3.4 Gaussian spectra

The Gaussian distribution provides a useful generalization of the δ-function
for modelling sources whose spectra, while essentially monochromatic, are
broadened by some physical process. For example, photons emitted by the
decay of elementary particles inside dark-matter halos would have their energies
Doppler-broadened by the circular velocity vc ≈ 220 km s−1 of their host
galaxies, giving rise to a spread σλ of order σλ (λ) = (2vc /c)λ ≈ 0.0015λ in the
SED. In the context of galaxies, this extra degree of freedom provides a simple
way to model the width of the bright part of the spectrum. If we take this to cover
the B-band (3600–5500 Å) then σλ ∼ 1000 Å. The Gaussian SED reads:

L(z) 1 λ − λp 2
F(λ, z) = √ exp − (3.19)
2πσλ 2 σλ
where λp is the wavelength at which the galaxy emits most of its light. We take
λ
p∞= 4400 Å as before, and note that integration over λ0 confirms that L(z) =
0 F(λ, z) dλ as required. Once again we can make the simplifying assumption
that L(z) = L 0 = constant; or we can use the empirical fit ˜ (z) ≡ n(z)L(z)/0
to the HDF data of Sawicki et al [9]. Taking the latter course and substituting
(3.19) into (3.6), we obtain
zf
˜(z) 1 λ0 /(1 + z) − λp 2
Iλ (λ0 ) = Ig exp − dz. (3.20)
0 (1 + z)3 H̃ (z) 2 σλ
The dimensional content of this integral has been pulled into a prefactor Ig =
Ig (λ0 ), defined by

0 λ0 λ0
Ig = √ = 390 CUs . (3.21)
32π 3 h H0 σλ σλ
Here we have divided (3.20) by the photon energy E 0 = hc/λ0 to put the result
into CUs, as before.
Results are shown in figure 3.3, where we have taken λp = 4400 Å,
σλ = 1000 Å and z f = 6. Aside from the fact that the short-wavelength
cut-off has disappeared, the situation is qualitatively similar to that obtained
using a δ-function approximation. (This similarity becomes formally exact as σλ
approaches zero.) One sees, as before, that the expected EBL signal is brightest at
optical wavelengths in an EdS Universe (a), but that the infrared hump due to the
redshifted peak of galaxy formation begins to dominate for higher-,0 models
Blackbody spectra 51
(a) (b)

LW76 LW76
SS78 SS78
1000 D79 1000 D79
Iλ ( CUs )
Iλ ( CUs )
T83 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
(c) (d)

LW76 LW76
SS78 SS78
1000 D79 1000 D79
Iλ ( CUs )
Iλ ( CUs )
T83 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
Figure 3.3. The spectral EBL intensity of galaxies whose spectra has been represented
by Gaussian distributions with rest-frame peak wavelength 4400 Å and standard deviation
1000 Å, calculated for the (a) EdS, (b) OCDM, (c) CDM and (d) BDM cosmologies
and compared with observational upper limits (full symbols and bold lines) and reported
detections (empty symbols).
(b) and (c), becoming overwhelming in the de Sitter-like model (d). Overall, the
best agreement between calculated and observed EBL levels occurs in the CDM
model (c). The matter-dominated EdS (a) and OCDM (b) models appear to
contain too little light (requiring one to postulate an additional source of optical or
near-optical background radiation besides that from galaxies), while the BDM
model (d) comes uncomfortably close to containing too much light. This is an
interesting situation and one which motivates us to reconsider the problem with
more realistic models for the galaxy SED.
3.5 Blackbody spectra
The simplest non-trivial approach to a galaxy spectrum is to model it as a

blackbody, and this was done by previous workers such as McVittie and Wyatt [2],
Whitrow and Yallop [3, 4] and Wesson [6]. Let us suppose that the galaxy SED
is a product of the Planck function and some wavelength-independent parameter
C(z):
2πhc2 C(z)/λ5
F(λ, z) = . (3.22)
σSB exp[hc/kT (z)λ] − 1
Here σSB ≡ 2π 5 k 4 /15c2h 3 = 5.67 × 10−5 erg cm−2 s−1 K−1 is the Stefan–
Boltzmann constant. The function F is normally regarded as an increasing
function of redshift (at least out to the redshift of galaxy formation). This can, in
principle, be accommodated by allowing C(z) or T (z) to increase with z in (3.22).
The former choice would correspond to a situation in which galaxy luminosity
decreases with time while its spectrum remains unchanged, as might happen if
stars were simply to die. The second choice corresponds to a situation in which
galaxy luminosity decreases with time as its spectrum becomes redder, as may
happen when its stellar population ages. The latter scenario is more realistic and
will be adopted here. The luminosity L(z) is found by integrating F(λ, z) over
all wavelengths:
∞
2πhc2 λ−5 dλ
L(z) = C(z) = C(z)[T (z)]4 (3.23)
σSB 0 exp[hc/kT (z)λ] − 1
so that the unknown function C(z) must satisfy C(z) = L(z)/[T (z)]4 . If we
require that Stefan’s law (L ∝ T 4 ) hold at each z, then
C(z) = constant = L 0 /T04 (3.24)
where T0 is the present ‘galaxy temperature’ (i.e. the blackbody temperature

corresponding to a peak wavelength in the B-band). Thus the evolution of
galaxy luminosity in this model is just that which is required by Stefan’s law
for blackbodies whose temperatures evolve as T (z). This is reasonable, since
galaxies are made up of stellar populations which cool and redden with time as
hot massive stars die out.
Let us supplement this with the minimal assumption that galaxy comoving
number density is conserved with redshift, n(z) = n 0 = constant. This is
sometimes referred to as the pure luminosity evolution or PLE scenario; and
while there is some controversy on this point, PLE has been found by many
workers to be roughly consistent with observed numbers of galaxies at faint
magnitudes, especially if there is a significant vacuum energy density ,0 > 0.
Proceeding on this assumption, the comoving galaxy luminosity density can be
written
n(z)L(z) L(z) T (z) 4
(z) ≡
˜ = = . (3.25)
0 L0 T0
This expression can then be inverted to give the blackbody temperature T (z) as a
function of redshift, since the form of ˜(z) is fixed by our fit to the photometric
HDF data (figure 3.1):
T (z) = T0 [˜ (z)]1/4. (3.26)
Blackbody spectra 53
We can check this by choosing T0 = 6600 K (i.e. a present peak wavelength

of 4400 Å) and reading off values of ˜(z) = (z)/0 at the peaks of the
curves marked ‘weak’, ‘moderate’ and ‘strong’ evolution in figure 3.1. Putting
these numbers into (3.26) yields blackbody temperatures (and corresponding peak
wavelengths) of 10 000 K (2900 Å), 11 900 K (2440 Å) and 13 100 K (2210 Å)
respectively at the galaxy-formation peak. These results are consistent with the
idea that galaxies would have been dominated by hot UV-emitting stars at this
early time.
Inserting the expressions (3.24) for C(z) and (3.26) for T (z) into the SED
(3.22), and substituting the latter into the EBL integral (3.6), we obtain
zf
(1 + z)2 dz
Iλ (λ0 ) = Ib . (3.27)
0 {exp[hc(1 + z)/kT (z)λ0 ] − 1} H̃(z)
The dimensional prefactor Ib = Ib (T0 , λ0 ) reads in this case

−4 −4
c2 0 T0 λ0
Ib = = 90 100 CUs . (3.28)
2H0σSB T04 λ40 6600 K 4400 Å
Numerically, the argument of the exponential term may be expressed in the form
hc(1 + z)/kT (z)λ0 = 4.95(1 + z)/[T (z)/6600 K]/(λ0 /4400 Å).
Results are shown in figure 3.4, where we have set z f = 6 following recent
observational hints of an epoch of ‘first light’ at this redshift [24]. Overall EBL
intensity is, however, quite insensitive to this choice, provided that z f 3.
Between z f = 3 and z f = 6, Iλ (λ0 ) rises by less than 1% below λ0 = 10 000 Å
and less than ∼5% at λ0 = 20 000 Å (where most of the signal originates at high
redshifts). There is no further increase beyond z f > 6 at the three-figure level of
precision.
Figure 3.4 shows some qualitative differences from our earlier results
obtained using the δ-function and Gaussian SEDs. Most noticeably, the prominent
‘double-hump’ structure is no longer apparent. The key evolutionary parameter
is now blackbody temperature T (z) and this goes as [(z)]1/4 so that individual
features in the comoving luminosity density profile are suppressed. (A similar
effect can be achieved with the Gaussian SED by choosing larger values of σλ .)
One can make out the same general trend as before by focusing on the long-
wavelength tail in each panel. This climbs steadily up the right-hand side of the
figure as one moves from the ,0 = 0 models (a) and (b) to the ,0 -dominated
models (c) and (d), showing that more and more light is arriving from distant,
highly redshifted galaxies and boosting the NIR part of the spectrum.
Absolute EBL intensities in each of these four models are consistent with
what we have seen already. This is not surprising because changing the shape of
the SED merely shifts light from one part of the spectrum to another. It cannot
alter the total amount of light in the EBL, which is set by the comoving luminosity
density ˜ (z) of sources, once the background cosmology (and hence the source
lifetime) has been chosen. As before, the best match between calculated EBL
(a) (b)

LW76 LW76
1000 SS78 1000 SS78
D79 D79
Iλ ( CUs )
Iλ ( CUs )
T83 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
(c) (d)

LW76 LW76
1000 SS78 1000 SS78
D79 D79
Iλ ( CUs )
Iλ ( CUs )
T83 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
Figure 3.4. The spectral EBL intensity of galaxies, modelled as blackbodies whose
characteristic temperatures are such that their luminosities L ∝ T 4 combine to produce
the observed comoving luminosity density (z) of the Universe. Results are shown
for the (a) EdS, (b) OCDM, (c) CDM and (d) BDM cosmologies. Also shown are
observational upper limits (full symbols and bold lines) and reported detections (open
symbols).
intensities and the observational detections is found for the ,0 -dominated
models (c) and (d). The fact that the EBL is now spread across a broader spectrum
has pulled down its peak intensity slightly, so that the BDM model (d) no longer
threatens to violate observational limits and, in fact, fits them rather nicely. The
zero-,0 models (a) and (b) would again appear to require some additional
source of background radiation (beyond that produced by galaxies) if they are
to contain enough light to make up the levels of EBL intensity that have been
reported.
3.6 Normal and starburst galaxies

The previous sections have shown that simple models of galaxy spectra, combined
with data on the evolution of comoving luminosity density in the Universe, can
produce levels of spectral EBL intensity in rough agreement with observational
limits and reported detections, and even discriminate to a degree between different
cosmological models. However, the results obtained up to this point are somewhat
Normal and starburst galaxies 55
unsatisfactory in that they are sensitive to theoretical input parameters, such as

λp and T0 , which are hard to connect with the properties of the actual galaxy
population.
A more comprehensive approach would use observational data in
conjunction with theoretical models of galaxy evolution to build up an ensemble
of evolving galaxy SEDs F(λ, z) and comoving number densities n(z) which
would depend not only on redshift but on galaxy type as well. Increasingly
sophisticated work has been carried out along these lines over the years by
Partridge and Peebles [25], Tinsley [26], Bruzual [27], Code and Welch [28],
Yoshii and Takahara [29] and others. The last-named authors, for instance,
divided galaxies into five morphological types (E/SO, Sab, Sbc, Scd and Sdm),
with a different evolving SED for each type, and found that their collective EBL
intensity at NIR wavelengths was about an order of magnitude below the levels
suggested by observation.
Models of this kind, however, are complicated while at the same time
containing uncertainties. This makes their use somewhat incompatible with our
purpose here, which is primarily to obtain a first-order estimate of EBL intensity
so that the importance of expansion can be properly ascertained. Also, the
increasingly deeper observations of recent years have begun to make it clear
that these morphological classifications are of limited value at redshifts z 1,
where spirals and ellipticals are still in the process of forming [30]. As we have
already seen, this is precisely where much of the EBL may originate, especially
if luminosity density evolution is strong or if there is a significant ,0 -term.
What is needed, then, is a simple model which does not distinguish too finely
between the spectra of galaxy types as they have traditionally been classified, but
which can capture the essence of broad trends in luminosity density evolution over
the full range of redshifts 0 z z f . For this purpose we will group together
the traditional classes (spiral, elliptical, etc) under the single heading of quiescent
or normal galaxies. At higher redshifts (z 1), we will allow a second class of
objects to play a role: the active or starburst galaxies. Whereas normal galaxies
tend to be comprised of older, redder stellar populations, starburst galaxies are
dominated by newly-forming stars whose energy output peaks in the ultraviolet
(although much of this is absorbed by dust grains and subsequently reradiated
in the infrared). One signature of the starburst type is thus a decrease in F(λ)
as a function of λ over NUV and optical wavelengths, while normal types show
an increase [31]. Starburst galaxies also tend to be brighter, reaching bolometric
luminosities as high as 1012–1013 L , versus 1010–1011 L for normal types.
There are two ways to obtain SEDs for these objects: by reconstruction
from observational data; or as output from theoretical models of galaxy evolution.
The former approach has had some success but becomes increasingly difficult at
short wavelengths, so that results have typically been restricted to λ 1000 Å
[31]. This represents a serious limitation if we want to integrate out to redshifts
z f ∼ 6 (say), since it means that our results are only strictly reliable down to
λ0 = λ(1 + z f ) ∼ 7000 Å. In order to integrate out to z f ∼ 6 and still go
(a) (b)
No extinction No extinction
1e+07
With extinction 1e+09 With extinction
1e+08
Fn ( Lsun Å−1 )
Fs ( Lsun Å−1 )
1e+06
1e+07
100000
1e+06
10000 100000
100 1000 10000 100000 1e+06 100 1000 10000 100000 1e+06
λ(Å) λ(Å)
Figure 3.5. Typical galaxy SEDs for (a) normal and (b) starburst type galaxies with and
without extinction by dust. These figures are adapted from figures 9 and 10 of Devriendt et
al [32], where, however, the plotted quantity is log[λF(λ)], not F(λ) as shown here. For
definiteness we have normalized (over 100 − 3 × 104 Å) such that L n = 1 × 1010 h −2
0 L
and L s = 2 × 1011 h −20 L with h 0 = 0.75. (These values are consistent with what we
will later call ‘model 0’ for a comoving galaxy number density of n 0 = 0.010h 30 Mpc−3 .)
down as far as the NUV (λ0 ∼ 2000 Å), we require SEDs which are good to
λ ∼ 300 Å in the galaxy rest-frame. For this purpose we will make use of
theoretical galaxy-evolution models, which have advanced to the point where they
cover the entire spectrum from the far ultraviolet to radio wavelengths. This broad
range of wavelengths involves diverse physical processes such as star formation,
chemical evolution and (of special importance here) dust absorption of ultraviolet
light and re-emission in the infrared. Typical normal and starburst galaxy SEDs
based on such models are now available down to ∼100 Å [32]. These functions,
displayed in figure 3.5, will constitute our normal and starburst galaxy SEDs,
Fn (λ) and Fs (λ).
Figure 3.5 shows the expected increase in Fn (λ) with λ at NUV wavelengths
(∼2000 Å) for normal galaxies, as well as the corresponding decrease for
starbursts. What is most striking about both templates, however, is their overall
multi-peaked structure. These objects are far from pure blackbodies, and the
primary reason for this is dust. This effectively removes light from the shortest-
wavelength peaks (which are due mostly to star formation), and transfers it to the
longer-wavelength ones. The dotted lines in figure 3.5 show what the SEDs would
look like if this dust reprocessing were ignored. The main difference between
normal and starburst types lies in the relative importance of this process. Normal
galaxies emit as little as 30% of their bolometric intensity in the infrared, while
the equivalent fraction for the largest starburst galaxies can reach 99%. (The
latter are, in fact, usually detected because of their infrared emission, despite the
fact that they contain mostly hot, blue stars.) In the full model of Devriendt et
al [32], these variations are incorporated by modifying input parameters such as
star formation timescale and gas density, leading to spectra which are broadly
similar in shape to those in figure 3.5 (though they differ in normalization and
‘tilt’ toward longer wavelengths). The results, it should be emphasized, have
been successfully matched to a wide range of real galaxy spectra. Here we will
be content with the single pair of prototypical SEDs shown in figure 3.5.
We now proceed to calculate the EBL intensity using Fn (λ) and Fs (λ), with
the
characteristic luminosities
of these two types found as usual by normalization,
Fn (λ) dλ = L n and Fs (λ) dλ = L s . Let us assume that the comoving
luminosity density of the Universe at any redshift z is a combination of normal
and starburst components
(z) = n n (z)L n + n s (z)L s (3.29)
where the comoving number densities are
n n (z) ≡ [1 − f (z)]n(z) and n s (z) ≡ f (z)n(z). (3.30)
In other words, we will account for evolution in (z) solely in terms of the
changing starburst fraction f (z), and a single comoving number density n(z)
as before. L n and L s are awkward to work with for dimensional reasons, and
we will find it more convenient to specify the SED instead by two dimensionless
adjustable parameters, the local starburst fraction f 0 and luminosity ratio 0 :
f 0 ≡ f (0) 0 ≡ L s /L n . (3.31)
Observations indicate that f 0 ≈ 0.05 in the local population [31] and Devriendt
et al were able to fit their templates to a range of normal and starburst galaxies
with 40 º 0 º 890 [32]. We will allow these two parameters to vary in the
ranges 0.01 f0 0.1 and 10 0 1000. This, in combination with our
‘strong’ and ‘weak’ limits on comoving luminosity-density evolution, gives us the
flexibility to obtain upper and lower bounds on EBL intensity.
The functions n(z) and f (z) can now be fixed by equating (z) as defined
by (3.29) to the comoving luminosity-density curves inferred from HDF data
(figure 3.1), and requiring that f → 1 at peak luminosity (i.e. assuming
that the galaxy population is entirely starburst-dominated at the redshift z p of
peak luminosity). These conditions are not difficult to set up. One finds that
modest number-density evolution is required in general, if f (z) is not to over or
undershoot unity at z p . We follow [33] and parametrize this with the function
n(z) = n 0 (1 + z)η for z z p . Here η can be termed the merger parameter
since a value of η > 0 would imply that the comoving number density of galaxies
decreases with time.
Pulling these requirements together, one obtains a model with


f (z) =


1
− 1
[0 (1 + z)−η (z) − 1] (z zp)
 0
1 (z > z p )

(1 + z)η (z z p )
n(z) = n 0 ×
(z) (z > z p ) .
(3.32)

Here (z) ≡ [1/0 + (1 − 1/0 ) f 0 ] ˜ (z) and η = ln[ (z p )]/ ln(1 + z p ). The
evolution of f (z), n n (z) and n s (z) is plotted in figure 3.6 for five models: a best-
fit model 0, corresponding to the moderate evolution curve in figure 3.1 with
f 0 = 0.05 and 0 = 20, and four other models chosen to produce the widest
possible spread in EBL intensities across the optical band. Models 1 and 2 are
the most starburst-dominated, with initial starburst fraction and luminosity ratio
at their upper limits ( f 0 = 0.1 and 0 = 1000). Models 3 and 4 are the least
starburst-dominated, with the same quantities at their lower limits ( f 0 = 0.01 and
0 = 10). Luminosity density evolution is set to ‘weak’ in the odd-numbered
models 1 and 3, and ‘strong’ in the even-numbered models 2 and 4. (In principle
one could identify four other ‘extreme’ combinations, such as maximum f 0 with
minimum 0 , but these will be intermediate to models 1–4.) We find merger
parameters η between +0.4, 0.5 in the strong-evolution models 2 and 4, and
−0.5, −0.4 in the weak-evolution models 1 and 3, while η = 0 for model 0.
These are well within the usual range [34].
The information contained in figure 3.6 can be summarized in words as
follows: starburst galaxies formed near z f ∼ 4 and increased in comoving number
density until z p ∼ 2.5 (the redshift of peak comoving luminosity density in
figure 3.1). They then gave way to a steadily growing population of fainter normal
galaxies which began to dominate between 1 º z º 2 (depending on the model)
and now make up 90–99% of the total galaxy population at z = 0. This is a
reasonable scenario and agrees with others that have been constructed to explain
the observed faint blue excess in galaxy number counts [35].
We are now in a position to compute the total spectral EBL intensity, which
is the sum of normal and starburst components:
Iλ (λ0 ) = Iλn (λ0 ) + Iλs (λ0 ). (3.33)
These components are found as usual by substituting the SEDs (Fn , Fs ) and
comoving number densities (3.30) into (3.6). This leads to:

zf λ0 dz
Iλn (λ0 ) = Ins ñ(z)[1 − f (z)]Fn
0 1+z (1 + z)3 H̃ (z)
zf
λ0 dz
Iλ (λ0 ) = Ins
s
ñ(z) f (z)Fs . (3.34)
0 1 +z (1 + z)3 H̃ (z)
(a) (b)
Model 0 (f0 = 0.05, l0 = 20) (evolution)

Model 1 (f0 = 0.1, l0 = 1000) (weak evolution)
Model 2 (f0 = 0.1, l0 = 1000) (strong evolution)
0.03
Model 3 (f0 = 0.01, l0 = 10) (weak evolution)
Model 4 (f0 = 0.01, l0 = 10) (strong evolution)
n ( h30 Mpc-3 )
1
0.02
f
0.01
0 0
0 1 2 0 1 2 3 4
z z
Figure 3.6. Evolution of (a) starburst fraction f (z) and (b) comoving normal and
starburst galaxy number densities n n (z) and n s (z), where total comoving luminosity
density (z) = n n (z)L n + n s (z)L s is matched to the ‘moderate’, ‘weak’ and ‘strong’
evolution curves in figure 3.1. Model 0 lies midway between models 1–4, which have
maximum and minimum possible values of the two adjustable parameters f0 ≡ f (0) and
0 ≡ L s /L n .
Here ñ(z) ≡ n(z)/n 0 is the relative comoving number density and the
dimensional content of both integrals has been pulled into the prefactor

0 λ0 λ0
Ins = Ins (λ0 ) = = 970 CUs . (3.35)
4πh H0 Å Å
This is explicitly independent of h 0 , as before, and it is important to realize how
this comes about. There is an implicit factor of L 0 contained in the galaxy SEDs
Fn (λ) and Fs (λ) via their normalizations (i.e. via the fact that we require the
galaxy population to make up the observed comoving luminosity density of the
Universe). To see this, note that equation (3.29) reads 0 = n 0 L n [1 + (0 − 1) f 0 ]
at z = 0. Since 0 ≡ n 0 L 0 , it follows that L n = L 0 /[1 + (0 − 1) f 0 ] and
L s = L 0 0 /[1 +(0 −1) f 0 ]. If the factor of L 0 is divided out of these expressions
when normalizing the functions Fn and Fs , it can be put directly into the integrals
(3.34), forming the quantity 0 in (3.35) and cancelling out the factor of h 0 in H0
as required.
The spectral intensity (3.33) is plotted in figure 3.7, where we have set
z f = 6 as usual. (Results are insensitive to this choice, increasing by less than
5% as one moves from z f = 3 to z f = 6, with no further increase for z f 6
at three-figure precision.) These plots show that the most starburst-dominated
models (1 and 2) produce the bluest EBL spectra, as might be expected. For these
two models, EBL contributions from normal galaxies remain well below those
from starbursts at all wavelengths, so that the bump in the observed spectrum at
(a) (b)
Model 0 Model 0
Model 1 Model 1
Model 2 Model 2
Model 3 Model 3
1000 Model 4 1000 Model 4
LW76 LW76
SS78 SS78
Iλ ( CUs )
Iλ ( CUs )
D79 D79
T83 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
(c) (d)
Model 0 Model 0
Model 1 Model 1
Model 2 Model 2
Model 3 Model 3
1000 Model 4 1000 Model 4
LW76 LW76
SS78 SS78
Iλ ( CUs )
Iλ ( CUs )
D79 D79
T83 T83
J84 J84
BK86 BK86
T88 T88
M90 M90
B98 B98
H98 H98
WR00 WR00
C01 C01
100 100
1000 10000 1000 10000
λ0 ( Å ) λ0 ( Å )
Figure 3.7. The spectral EBL intensity of a combined population of normal and starburst
galaxies, with SEDs as shown in figure 3.5. The evolving number densities are such as to
reproduce the total comoving luminosity density seen in the HDF (figure 3.1). Results are
shown for the (a) EdS, (b) OCDM, (c) CDM and (d) BDM cosmologies. Also shown
are observational upper limits (full symbols and bold lines) and reported detections (open
symbols).
λ0 ∼ 4000 Å is essentially an echo of the peak at ∼ 1100 Å in the starburst SED

(figure 3.5), redshifted by a factor (1 + z p ) from the epoch z p ≈ 2.5 of maximum
comoving luminosity density. In the least starburst-dominated models (3 and 4),
by contrast, EBL contributions from normal galaxies catch up to and exceed those
from starbursts at λ0 10 000 Å, giving rise to the bump seen at λ0 ∼ 20 000 Å in
these models. Absolute EBL intensities are highest in the strong-evolution models
(2 and 4) and lowest in the weak-evolution models (1 and 3). We emphasize that,
for a given cosmology, the total amount of light in the EBL is set by a choice of
luminosity density profile. A choice of SED merely shifts this from one part of
the spectrum to another. Within the context of the simple two-component model
adopted here, and the constraints imposed on luminosity density by the HDF data
(section 3.3), the curves in figure 3.7 represent upper and lower limits on the
spectral intensity of the EBL at near-optical wavelengths.
These curves are spread over a broader range of wavelengths than those
obtained earlier using single-component Gaussian and blackbody spectra. This
leads to a drop in peak intensity, as we can appreciate by noting that there now
Back to Olbers 61
appears to be a significant gap between theory and observation in all but the most
vacuum-dominated cosmology, BDM (d). This is so even for the models with
the strongest luminosity density evolution (models 2 and 4). In the case of the
EdS cosmology (a), this gap is nearly an order of magnitude, as found by Yoshii
and Takahara [29]. Similar conclusions have been reached more recently from
an analysis of Subaru Deep Field data by Totani et al [36], who suggest that the
shortfall could be made up by a very diffuse, previously undetected component of
background radiation not associated with galaxies. However, other workers have
argued that existing galaxy populations are enough to explain the data, if different
assumptions are made about their SEDs [37] or if allowance is made for faint low-
surface-brightness galaxies below the detection limit of existing surveys [23].
In this connection it may be worthwhile to remember that there are preliminary
indications from the Sloan Digital Sky Survey that the traditional value of 0
(as we have used it here) may go up by as much as 40% when corrected for a
downward bias which is inherent in the way that galaxy magnitudes have been
measured at high redshift [38].
3.7 Back to Olbers

Having obtained quantitative estimates of the spectral intensity of the near-optical
EBL which are in reasonable agreement with observation, let us return to the
question with which we began: why precisely is the sky dark at night? By ‘dark’
we now mean specifically dark at near-optical wavelengths. We can provide a
quantitative answer to this question by using a spectral version of the bolometric
argument in chapter 2. That is, we compute the EBL intensity Iλ,stat in model
universes which are equivalent to expanding ones in every way except expansion,
and then take the ratio Iλ /Iλ,stat . If this is of order unity, then expansion plays a
minor role and the darkness of the optical sky (like the bolometric one) must be
attributed mainly to the fact that the Universe is too young to have filled up with
light. If Iλ /Iλ,stat 1, however, then we would have a situation qualitatively
different from the bolometric one, and expansion would play a crucial role in the
resolution to Olbers’ paradox.
The spectral EBL intensity for the equivalent static model is obtained by
putting the functions ñ(z), f (z), Fn (λ), Fs (λ) and H̃ (z) into (3.8) rather than
(3.6). This results in
n s
Iλ,stat (λ0 ) = Iλ, stat (λ0 ) + Iλ,stat (λ0 ) (3.36)
where the normal and starburst contributions are given by

zf
ñ(z)[1 − f (z)] dz
Iλ,stat (λ0 ) = Ins Fn (λ0 )
n
0 (1 + z) H̃ (z)
zf
ñ(z) f (z) dz
stat (λ0 ) = Ins Fs (λ0 ) .
s
Iλ, (3.37)
0 (1 + z) H̃ (z)
(a) (b)
Model 0 Model 0
1 Model 1 1 Model 1
Model 2 Model 2
Model 3 Model 3
Iλ / Iλ,stat
Iλ / Iλ,stat
Model 4 Model 4
0.5 0.5
0 0
10000 10000
λ0 ( Å ) λ0 ( Å )
(c) (d)
Model 0 Model 0
1 Model 1 1 Model 1
Model 2 Model 2
Model 3 Model 3
Iλ / Iλ,stat
Iλ / Iλ,stat
Model 4 Model 4
0.5 0.5
0 0
10000 10000
λ0 ( Å ) λ0 ( Å )
Figure 3.8. The ratio Iλ /Iλ,stat of spectral EBL intensity in expanding models to that in
equivalent static models, for the (a) EdS, (b) OCDM, (c) CDM and (d) BDM models.
The fact that this ratio lies between 0.3 and 0.6 in the B-band (4000–5000 Å) tells us
that expansion reduces the intensity of the night sky at optical wavelengths by a factor of
between two and three.
Despite a superficial resemblance to their counterparts (3.34) in the expanding

Universe, these are vastly different expressions. Most importantly, the SEDs
Fn (λ0 ) and Fs (λ0 ) no longer depend on z and have been pulled out of the integrals.
The quantity Iλ,stat (λ0 ) is effectively a weighted mean of the SEDs Fn (λ0 ) and
Fs (λ0 ). The weighting
z factors (i.e. the integrals over z) are related to the age
of the galaxies, 0 f dz/(1 + z) H̃ (z), but modified by factors of n n (z) and n s (z)
under the integral. This latter modification is important because it prevents the
integrals from increasing without limit as z f becomes arbitrarily large, a problem
that would otherwise introduce considerable uncertainty into any attempt to put
bounds on the ratio Iλ,stat /Iλ [6]. A numerical check confirms that Iλ,stat is nearly
as insensitive to the value of z f as Iλ , increasing by up to 8% as one moves from
z f = 3 to z f = 6, but with no further increase as z f 6.
The ratio of Iλ /Iλ,stat is plotted over 2000–25 000 Å in figure 3.8, where we
have set z f = 6. (Results are insensitive to this choice, as we have mentioned
earlier and it may be noted that they are also independent of uncertainty in
constants such as 0 since these are common to both Iλ and Iλ,stat .) Several
features in this figure deserve notice. First, the average value of Iλ /Iλ,stat across
Back to Olbers 63
the spectrum is about 0.6, consistent with bolometric expectations (chapter 2).
Second, the diagonal, bottom-left to top-right orientation arises largely because
Iλ (λ0 ) drops off at short wavelengths, while Iλ,stat (λ0 ) does so at long ones. The
reason why Iλ (λ0 ) drops off at short wavelengths is that ultraviolet light reaches
us only from the nearest galaxies; anything from more distant ones is redshifted
into the optical. The reason why Iλ,stat (λ0 ) drops off at long wavelengths is
because it is a weighted mixture of the galaxy SEDs, and drops off at exactly the
same place that they do: λ0 ∼ 3 × 104 Å. In fact, the weighting is heavily tilted
toward the dominant starburst component, so that the two sharp bends apparent in
figure 3.8 are essentially (inverted) reflections of features in Fs (λ0 ); namely, the
small bump at λ0 ∼ 4000 Å and the shoulder at λ0 ∼ 11 000 Å (figure 3.5).
Finally, the numbers: figure 3.8 shows that the ratio of Iλ /Iλ,stat is
remarkably consistent across the B-band (4000–5000 Å) in all four cosmological
models, varying from a high of 0.46±0.10 in the EdS model to a low of 0.39±0.08
in the BDM model. These numbers should be compared with the bolometric
result of Q/Q stat ≈ 0.6 ± 0.1 from chapter 2. They tell us that expansion does
play a greater role in determining B-band EBL intensity than it does across the
spectrum as a whole—but not by much. If its effects were removed, the night sky
at optical wavelengths would be anywhere from twice as bright (in the EdS model)
to three times brighter (in the BDM model). These results depend modestly on
the makeup of the evolving galaxy population and figure 3.8 shows that Iλ /Iλ,stat
in every case is highest for the weak-evolution model 1, and lowest for the strong-
evolution model 4. This is as we would expect, based on our discussion at
the beginning of this chapter: models with the strongest evolution effectively
‘concentrate’ their light production over the shortest possible interval in time,
so that the importance of the lifetime factor drops relative to that of expansion.
Our numerical results, however, prove that this effect cannot qualitatively alter the
resolution of Olbers’ paradox. Whether intensity is reduced by two times or three
due to expansion, its order of magnitude is still set by the lifetime of the galaxies.
There is one factor which we have not considered in this chapter, and that is
the possibility of absorption by intergalactic dust and neutral hydrogen, both of
which are strongly absorbing at ultraviolet wavelengths. The effect of this would
primarily be to remove ultraviolet light from high-redshift galaxies and transfer
it into the infrared—light that would otherwise be redshifted into the optical and
contribute to the EBL. (Dust extinction thus acts in the opposite sense to enhanced
galaxy evolution, which effectively injects ultraviolet light at high redshifts.) The
EBL intensity Iλ (λ0 ) would therefore drop, and one could expect reductions over
the B-band in particular. The size of this effect is difficult to assess because
we have limited data on the character and distribution of dust beyond our own
galaxy. We will find indications in chapter 7, however, that the reduction could
be significant at the shortest wavelengths considered here (λ0 ≈ 2000 Å) for the
most extreme dust models. This would further widen the gap between observed
and predicted EBL intensities noted at the end of section 3.6.
Absorption would play far less of a role in the equivalent static models,
where there is no redshift. (Ultraviolet light would still be absorbed, but the
effect would not carry over into the optical.) Therefore, the ratio Iλ /Iλ,stat would
be expected to drop in nearly direct proportion to the drop in Iλ . In this sense
Olbers may actually have had part of the solution after all—not (as he thought)
because intervening matter ‘blocks’ the light from distant sources but because
it transfers it out of the optical. The importance of this effect, which would be
somewhere below that of expansion, is, however, a separate issue from the one
we have concerned ourselves with in this chapter. What we have shown is that the
optical sky, like the bolometric one, is dark at night mainly because it has not had
time to fill up with light from distant galaxies. Cosmic expansion darkens it further
by a factor which depends on background cosmology and galaxy evolution, but
which lies between two and three in any case.
References
[1] Peacock J A 1999 Cosmological Physics (Cambridge: Cambridge University Press)
p 354
[2] McVittie G C and Wyatt S P 1959 Astrophys. J. 130 1
[3] Whitrow G J and Yallop B D 1964 Mon. Not. R. Astron. Soc. 127 130
[4] Whitrow G J and Yallop B D 1965 Mon. Not. R. Astron. Soc. 130 31
[6] Wesson P S 1991 Astrophys. J. 367 399
[7] Fukugita M, Hogan C J and Peebles P J E 1996 Nature 381 489
[9] Sawicki M J, Lin H and Yee H K C 1997 Astron. J. 113 1
[10] Toller G N 1983 Astrophys. J. 266 L79
[11] Henry R C 1999 Astrophys. J. 516 L49
[12] Lillie C F and Witt A N 1976 Astrophys. J. 208 64
[13] Spinrad H and Stone R P S 1978 Astrophys. J. 226 609
[15] Boughn S P and Kuhn J R 1986 Astrophys. J. 309 33
[16] Jakobsen P et al 1984 Astron. Astrophys. 139 481
[17] Tennyson P D et al 1988 Astrophys. J. 330 435
[18] Murthy J et al 1990 Astron. Astrophys. 231 187
[19] Hauser M G et al 1998 Astrophys. J. 508 25
[20] Wright E L and Reese E D 2000 Astrophys. J. 545 43
[21] Cambrésy L et al 2001 Astrophys. J. 555 563
[22] Bernstein R A 1998 PhD Thesis California Institute of Technology
[24] Ellis R et al 2001 Astrophys. J. 560 L119
[25] Partridge R B and Peebles P J E 1967 Astrophys. J. 148 377
[26] Tinsley N M 1973 Astron. Astrophys. 24 89
[27] Bruzual A G 1981 PhD Thesis University of California, Berkeley
Back to Olbers 65
[28] Code A D and Welch G A 1982 Astrophys. J. 256 1

[29] Yoshii Y and Takahara F 1988 Astrophys. J. 326
[30] Abraham R G and van den Bergh S 2001 Science 293 1273
[31] Kinney A L et al 1996 Astrophys. J. 467 38
[32] Devriendt J E G, Guiderdoni B and Sadat R 1999 Astron. Astrophys. 350 381
[33] Totani T and Yoshii Y 2000 Astrophys. J. 540 81
[34] Totani T and Yoshii Y 1998 Astrophys. J. 501 L177
[35] Pearson C and Rowan-Robinson M 1996 Mon. Not. R. Astron. Soc. 283 174
[36] Totani T et al 2001 Astrophys. J. 550 L137
[37] Jimenez R and Kashlinsky A 1999 Astrophys. J. 511 16
[38] Yasuda N et al 2001 Astron. J. 122 1104
Chapter 4
The dark matter
4.1 From light to dark matter

In calculating the intensity of the summed light from distant galaxies, we have
effectively taken a census of one component of the Universe: its luminous matter.
In astronomy, where nearly everything we know comes to us in the form of light
signals from vast distances, one might be forgiven for thinking that luminous
matter was the only kind that counted. This supposition, however, turns out to
be spectacularly wrong. The density lum of luminous matter is now thought
to comprise less than 1% of the total density tot of all forms of matter and
energy put together. (Here as in chapters 2 and 3, we express densities in units
of the critical density, equation (2.36), and denote them with the symbol .) The
remaining 99% or more consists of dark matter which, while not seen directly,
is inferred to exist from its gravitational influence on luminous matter as well as
the geometry of the Universe.
The identity of this unseen material, whose existence was first suspected in
the 1920s and 1930s by astronomers such as Kapteyn [1], Oort [2] and Zwicky
[3], has now become the central mystery of modern cosmology. In subsequent
chapters, we will examine some of the theoretical dark-matter candidates which
have been proposed, concentrating on the fact that many are not perfectly black.
For those which can decay either directly or indirectly into photons (and this
turns out to be most of them), we can use the same formalism we have applied
to galaxies in chapters 2 and 3. That is, we can calculate the intensity of
their contributions to the spectrum of background radiation reaching our Galaxy,
and compare with observational bounds over wavebands stretching from long-
wavelength radio waves to high-energy γ -rays. We will find that this procedure
allows us to narrow down the list of plausible contenders quite substantially.
In the present chapter, we will outline what is known about dark matter in
general, and how it is that we know it exists when it does not shine. Observation
and experiment over the past few years have increasingly suggested that there
are, in fact, four distinct categories of dark matter, three of which imply new
66
The four elements of modern cosmology 67
physics beyond the existing standard model of particle interactions. This is an

unexpected development, and one whose supporting evidence deserves to be
carefully scrutinized. We proceed to give a brief overview of the current situation,
followed by a closer look at the arguments for all four parts of nature’s ‘dark side’.
4.2 The four elements of modern cosmology
At least some of the dark matter, such as that contained in planets and ‘failed stars’
too dim to see, must be composed of ordinary atoms and molecules. The same
applies to dark gas and dust (although these can sometimes be seen in absorption,
if not emission). Such contributions are termed baryonic dark matter (BDM),
which combined together with luminous matter gives the total baryonic matter
density bar ≡ lum + bdm . If our understanding of big-bang theory and the
formation of the light elements is correct, then we will see that bar cannot
represent more than 5% of the critical density.
What, then, makes up the bulk of the Universe? Besides the dark baryons, it
now appears that three other varieties of dark matter play a role. The first of these
is cold dark matter (CDM), the existence of which has been inferred from the
behaviour of visible matter on scales larger than the Solar System (e.g. galaxies
and clusters of galaxies). CDM is thought to consist of particles (sometimes
referred to as ‘exotic’ dark-matter particles) whose interactions with ordinary
matter are so weak that they are seen primarily via their gravitational influence.
While they have not been detected (and are indeed hard to detect by definition),
such particles are predicted by plausible extensions of the standard model. The
overall CDM density cdm is believed by many cosmologists to exceed that of the
baryons (bar ) by at least an order of magnitude.
Another piece of the puzzle is provided by neutrinos, particles whose
existence is unquestioned but whose collective density (ν ) depends on their
rest mass, which is not yet known. If neutrinos are massless, or nearly so, then
they remain relativistic throughout the history of the Universe and behave for
dynamical purposes like photons. In this case neutrino contributions combine
with those of photons (γ ) to give the present radiation density as r,0 =
ν + γ . This is known to be very small. If, however, neutrinos are sufficiently
massive, then they are no longer relativistic on average, and belong together with
baryonic and cold dark matter under the category of pressureless matter, with
present density m,0 = bar +cdm +ν . These neutrinos could play a significant
dynamical role, especially in the formation of large-scale structure in the early
Universe, where they are sometimes known as hot dark matter (HDM). Recent
experimental evidence suggests that neutrinos do contribute to m,0 but at levels
below those of the baryons.
Influential only over the largest scales—those comparable to the
cosmological horizon itself—is the last component of the unseen Universe:
vacuum energy. Its many alternative names (the zero-point field, quintessence,
68 The dark matter
dark energy and the cosmological constant ) betray the fact that there is, at
present, no consensus as to where vacuum energy originates; or how to calculate
its energy density ( ) from first principles. Existing theoretical estimates of
this latter quantity range over some 120 orders of magnitude, prompting many
cosmologists until very recently to disregard it altogether. Observations of distant
supernovae and CMB fluctuations, however, increasingly imply that vacuum
energy is not only real but that its present energy density (,0 ) exceeds that
of all other forms of matter (m,0 ) and radiation (r,0 ) put together.
If this account is correct, then the real Universe hardly resembles the one
we see at night. It is composed, to a first approximation, of invisible vacuum
energy whose physical origin remains obscure. A significant minority of its total
energy density is found in CDM particles, whose ‘exotic’ nature is also not yet
understood. Close inspection is needed to make out the further contribution of
neutrinos, although this too is non-zero. And baryons, the stuff of which we
are made, are little more than a cosmic afterthought. This picture, if confirmed,
constitutes a revolution of Copernican proportions, for it is not only our location
in space which turns out to be undistinguished, but our very makeup. The ‘four
elements’ of modern cosmology are shown schematically in figure 4.1.
4.3 Baryons
Let us now go over the evidence for these four species of dark matter more
carefully, beginning with the baryons. The total present density of luminous
baryonic matter can be inferred from the observed luminosity density of the
Universe, if various reasonable assumptions are made about the fraction of
galaxies of different morphological type, their ratios of disc-type to bulge-type
stars, and so on. A recent and thorough such estimate is that of Fukugita et al [4]:
lum = (0.0027 ± 0.0014)h −1

0 . (4.1)
Here h 0 is, as usual, the value of Hubble’s constant expressed in units

of 100 km s−1 Mpc−1 . While this parameter (and hence the experimental
uncertainty in H0) factored out of the EBL intensities in chapters 2 and 3, it must
be squarely faced where densities are concerned. We therefore digress briefly to
discuss the observational status of h 0 .
Using various relative-distance methods, all calibrated against the distance to
Cepheid variables in the Large Magellanic Cloud (LMC), the Hubble Key Project
(HKP) team has determined that h 0 = 0.72 ± 0.08 [5]. Independent ‘absolute’
methods (e.g. time delays in gravitational lenses, the Sunyaev–Zeldovich effect in
the CMB, and the Baade–Wesselink method applied to supernovae) have higher
uncertainties but are roughly consistent with this, giving h 0 ≈ 0.55–0.74 [6].
This level of agreement is a great improvement over the factor-two discrepancies
of previous decades.
Baryons 69
Figure 4.1. Top: the ‘four elements’ of ancient cosmology. These are widely attributed to
the Greek philosopher Empedocles, who introduced them as follows, ‘Hear first the roots
of all things: bright Zeus (fire), life-giving air (Hera), Aidoneus (earth) and Nestis (water),
who moistens the springs of mortals with her tears’ (Fragments, c. 450 BC). Bottom:
the modern counterparts. (Figure adapted from a 1519 edition of Aristotle’s De caelo
by Johann Eck.)
There are signs, however, that we are still some way from ‘precision’ values
with uncertainties of less than 10%. A recalibrated LMC Cepheid period–
luminosity relation based on a much larger sample (from the OGLE microlensing
survey) leads to considerably higher values, namely h 0 = 0.85 ± 0.05 [7].
A new, purely geometric technique, based on the use of long-baseline radio
interferometry to measure the transverse velocity of water masers [8], also implies
that the traditional calibration is off, raising all Cepheid-based estimates by
12 ± 9% [9]. This would boost the HKP value to h 0 = 0.81 ± 0.09. There
is some independent support for such a recalibration in recent observations of
eclipsing binaries [10] and ‘red-clump stars’ in the LMC [11], although this—
like most conclusions involving the value of h 0 —has been disputed [12]. If the
LMC-based numbers do go up, then one would be faced with explaining the lower
h 0 values from the absolute methods. Here the choice of cosmological model
70 The dark matter
(e.g. as defined in table 3.1) can be important. Gravitational lensing values of h 0

(routinely reported assuming an EdS model) rise on average by 9% in OCDM,
7% in CDM and 3% in BDM for the six lens systems whose time delays have
been measured so far.
On this subject, history encourages caution. Where it is necessary to specify
the value of h 0 in this book, we will adopt the range
h 0 = 0.75 ± 0.15. (4.2)
Values at the edges of this range can discriminate powerfully between different
cosmological models. To a large extent this is a function of their ages, which can
be computed by integrating (2.44) or (in the case of flat models) directly from
(2.70). Alternatively, one can integrate the Friedmann–Lemaı̂tre equation (2.40)
numerically backward in time. Since this equation defines the expansion rate
H ≡ Ṙ/R, its integral gives the scale factor R(t). We show the results in
figure 4.2 for our four standard cosmological test models (EdS, OCDM, CDM
and BDM). These are seen to have ages of 7h −1 −1 −1 −1
0 , 8h 0 , 10h 0 and 17h 0 Gyr
respectively. Abundances of radioactive thorium and uranium in metal-poor halo
stars imply a Milky Way age 16 ± 5 Gyr [13], setting a firm lower limit of 11 Gyr
on the age of the Universe. If h 0 lies at the upper end of its allowed range
(h 0 = 0.9), then the EdS and OCDM models would be ruled out on the basis
that they are not old enough to contain these stars (this is known as the age crisis
in low-,0 models). With h 0 at the bottom of the range (h 0 = 0.6), however,
only EdS comes close to being excluded. The EdS model thus defines one edge
of the spectrum of observationally viable models.
The BDM model faces the opposite problem: figure 4.2 gives its age as
17h −1
0 Gyr, or as high as 28 Gyr (if h 0 = 0.6). The latter number, in particular,
is well beyond the age of anything seen in our Galaxy. Of course, upper limits
on age are not as easy to set as lower ones. But following Copernican reasoning,
we do not expect to live in a galaxy which is unusually young. The age of a
‘typical’ galaxy can be estimated by recalling from chapter 3 that most galaxies
appear to have formed in the redshift range 2 º z f º 4. Since R/R0 = (1 + z)−1
from (2.15), this corresponds to a range of scale factors 0.33 R/R0 0.2.
For the BDM model, figure 4.2 shows that R(t)/R0 does not reach these values
until (5 ± 2) h −1
0 Gyr after the big bang. Thus galaxies would have an age of
about (12 ± 2)h −10 Gyr in the BDM model, and not more than 21 Gyr in any
case. This is close to upper limits which have been set on the age of the Universe
in models of this type, t0 < 24 ± 2 Gyr [14]. The BDM model, or something
close to it, probably defines a position opposite that of EdS on the spectrum of
observationally viable models.
For the other three models, figure 4.2 shows that galaxy formation must be
accomplished within less than 2 Gyr after the big bang. The reason this is able
to occur so quickly is that these models all contain significant amounts of CDM,
which decouples from the primordial plasma before the baryons and prepares
Baryons 71
1.5 EdS ( Ωm,0=1 , ΩΛ,0=0 )

OCDM ( Ωm,0=0.3 , ΩΛ,0=0 )
ΛCDM ( Ωm,0=0.3 , ΩΛ,0=0.7 )
ΛBDM ( Ωm,0=0.03 , ΩΛ,0=1 )
Now
1
z=2
R/R0
z=4
0.5
0
-15 -10 -5 0 5
t − t0 ( h0-1 Gyr )
Figure 4.2. Evolution of the cosmological scale factor R̃(t) ≡ R(t)/R0 as a function of
time for the cosmological test models introduced in table 3.1. Triangles indicate the range
2 z f 4 where the bulk of galaxy formation may have taken place.
potential wells for the baryons to fall into. This, as we will see, is one of the main
motivations for CDM.
Returning now to the density of luminous matter, we find with our values for
h 0 that equation (4.1) gives
lum = 0.0036 ± 0.0020. (4.3)
This is the basis for our statement (section 4.1) that the visible Universe makes up
less than 1% of the critical density.
It is, however, conceivable that most of the baryons are not visible. How
significant could such dark baryons be? The theory of primordial big-bang
nucleosynthesis provides us with an independent method for determining the
density of total baryonic matter in the Universe, based on the assumption that
the light elements we see today were forged in the furnace of the hot big bang.
Results using different light elements are roughly consistent, which is impressive
in itself. The primordial abundances of 4 He (by mass) and 7 Li (relative to
H) imply a baryon density of bar = (0.011 ± 0.005)h −2 0 [15]. By contrast,
new measurements based exclusively on the primordial D/H abundance give a
higher value with lower uncertainty: bar = (0.019 ± 0.002)h −2 0 [16]. Since
it appears premature at present to exclude either of these results, we choose an
intermediate value of bar = (0.016 ± 0.005)h −2 0 . Combining this with our range
72 The dark matter
of values (4.2) for h 0 , we conclude that
bar = 0.028 ± 0.012. (4.4)
This agrees very well with an entirely independent estimate of bar obtained by
adding up individual mass contributions from all known repositories of baryonic
matter via their estimated mass-to-light ratios [4]. It also provides the rationale
for our choice of m,0 = 0.03 in the baryonic dark-matter model (BDM). If
tot,0 is close to unity, then it follows from equation (4.4) that all the atoms and
molecules in existence combine to make up less than 5% of the Universe by mass.
The vast majority of these baryons, moreover, are invisible. The baryonic
dark-matter fraction f bdm (≡ bdm /bar ) = 1 − lum /bar is
f bdm = (87 ± 8)%. (4.5)
Here we have used equations (4.1) and (4.2), together with the range of values
for bar h 20 from nucleosynthesis. Where could these dark baryons be? One
possibility is that they are smoothly distributed in a gaseous intergalactic medium,
which would have to be strongly ionized in order to explain why it has not left a
more obvious absorption signature in the light from distant quasars. Observations
using OVI absorption lines as a tracer of ionization suggest that the contribution of
such material to bar is at least 0.003h −1
0 [17], comparable to lum . Simulations
are able to reproduce many observed features of the ‘forest’ of Lyman-α (Lyα)
absorbers with as much as 80–90% of the baryons in this form [18].
Dark baryonic matter could also be bound up in clumps of matter such as
substellar objects (jupiters, brown dwarfs) or stellar remnants (white, red and
black dwarfs, neutron stars, black holes). Substellar objects are not likely to make
a large contribution, given their small masses. Black holes are limited in the
opposite sense: they cannot be more massive than about 105 M since this would
lead to dramatic tidal disruptions and lensing effects which are not seen [19]. The
baryonic dark-matter clumps of most interest are therefore ones whose mass is
within a few orders of magnitude of M . Gravitational microlensing constraints
based on quasar variability do not seriously limit the cosmological density of
such objects at present, setting an upper bound of 0.1 (well above bar ) on their
combined contributions to m,0 in an EdS Universe [20].
The existence of dark massive compact halo objects (MACHOs) within
our own galactic halo has been confirmed by the MACHO microlensing survey of
LMC stars [21]. The inferred lensing masses lie in the range (0.15–0.9)M and
would account for between 8% and 50% of the high rotation velocities seen in the
outer parts of the Milky Way. (This depends on the choice of halo model and is an
extrapolation from the 15 or so events actually seen.) The identity of these objects
has been hotly debated, with some authors linking them to faint, fast-moving
objects apparently detected in the HDF [22]. It is unlikely that they could be
traditional white dwarfs, since these are formed from massive progenitors whose
metal-rich ejecta we do not see [23]. Ancient, low-mass (º0.6M ) red dwarfs
Cold dark matter 73
which have cooled into invisibility are one possibility. Existing bounds on such
objects [24] depend on many extrapolations from known populations. Degenerate
‘beige dwarfs’, which might be able to form above the hydrogen-burning mass
limit of 0.08M [25], have also been suggested.
4.4 Cold dark matter

The introduction of a second species of unseen dark matter into the Universe has
been primarily motivated in three ways:
(1) a range of observational arguments imply that the density parameter of total
gravitating non-relativistic matter (m,0 = bar + cdm + ν ) is higher than
that provided by baryons and neutrinos alone;
(2) our current understanding of large-scale structure formation requires the
process to be helped along by quantities of non-relativistic, weakly
interacting matter in the early Universe, creating the potential wells for
infalling baryons; and
(3) theoretical physics supplies several plausible (albeit still undetected)
candidate CDM particles with the right properties.
Since our ideas on structure formation may change, and the candidate
particles may not materialize, the case for CDM turns at present on the
observational arguments. At one time, these were compatible with cdm ≈ 1,
raising hopes that CDM would resolve two of the biggest challenges in cosmology
at a single stroke: accounting for observations of large-scale structure and
providing all the dark matter necessary to make m,0 = 1, vindicating the EdS
model (and with it, the simplest models of inflation). Observations, however, no
longer support values of m,0 this high, and independent evidence now points
to at least two other kinds of non-baryonic dark matter (neutrinos and vacuum
energy). The CDM hypothesis is, therefore, no longer as compelling as it
once was. With this in mind we will pay special attention to the observational
arguments in this section. The lower limits on m,0 , in particular, are crucial:
only if m,0 > bar + ν do we require cdm > 0.
The arguments can be broken into two classes: those which are purely
empirical; and those which assume, in addition, the validity of the gravitational
instability theory of structure formation. Let us begin with the empirical
arguments. The first has already been encountered in section 4.3: the spiral
galaxy rotation curve. If the MACHO results are taken at face value, and if
the Milky Way is typical, then compact objects make up less than 50% of the
mass of the halos of spiral galaxies. If, as has been argued [26], the remaining
halo mass cannot be attributed to baryonic matter in known forms such as dust,
rocks, planets, gas or hydrogen snowballs, then a more exotic form of dark matter
is required.
The total mass of dark matter in galaxies, however, is limited. The easiest
way to see this is to compare the mass-to-light ratio (M/L) of our own Galaxy
74 The dark matter
to that of the Universe as a whole. If the latter is flat, then its M/L ratio is just
the ratio of the critical density to its luminosity density. This is (M/L)crit,0 =
ρcrit,0 /0 = (1040 ± 230)M /L , where we have used (2.24) for 0 , (2.36) for
ρcrit,0 and (4.2) for h 0 . The corresponding value for the Milky Way is (M/L)mw =
(21 ± 7)M /L , since the latter’s luminosity is L mw = (2.3 ± 0.6) × 1010 L
(in the B-band) and its total dynamical mass (including that of any unseen halo
component) is Mmw = (4.9 ± 1.1) × 1011 M inside 50 kpc from the motions
of galactic satellites [27]. The ratio of (M/L)mw to (M/L)crit,0 is thus less than
3%, and even if we multiply this by a factor of a few (to account for possible halo
mass outside 50 kpc), it is clear that galaxies like our own cannot make up more
than 10% of the critical density.
Most of the mass of the Universe, in other words, is spread over scales larger
than galaxies, and it is here that the arguments for CDM take on the most force.
The M/L-ratio method is, in fact, the most straightforward: one measures M/L
for a chosen region, corrects for the corresponding value in the ‘field’ and divides
by (M/L)crit,0 to obtain m,0 . Much, however, depends on the choice of region.
A widely respected application of this approach is that of the CNOC team [28],
which uses rich clusters of galaxies. These systems sample large volumes of
the early Universe, have dynamical masses which can be measured by three
independent methods (the virial theorem, x-ray gas temperatures and gravitational
lensing) and are subject to reasonably well-understood evolutionary effects. They
are found to have M/L ∼ 200M /L on average, giving m,0 = 0.19 ± 0.06
when ,0 = 0 [28]. This result scales as (1 − 0.4,0) [29], so that m,0 drops
to 0.11 ± 0.04 in a model with ,0 = 1.
The weak link in this chain of inference is that rich clusters may not be
characteristic of the Universe as a whole. Only about 10% of galaxies are found
in such systems. If individual galaxies (like the Milky Way, with M/L ≈
21M /L ) are substituted for clusters, then the inferred value of m,0 drops by a
factor of ten, approaching bar and removing the need for CDM. A recent effort to
address the impact of scale on M/L arguments concludes that m,0 = 0.16±0.05
(in flat models), when regions of all scales are considered from individual galaxies
to superclusters [30].
Another line of argument uses the ratio of baryonic to total gravitating mass
in clusters, or cluster baryon fraction (Mbar /M tot ). Baryonic matter is defined as
the sum of visible galaxies and hot cluster gas (the mass of which can be inferred
from the x-ray temperature). Total cluster mass is measured by one or all of the
three methods listed earlier (virial theorem, x-ray temperature or gravitational
lensing). At sufficiently large radii, the cluster may be taken as representative
of the Universe as a whole, so that m,0 = bar /(Mbar /M tot ), where bar is
fixed by big-bang nucleosynthesis (section 4.3). Applied to various clusters, this
procedure leads to m,0 = 0.3 ± 0.1 [31]. This result is probably an upper limit,
partly because baryon enrichment is more likely to take place inside the cluster
than out, and partly because dark baryonic matter (such as MACHOs) is not taken
into account; this would raise Mbar and lower m,0 .
Cold dark matter 75
A final, recent entry into the list of purely empirical methods uses the
separation of radio galaxy lobes as standard rulers in a variation on the classical
angular size–distance test in cosmology. The widths, propagation velocities and
magnetic field strengths of these lobes have been calibrated for 14 radio galaxies
with the aid of long-baseline radio interferometry. This leads to the constraints
m,0 < 0.10 (for ,0 = 0) and m,0 = 0.10+0.25 −0.10 for flat models with
,0 = 1 − m,0 [32].
Let us consider, next, the measurements of m,0 based on the idea that large-
scale structure formed via gravitational instability from a Gaussian spectrum of
primordial density fluctuations (GI theory for short). These are somewhat circular,
in the sense that such a process could not have taken place as it did unless m,0
is considerably larger than bar . But inasmuch as GI theory is the only structure-
formation theory we have which is both fully worked out and in good agreement
with observation (with some potential difficulties on small scales [33, 34]), this
way of measuring m,0 should be taken seriously.
According to GI theory, formation of large-scale structure is more or less
complete by z ≈ −1 m,0 − 2 [35]. Therefore, one way to constrain m,0 is to look
for number density evolution in large-scale structures such as galaxy clusters.
In a low-matter-density Universe, this would be relatively constant out to at least
z ∼ 1, whereas in a high-matter-density Universe one would expect the abundance
of clusters to drop rapidly with z because they are still in the process of forming.
The fact that massive clusters are seen at redshifts as high as z = 0.83 leads to
limits m,0 = 0.17+0.14 +0.13
−0.09 for ,0 = 0 models and m,0 = 0.22−0.07 for flat
ones [36].
Studies of the Fourier power spectra P(k) of the distributions of galaxies
or other structures can be used in a similar way. In GI theory, structures of a given
mass form by the collapse of large volumes in a low-matter-density Universe or
smaller volumes in a high-matter-density Universe. Thus m,0 can be constrained
by changes in P(k) between one redshift and another. Comparison of the mass
power spectrum of Lyα absorbers at z ≈ 2.5 with that of local galaxy clusters at
z = 0 has led to an estimate of m,0 = 0.46+0.12 −0.10 for ,0 = 0 models [37].
This result goes as approximately (1 − 0.4,0), so that the central value of m,0
drops to 0.34 in a flat model and 0.28 if ,0 = 1. One can also constrain
m,0 from the local galaxy power spectrum alone, although this involves some
assumptions about the extent to which ‘light traces mass’ (i.e. to which visible
galaxies trace the underlying density field). Early results from the 2dF survey
give m,0 = (0.20 ± 0.03)h −1 0 for flat models [38], or m,0 = 0.27 ± 0.07 with
our range of h 0 values as given by (4.2).
A final group of measurements, and one which has traditionally yielded the
highest estimates of m,0 , comes from the analysis of galaxy peculiar velocities.
These are produced by the gravitational potential of locally over- or under-
dense regions relative to the mean matter density. The power spectra of the
velocity and density distributions can be related to each other in the context of
GI theory in a way which depends explicitly on m,0 . Tests of this kind probe
76 The dark matter
relatively small volumes and are hence insensitive to ,0 , but they can depend
significantly on h 0 as well as the spectral index n of the density distribution.
In [39], where the latter is normalized to CMB fluctuations, results take the
form m,0 h 01.3 n 2 ≈ 0.33 ± 0.07 or (taking n = 1 and using our values of h 0 )
m,0 ≈ 0.48 ± 0.15.
In summarizing these results, one is struck by the fact that arguments based
on gravitational instability (GI) theory favour values of m,0 0.2 and higher,
whereas purely empirical arguments require m,0 0.4 and lower. The latter
are, in fact, compatible in some cases with values of m,0 as low as bar , raising
the possibility that CDM might not, in fact, be necessary. The results from GI-
based arguments, however, cannot be stretched this far. What is sometimes done
is to ‘go down the middle’ and blend the results of both kinds of arguments into
a single bound of the form m,0 ≈ 0.3 ± 0.1. Any such bound with m,0 > 0.05
constitutes a proof of the existence of CDM, since bar 0.04 from (4.4).
(Massive neutrinos do not affect this argument, as we will indicate in section 4.5.)
A more conservative interpretation of the data, bearing in mind the full range of
m,0 values implied above (bar m,0 0.6), is
cdm = 0.3 ± 0.3. (4.6)
Values of cdm at the bottom of this range, however, carry with them
the (uncomfortable) implication that the conventional picture of structure
formation via gravitational instability is incomplete. Conversely, if our current
understanding of structure formation is correct, then CDM must exist and
cdm > 0.
The question, of course, becomes moot if CDM is discovered in the
laboratory. From a large field of theoretical particle candidates, two have emerged
as frontrunners: the axion and the supersymmetric weakly-interacting massive
particle (WIMP). The plausibility of both candidates rests on three properties.
Either one, if it exists, is:
(1) weakly interacting (i.e. ‘noticed’ by ordinary matter primarily via its
gravitational influence);
(2) cold (i.e. non-relativistic in the early Universe, when structures begin to
form); and
(3) expected on theoretical grounds to have a collective density within a few
orders of magnitude of the critical one.
We will return to these particles in chapters 6 and 8 respectively.
4.5 Neutrinos
Since neutrinos indisputably exist in great numbers (their number density n ν is
3/11 times that of the CMB photons, or 112 cm−3 per species [40]), they have
been leading dark-matter candidates for longer than either the axion or the WIMP.
Neutrinos 77
They gained prominence in 1980 when teams in the USA and the Soviet Union
both reported evidence of non-zero neutrino rest masses. While these claims did
not stand up, a new round of experiments now indicates once again that m ν (and
hence ν ) > 0.
Dividing n ν m ν by the critical density (2.36), one obtains

ν = m ν c2 /93.8 eV h −2
0 (4.7)
where the sum is over the three neutrino species. Here, we follow convention
and specify particle masses in units of eV/c2 , where 1 eV/c2 = 1.602 ×
10−12 erg/c2 = 1.783 × 10−33 g. The calculations in this section are strictly
valid only for m ν c2 º 1 MeV. More massive neutrinos with m ν c2 ∼ 1 GeV were
once considered as CDM candidates but are no longer viable since experiments
at the LEP collider rule out additional neutrino species with masses up to at least
half of that of the Z 0 (m Z0 c2 = 91 GeV).
Current laboratory upper bounds on neutrino rest masses are m νe c2 < 3 eV,
m νµ c2 < 0.19 MeV and m ντ c2 < 18 MeV, so it would appear feasible in principle
for these particles to close the Universe. In fact m νµ and m ντ are limited far more
stringently by (4.7) than by laboratory bounds. Perhaps the best-known theory
along these lines is that of Sciama [41], who postulated a population of 29 eV
τ neutrinos which, if they decayed into lighter neutrinos on suitable timescales,
could be shown to solve a number of longstanding astrophysical puzzles related to
interstellar and intergalactic ionization. Equation (4.7) shows that such neutrinos
would also account for much of the dark matter, contributing a minimum collective
density of ν 0.38 (assuming as usual that h 0 0.9). We will consider the
decaying-neutrino hypothesis in more detail in chapter 7.
Strong upper limits can be set on ν within the context of the gravitational
instability picture. Neutrinos constitute hot dark matter (i.e. they are relativistic
when they decouple from the primordial fireball) and are therefore able to stream
freely out of density perturbations in the early Universe, erasing them before they
have a chance to grow. Good agreement with observations of large-scale structure
can be achieved in models with ν as high as 0.2, but only if bar + cdm = 0.8
and h 0 = 0.5 [42]. Statistical exploration of a larger set of (flat) models leads
to an upper limit on m ν which is roughly proportional to the CDM density cdm
and peaks (when cdm = 0.6) at m ν c2 5.5 eV [43]. Since 0 cdm 0.6
from (4.6), this implies
m ν c2 (9.2 eV)cdm . (4.8)
If h 0 0.6 and cdm 0.6, and if structure grows via gravitational instability as
is generally assumed, then equations (4.7) and (4.8) together allow a collective
neutrino density as high as ν = 0.16. Thus neutrino contributions could,
in principle, be as much as four to ten times as high as those from baryons,
equation (4.4). This is perfectly consistent with the neglect of relativistic matter
during the matter-dominated era (section 2.5). Neutrinos, while relativistic
at decoupling, lose energy and become non-relativistic on timescales tnr ≈
78 The dark matter
190 000 yr (m ν c2 /eV)−2 [44]. This is well before the present epoch for neutrinos
which are massive enough to be of interest.
New lower limits on ν have been reported in recent years from atmospheric
(Super-Kamiokande [45]), Solar (SAGE [46], Homestake [47], GALLEX [48]),
and accelerator-based (LSND [49]) neutrino experiments. In each case it
appears that two species are converting into each other in a process known
as neutrino oscillation, which can only take place if both have non-zero rest
masses. The strongest evidence comes from Super-Kamiokande, which has
reported oscillations between τ and µ neutrinos with 5 × 10−4 eV2 < m 2τ µ c2 <
6×10−3 eV2 , where m 2τ µ ≡ |m 2ντ −m 2νµ | [45]. Oscillation experiments measure
the square of the mass difference between two species, and cannot fix the mass
of any one species unless they are combined with other evidence such as that
from neutrinoless double-beta decay [50]. Nevertheless, if neutrino masses
are hierarchical, like those of other fermions, then m ντ m νµ and the Super-
Kamiokande results imply that m ντ c2 > 0.02 eV. In this case it follows from (4.7)
that ν 0.0003 (with h 0 0.9 as usual). If, instead, neutrino masses are nearly
degenerate, then ν could, in principle, be much larger than this but will still lie
below the upper bound imposed by structure formation. Putting (4.8) into (4.7),
we conclude that
0.0003 ν < 0.3cdm . (4.9)
The neutrino contribution to tot,0 is anywhere from an order of magnitude below

that of the visible stars and galaxies (section 4.3) up to nearly one-third the amount
conventionally attributed to CDM (section 4.4). We emphasize that low values of
cdm imply low values of ν also. Physically, this reflects the fact that neutrinos
interfere with structure formation in the early Universe unless something like
CDM is present to help hold primordial density perturbations together. In theories
(like that to be discussed in chapter 6) where the density of neutrinos exceeds that
in (4.9), one would need to modify the standard gravitational instability picture
by encouraging the growth of early structures in some other way, as for instance
by ‘seeding’ them with loops of cosmic string.
4.6 Vacuum energy

There are at least four reasons to include a cosmological constant () in
Einstein’s field equations (2.1). The first is mathematical: plays a role in these
equations similar to that of the additive constant in an indefinite integral [51]. The
second is dimensional: specifies the radius of curvature R ≡ −1/2 in closed
models at the moment when the matter density parameter m goes through its
maximum, thereby providing a fundamental length scale for cosmology [52]. The
third is dynamical: determines the asymptotic expansion rate of the Universe
according to equation (2.41). And the fourth is material: is related to the energy
density of the vacuum via equation (2.38), ,0 = c2 /3H02.
Vacuum energy 79
Table 4.1. Theoretical estimates of ,0 .
Theory Estimated value of ρ ,0
QCD (0.3 GeV)4 −3 c−5 = 1016 g cm−3 1044 h −2

0
EW (200 GeV)4 −3 c−5 = 1026 g cm−3 1055 h −2
0
GUTs (1019 GeV)4 −3 c−5 = 1093 g cm−3 10122 h −2
0
With all these reasons to take this term seriously, why have many
cosmologists since Einstein set = 0? Mathematical simplicity is one
explanation. Another is the smallness of most effects associated with the -term.
Einstein himself set = 0 in 1931 ‘for reasons of logical economy’, because
he saw no hope of measuring this quantity experimentally at the time. He is
often quoted as adding that its introduction in 1915 was the biggest blunder of his
life. This comment (which does not appear in Einstein’s writings but was rather
attributed to him by Gamow [53]) is sometimes interpreted as a rejection of the
very idea of a cosmological constant. It more likely represents Einstein’s rueful
recognition that, by invoking the -term solely to obtain a static solution of the
field equations, he had narrowly missed what would surely have been one of the
greatest triumphs of his life: the prediction of cosmic expansion.
The relation between and the energy density of the vacuum has led to a
quandary in more recent times: the fact that ρ as estimated in the context of
quantum field theories such as quantum chromodynamics (QCD), electroweak
(EW) and grand unified theories (GUTs) implies impossibly large values of
,0 (table 4.1). These theories have been very successful in the microscopic
realm. Here, however, they are in gross disagreement with the observed facts
of the macroscopic world, which tell us that ,0 cannot be much larger than
order unity. This cosmological-constant problem has been reviewed by several
workers but there is no consensus on how to solve it [54]. It is undoubtedly
another reason why many cosmologists have preferred to set = 0, rather than
deal with a parameter whose microphysical origins are still unclear.
Setting to zero, however, is not really an appropriate response because
observations indicate that ,0 , while nowhere near the size suggested by
table 4.1, is nevertheless greater than zero. The cosmological constant problem
has therefore become more baffling, in that an explanation of this parameter must
apparently contain a cancellation mechanism which is not only good to some 44
(or 122) decimal places, but which begins to fail at precisely the 45th (or 123rd).
One suggestion for understanding the possible nature of such a cancellation
has been to treat the vacuum energy field literally as an Olbers-type summation
of contributions from different places in the Universe [55]. It can then be
handled with the same formalism that we have developed in chapters 2 and 3
for background radiation. This has the virtue of framing the problem in concrete
80 The dark matter
terms, and raises some interesting possibilities, but does not in itself explain why
the energy density inherent in such a field does not gravitate in the conventional
way [56]. Another idea is that theoretical expectations for the value of of
might refer only to the latter’s primordial or ‘bare’ value, which could have
been progressively ‘screened’ over time. The cosmological constant would thus
become a variable cosmological term (see [57] for a review). In such a scenario
the present ‘low’ value of ,0 would simply reflect the fact that the Universe is
old. In general, however, this means modifying Einstein’s field equations (2.1)
and/or introducing new forms of matter such as scalar fields. We will look at this
suggestion in more detail in chapter 5.
A third possibility occurs in higher-dimensional gravity, where the
cosmological constant can arise as an artefact of dimensional reduction (i.e.
in extracting the appropriate four-dimensional limit from the theory). In such
theories it is possible that the ‘effective’ 4 could be small while its N-
dimensional analogue N is large [58]. We will consider some aspects of
higher-dimensional gravity in chapter 9. Some workers, finally, have argued
that a Universe in which was too large might be incapable of giving rise to
intelligent observers. That is, the fact of our own existence might already ‘require’
,0 ∼ 1 [59]. This is an application of the anthropic principle whose status,
however, remains unclear.
Let us pass now to what is known about the value of ,0 from cosmology.
It is widely believed that the Universe originated in a big-bang singularity rather
than passing through a ‘big bounce’ at the beginning of the current expansionary
phase. By differentiating the Friedmann–Lemaı̂tre equation (2.40) and setting
both the expansion rate and its time derivative to zero, one finds that this implies
an upper limit (sometimes called the Einstein limit ,E ) on ,0 as a function
of m,0 . Models with = ,E , known as Eddington–Lemaı̂tre models, are
asymptotic to Einstein’s original static model in the infinite past. The quantity
,E can be expressed as follows for cases with m,0 < 0.5 [60]:
1/3
2/3
,E = 1 − m,0 + 32 m,0 1 − m,0 + 1 − 2m,0
1/3
+ 1 − m,0 − 1 − 2m,0 . (4.10)
For m,0 = 0.3 the requirement that ,0 < ,E implies ,0 < 1.71, a limit
that tightens to ,0 < 1.16 for m,0 = 0.03.
A slightly stronger constraint can be formulated (for closed models) in
terms of the antipodal redshift (z a ). The antipodes are defined as the set of
points located on the ‘other side of the Universe’ at χ = π, where dχ (the
radial coordinate distance element) is given by dχ = (1 − kr 2 )−1/2 dr from
the metric (2.2). Using (2.10) and (2.18) this can be put into the form dχ =
−(c/H0 R0 ) dz/ H̃ (z), which can be integrated from z = 0 to z a with the help of
(2.39) and (2.40). Gravitational lensing of sources beyond the antipodes cannot
Vacuum energy 81
give rise to normal (multiple) images [61], so the redshift z a of the antipodes
must exceed that of the most distant normally-lensed object, currently a galaxy at
z = 4.92 [62]. Requiring that z a > 4.92 leads to the upper bound ,0 < 1.57
if m,0 = 0.3. This tightens to ,0 < 1.15 for BDM-type models with
m,0 = 0.03.
Gravitational lensing also leads to a different and considerably more
stringent upper limit (for all models) based on lensing statistics. The increase in
path length to a given redshift in vacuum-dominated models (relative to, say, the
EdS model) means that there are more sources to be lensed, and presumably more
lensed objects to be seen. The observed frequency of lensed quasars, however,
is rather modest, leading to the bound ,0 < 0.66 for flat models [63]. Dust
could hide distant sources [64]. However, radio lenses should be far less affected
and these give only slightly weaker constraints: ,0 < 0.73 (for flat models)
or ,0 º 0.4 + 1.5m,0 (for curved ones) [65]. Uncertainties arise in lens
modelling and evolution as well as source redshifts and survey completeness. A
recent conservative limit from radio lenses is ,0 < 0.95 for flat models [66].
Tentative lower limits have been set on ,0 using faint galaxy number
counts. This method is based on a similar premise to that of lensing statistics:
the enhanced comoving volume at large redshifts in vacuum-dominated models
should lead to greater (projected) galaxy number densities at faint magnitudes. In
practice, it has proven difficult to disentangle this effect from galaxy luminosity
evolution. Early claims of a best fit at ,0 ≈ 0.9 [67] have been disputed on the
basis that the steep increase seen in numbers of blue galaxies is not matched in
the K-band, where luminosity evolution should be less important [68]. Attempts
to account for evolution in a comprehensive way have subsequently led to a lower
limit of ,0 > 0.53 [69] and, most recently, a reasonable fit (for flat models)
with a vacuum density parameter of ,0 = 0.8 [70].
Other evidence for a significant ,0 term has come from numerical
simulations of large-scale structure formation. Figure 4.3 shows the evolution
of massive structures between z = 3 and z = 0 in simulations by the VIRGO
Consortium [71]. Of the two models shown, CDM (top row) is qualitatively
closer to the observed distribution of galaxies in the real Universe than EdS
(‘SCDM’, bottom row). The improvement is especially marked at higher redshifts
(left-hand panels). Power spectrum analysis, however, reveals that the match is
not particularly good in either case [71]. This may be a reflection of bias (i.e. of
a systematic discrepancy between the distributions of mass and light). Different
combinations of m,0 and ,0 might also provide better fits. Simulations of
closed BDM-type models would be of particular interest [72, 73].
The first measurements to put both lower and upper bounds on ,0 have
come from Type Ia supernovae (SNIa). These objects have luminosities which
are both high and consistent, and they are thought not to evolve significantly
with redshift. All of these properties make them ideal for use in the classical
magnitude–redshift relation. Two independent groups (HZT [74] and SCP [75])
have reported a systematic dimming of SNIa at z ≈ 0.5 by about 0.25 magnitudes
82 The dark matter
Figure 4.3. Numerical simulations of structure formation. In the top row is the CDM
model with m,0 = 0.3, ,0 = 0.7 and h 0 = 0.7. The bottom row shows the EdS
(‘SCDM’) model with m,0 = 1, ,0 = 0 and h 0 = 0.5. The panel size is comoving
with the Hubble expansion, and time runs from left (z = 3) to right (z = 0). (Images
courtesy of J Colberg and the VIRGO Consortium.)
relative to that expected in an EdS model, suggesting that space at these redshifts
is ‘stretched’ by a significant vacuum term. The 2σ confidence intervals from
both studies may be summarized as 0.8m,0 − 0.6,0 = −0.2 ± 0.3, or
,0 = 43 m,0 + 1
3 ± 12 . (4.11)
From this result it follows that ,0 > 0 for any m,0 > 1/8, and that ,0 > 0.2
if m,0 = 0.3. These numbers are in disagreement with both the EdS and OCDM
models. To extract limits on ,0 alone, we recall that m,0 bar 0.02
(section 4.3) and m,0 0.6 (section 4.4). If we combine these conservative
bounds with a single limit, m,0 = 0.31 ± 0.29, then (4.11) gives
,0 = 0.75 ± 0.63. (4.12)
This is not a high-precision measurement but it is enough to establish that

,0 > 0.1 and hence that the vacuum energy is real. Several words of caution
are in order, however. Intergalactic dust could also cause a systematic dimming
(without reddening) if it were ‘sifted’ during the process of ejection from galaxies
[76]. The neglect of evolution may be more serious than claimed [77]. And
much remains to be understood about the physics of SNIa explosions [78]. It is
likely that observations will have to reach z ∼ 2 before the SNIa magnitude–
Vacuum energy 83
redshift relation is able to discriminate statistically between models (like CDM

and BDM) with different ratios of m,0 to ,0 .
Further support for the existence of ‘dark energy’ has arisen from a
completely independent source: the angular power spectrum of CMB
fluctuations. These are produced by density waves in the primordial plasma,
termed by Lineweaver ‘the oldest music in the Universe’ [79]. The first peak
in their power spectrum picks out the angular size of the largest fluctuations
in this plasma at the moment when the Universe became transparent to light.
Because it is seen through the ‘lens’ of a curved Universe, the location of this
peak is sensitive to the latter’s total density ,0 + m,0 . Results have been
reported from three different experiments (BOOMERANG [80], MAXIMA [81]
and DASI [82]) and combined with earlier data from the COBE satellite [83].
BOOMERANG and MAXIMA conclusions have been analysed together to yield
m,0 + ,0 = 1.11+0.13−0.12 [84], while those from the more recent DASI team give
m,0 +,0 = 1.047+0.103−0.120 [82]. (These are reported 95% uncertainties using the
most conservative statistical priors.) Let us summarize these results as follows:
m,0 + ,0 = 1.08 ± 0.16. (4.13)
The Universe is therefore spatially flat or very close to it. To extract a value
for ,0 alone, we can do as in the SNIa case and substitute our matter-density
bounds (m,0 = 0.31 ± 0.29) into (4.13) to obtain
,0 = 0.77 ± 0.33. (4.14)
This is consistent with (4.12), but has error bars which have been reduced
by 50% and are now due almost entirely to the uncertainty in m,0 . This
measurement is impervious to most of the uncertainties of the earlier ones,
because it leapfrogs ‘local’ systems whose interpretation is complex (supernovae,
galaxies, and quasars), going directly back to the radiation-dominated era when
physics was simpler. Equation (4.14) is sufficient to establish that ,0 > 0.4,
and hence that vacuum energy not only exists, but may very well dominate the
energy density of the Universe.
The CMB power spectrum favours vacuum-dominated models but is not yet
resolved with sufficient precision to discriminate (on its own) between models
which have exactly the critical density (like CDM) and those which are close
to the critical density (like BDM). As it stands, the location of the first peak
in these data actually hints at ‘marginally closed’ models, although the implied
departure from flatness is not statistically significant and could also be explained
in other ways [85].
Much attention is focused on the second- and higher-order peaks of the
spectrum, which contain valuable clues about the matter component. Odd-
numbered peaks are produced by regions of the primordial plasma which have
been maximally compressed by infalling material, and even ones correspond to
maximally rarefied regions which have rebounded due to photon pressure. A high
84 The dark matter
Figure 4.4. Observational constraints on the values of m,0 and ,0 from supernovae
data (SNIa) and the power spectrum of CMB fluctuations (BOOMERANG, MAXIMA).
Shown are 68%, 95% and 99.7% confidence intervals inferred both separately and jointly
from the data. The dashed line indicates spatially flat models with k = 0; models on the
lower left are closed while those on the upper right are open. (Reprinted from [84] by
permission of A Jaffe and P L Richards.)
baryon-to-photon ratio enhances the compressions and retards the rarefractions,

thus suppressing the size of the second peak relative to the first. The strength of
this effect depends on the fraction of baryons (relative to the more weakly-bound
neutrinos and CDM particles) in the overdense regions. The BOOMERANG
and MAXIMA data show an unexpectedly weak second peak. While there are
a number of ways to account for this in CDM models (e.g. by ‘tilting’ the
primordial spectrum), the data are fit most naturally by the BDM model with
m,0 = bar , cdm = 0 and ,0 ≈ 1 [86]. Models of this kind have been
advocated on other grounds over the years, notably in connection with an analysis
of Lyα absorption spectra [87] (see [88] for a review). Subsequent DASI data,
however, show a stronger second peak and are better fit by CDM [82]. These
issues will be resolved as data begin to come in from other experiments such as
the MAP and Planck satellites.
The best constraints on ,0 come from taking both the supernovae
and microwave background results at face value and substituting one into the
other. This provides a valuable cross-check on the matter density, because the
SNIa and CMB constraints are very nearly orthogonal in the m,0 –,0 plane
(figure 4.4). Thus, forgetting all about our conservative bounds on m,0 and
The coincidental Universe 85
merely substituting (4.13) into (4.11), we find

,0 = 0.76 ± 0.23. (4.15)
Alternatively, extracting the matter density parameter, we obtain
m,0 = 0.32 ± 0.22. (4.16)
These results further tighten the case for a vacuum-dominated Universe. Equation
(4.15) also implies that ,0 < 1, which begins to put pressure on the BDM
model. Perhaps most importantly, equation (4.16) establishes that m,0 0.1,
which is inconsistent with BDM and requires the existence of CDM. Moreover,
the fact that the range of values picked out by (4.16) agrees so well with
that derived in section 4.4 constitutes solid evidence for the CDM model in
particular, and for the gravitational instability picture of large-scale structure
formation in general.
The depth of the change in thinking that has been triggered by these
developments on the observational side can hardly be exaggerated. Only a few
years ago, it was still routine to set = 0 and cosmologists had two main choices:
the ‘one true faith’ (flat, with m,0 ≡ 1) or the ‘reformed’ (open, with individual
believers being free to choose their own values near m,0 ≈ 0.3). All this has
been irrevocably altered by the CMB experiments. If there is a guiding principle
now, it is no longer m,0 ≈ 0.3, and certainly not ,0 = 0; it is tot,0 ≈ 1 from
the power spectrum of the CMB. Cosmologists have been obliged to accept a -
term, and it is not so much a question of whether or not it dominates the energy
budget of the Universe but by how much.
4.7 The coincidental Universe

The observational evidence reviewed in the foregoing sections has led us
progressively toward the corner of parameter space occupied by vacuum-
dominated models with close to (or exactly) the critical density. The resulting
picture is self-consistent and agrees with nearly all the data. Major questions,
however, remain on the theoretical side. Prime among these is the problem of the
cosmological constant, which (as previously described) is particularly acute in
models with non-zero values of , because one can no longer hope that a simple
symmetry of nature will eventually be found which requires = 0.
A related concern has to do with the evolution of the matter and vacuum
energy density parameters m and over time. Equations (2.32) and (2.35)
combine to give
ρm (t) m,0 ρ (t) ,0
m (t) ≡ = (t) ≡ = . (4.17)
ρcrit (t) R̃ (t) H̃ 2(t)
3 ρcrit (t) H̃ 2(t)
Here H̃ [z(t)] is given by (2.40) as usual and z(t) = 1/ R̃(t) − 1 from (2.15).
Equations (4.17) can be solved exactly for flat models using (2.68) and (2.69) for
86 The dark matter
(a) (b)
1.5 ΛCDM (Ωm,0=0.3, ΩΛ,0=0.7): Ωm(t) 1.5 ΛCDM (Ωm,0=0.3, ΩΛ,0=0.7): Ωm(t)
ΩΛ(t) ΩΛ(t)
Matter, vacuum density parameters
Matter, vacuum density parameters

Nucleosynthesis ΛBDM (Ωm,0=0.03, ΩΛ,0=1): Ωm(t)
Decoupling ΩΛ(t)
Now Now
1 1
0.5 0.5
0 0
1e−18 1e−12 1e−06 1 1e+06 1e+12 1e+18 −20 −10 0 10 20 30 40 50 60 70 80
−1 −1
t ( h0 Gyr ) t − t0 ( h0 Gyr )
Figure 4.5. The evolution of m (t) and (t) in vacuum-dominated models. The
left-hand panel (a) shows a single model (CDM) over 20 powers of time in either
direction (after a similar plot against scale factor R in [54]). Plotted this way, we are
seen to live at a very special time (marked ‘Now’). Standard nucleosynthesis (tnuc ∼ 1 s)
and matter–radiation decoupling times (tdec ∼ 1011 s) are included for comparison. The
right-hand panel (b) shows both the CDM and BDM models on a linear rather than
logarithmic scale, for the first 100h −1
0 Gyr after the big bang (i.e. the lifetime of the stars
and galaxies).
R̃(t) and H̃ (t). Results for the CDM model are illustrated in figure 4.5(a). At
early times, the cosmological term is insignificant ( → 0 and m → 1), while
at late ones it entirely dominates and the Universe is effectively empty ( → 1
and m → 0).
What is remarkable in this figure is the location of the present (marked
‘Now’) in relation to the values of m and . We have apparently arrived
on the scene at the precise moment when these two parameters are in the midst
of switching places. (We have not considered radiation density r here, but
similar considerations apply to it.) This has come to be known as the coincidence
problem and Carroll [54] has aptly described such a Universe as ‘preposterous’,
writing: ‘This scenario staggers under the burden of its unnaturalness, but
nevertheless crosses the finish line well ahead of any of its competitors by
agreeing so well with the data’. Cosmology may indeed be moving toward a
position like that of particle physics, in which there is a standard model which
accurately accounts for all observed phenomena but which appears to be founded
on a series of finely-tuned parameter values which leave one with the distinct
impression that the underlying reality has not yet been grasped.
Figure 4.5(b) is a close-up view of figure 4.5(a), with one difference: it is
plotted on a linear scale in time for the first 100h −1
0 Gyr after the big bang, rather
than a logarithmic scale over 10±20h −1 0 Gyr. The rationale for this is simple:
100 Gyr is approximately the lifespan of the galaxies (as determined by their
main-sequence stellar populations). One would not, after all, expect observers
to appear on the scene long after all the galaxies had disappeared or, for that
matter, in the early stages of the expanding fireball. Seen from the perspective
of figure 4.5(b), the coincidence, while still striking, is perhaps no longer so
preposterous. However, the CDM model still appears fine-tuned, in that ‘Now’
follows rather quickly on the heels of the epoch of matter–vacuum equality. In the
BDM model, m,0 and ,0 are closer to the cosmological time-averages of
m (t) and (t), namely zero and one respectively. In such a picture it might be
easier to believe that we have not arrived on the scene at a special time but merely
a late one. Whether or not this is a helpful way to approach the coincidence
problem is, to a certain extent, a matter of taste.
To summarize the contents of this chapter: we have learned that what can
be seen with our telescopes constitutes no more than 1% of the density of the
Universe. The rest is dark. A small portion (no more than 5%) of this dark
matter is made up of ordinary baryons. Many observational arguments hint at
the existence of a second, more exotic species known as cold dark matter (though
they do not quite establish its existence unless they are combined with ideas about
the formation of large-scale structure). Experiments also imply the existence of
a third dark-matter species, the massive neutrino, although its role appears to
be more limited. Finally, all these components are dwarfed in importance by a
newcomer whose physical origin remains shrouded in obscurity: the energy of
the vacuum.
In the chapters that follow, we will explore the leading contenders for the
dark matter in more depth: vacuum energy, axions, neutrinos, supersymmetric
weakly-interacting massive particles (WIMPs) and objects like black holes.
Our focus will be on what can be learned about each one from its possible
contributions to the extragalactic background light at all wavelengths, from the
radio region to the γ -ray. In the spirit of Olbers’ paradox and what we have done
so far, our main question for each candidate will thus be: Just how dark is it?
References
[1] Kapteyn J C 1922 Astrophys. J. 55 302
[2] Oort J H 1932 Bull. Astron. Inst. Neth. 6 249
[3] Zwicky F 1937 Astrophys. J. 86 217
[5] Freedman W L et al 2001 Astrophys. J. 553 47
[6] Peacock J A 1999 Cosmological Physics (Cambridge: Cambridge University Press)
pp 141–5
[7] Willick J A and Batra P 2001 Astrophys. J. 548 564
[8] Herrnstein J R et al 1999 Nature 400 539
[9] Maoz E et al 1999 Nature 401 351
[10] Guinan E F et al 1998 Astrophys. J. 509 L21
88 The dark matter
[11] Udalski A 2000 Astrophys. J. 531 L25

[12] Romaniello M et al 2000 Astrophys. J. 530 738
[13] Cowan J J et al 1999 Astrophys. J. 521 194
[14] Griffiths L M, Melchiorri A and Silk J 2001 Astrophys. J. 553 L5
[15] Olive K A 2000 Nucl. Phys. Proc. Suppl. 80 79
[16] Tytler D, O’Meara J M, Suzuki N and Lubin D 2000 Phys. Scr. 85 12
[17] Tripp T M, Savage B D and Jenkins E B 2000 Astrophys. J. 534 L1
[18] Miralda-Escudé J et al 1996 Astrophys. J. 471 582
[19] Carr B J and Sakellariadou M 1998 Astrophys. J. 516 195
[20] Schneider P 1993 Astron. Astrophys. 279 1
[21] Alcock C et al 2000 Astrophys. J. 542 281
[22] Ibata R A et al 1999 Astrophys. J. 524 L95
[23] Fields B D, Freese K and Graff D S 2000 Astrophys. J. 534 265
[24] Graff D S and Freese K 1996 Astrophys. J. 456 L49
[25] Hansen B M S 1999 Astrophys. J. 517 L39
[26] Hegyi D J and Olive K A 1986 Astrophys. J. 303 56
[27] Kochanek C S 1996 Astrophys. J. 457 228
[28] Carlberg R G, Yee H K C and Ellingson E 1997 Astrophys. J. 478 462
[29] Carlberg R G et al 1999 Astrophys. J. 516 552
[30] Bahcall N A et al 2000 Astrophys. J. 541 1
[31] Bahcall N A et al P J 1999 Science 284 1481
[32] Guerra E J, Daly R A and Wan L 2000 Astrophys. J. 544 659
[33] Moore B 1994 Nature 370 629
[34] Klypin A A et al 1999 Astrophys. J. 522 82
[35] Padmanabhan T 1993 Structure Formation in the Universe (Cambridge: Cambridge
University Press) pp 63, 135–6
[36] Bahcall N A and Fan X 1998 Astrophys. J. 504 1
[37] Weinberg D H et al 1999 Astrophys. J. 522 563
[38] Peacock J A et al 2001 Proc. 20th Texas Symposium on Relativistic Astrophysics ed
J C Wheeler and H Martel (New York: American Institute of Physics) p 938
[39] Zehavi I and Dekel A 1999 Nature 401 252
[40] Peebles P J E 1993 Principles of Physical Cosmology (Princeton, NJ: Princeton
University Press) pp 158–63
[41] Sciama D W 1993 Modern Cosmology and the Dark Matter Problem (Cambridge:
Cambridge University Press)
[42] Gawiser E and Silk J 1998 Science 280 1405
[43] Croft R A C, Hu W and Davé R 1999 Phys. Rev. Lett. 83 1092
[44] Kolb E W and Turner M S 1990 The Early Universe (Reading, MA: Addison-Wesley)
p 352
[45] Fukuda Y et al 1998 Phys. Rev. Lett. 81 1562
[46] Abdurashitov N et al 1999 Phys. Rev. Lett. 83 4686
[47] Cleveland B T et al 1998 Astrophys. J. 496 505
[48] Hampel W et al 1999 Phys. Lett. B 447 127
[49] Athanassopoulos C et al 1998 Phys. Rev. Lett. 81 1774
[50] Klapdor-Kleingrothaus H V, Päs H and Smirnov A Y 2001 Phys. Rev. D 63 073005
[51] Rindler W 1977 Essential Relativity (Berlin: Springer) p 226
[52] Priester W, Hoell J and Blome H-J 1995 Comments Astrophys. 17 327
[53] Gamow G 1970 My World Line (New York: Viking Press) p 44
[54] Carroll S M 2001 Living Reviews in Relativity 4

http://www.livingreviews.org/Articles/Volume4/2001-1carroll
[55] Puthoff H E 1989 Phys. Rev. A 40 4857
[56] Wesson P S 1991 Astrophys. J. 378 466
[57] Overduin J M and Cooperstock F I 1998 Phys. Rev. D 58 043506
[58] Wesson P S and Liu H 2001 Int. J. Mod. Phys. D 10 905
[59] Weinberg S 2001 Sources and Detection of Dark Matter in the Universe ed D B Cline
(Berlin: Springer) p 18
[60] Blome H-J and Priester W 1985 Astrophys. Sp. Sci. 117 327
[61] Gott J R III, Park M-G and Lee H M 1989 Astrophys. J. 338 1
[62] Franx M et al 1997 Astrophys. J. 486 L75
[64] Malhotra S, Rhoads J E and Turner E L 1997 Mon. Not. R. Astron. Soc. 288 138
[65] Falco E E, Kochanek C S and Muñoz J A 1998 Astrophys. J. 494 47
[66] Cooray A R 1999 Astron. Astrophys. 342 353
[67] Fukugita M et al 1990 Astrophys. J. 361 L1
[68] Gardner J P, Cowie L L and Wainscoat R J 1993 Astrophys. J. 415 L9
[69] Totani T, Yoshii Y and Sato K 1997 Astrophys. J. 483 L75
[70] Totani T and Yoshii Y 2000 Astrophys. J. 540 81
[71] Jenkins A et al 1998 Astrophys. J. 499 20
[72] Feldman H A and Evrard A E 1993 Int. J. Mod. Phys. D 2 113
[73] White M and Scott D 1996 Astrophys. J. 459 415
[74] Riess A G et al 1998 Astron. J. 116 1009
[75] Perlmutter S et al 1999 Astrophys. J. 517 565
[76] Aguirre A and Haiman Z 2000 Astrophys. J. 532 28
[77] Drell P S, Loredo T J and Wasserman I 2000 Astrophys. J. 530 593
[78] Hillebrandt W and Niemeyer J C 2000 Ann. Rev. Astron. Astrophys. 38 191
[79] Lineweaver C H 2001 Gravitational Lensing: Recent Progress and Future Goals
(Astronomical Society of the Pacific Conference Series, Volume 237) ed
T G Brainerd and C S Kochanek (San Francisco, CA: ASP)
[80] de Bernardis P et al 2000 Nature 404 955
[81] Hanany S et al 2000 Astrophys. J. 545 L5
[82] Pryke C et al 2001 Astrophys. J. 568 46
[83] Bennett C et al 1996 Astrophys. J. 464 L1
[84] Jaffe A H et al 2000 Phys. Rev. Lett. 86 3475
[85] White M, Scott D and Pierpaoli E 2000 Astrophys. J. 545 1
[86] McGaugh S S 2000 Astrophys. J. 541 L33
[87] Liebscher D-E, Priester W and Hoell J 1992 Astron. Astrophys. 261 377
[88] Overduin J M and Priester W 2001 Naturwissenschaften 88 229
Chapter 5
The vacuum
5.1 Vacuum decay

It may be surprising to find the vacuum, traditionally defined as the ‘absence
of anything’, playing a large and possibly dominant role in the dynamics of the
Universe. But quantum mechanics has taught us to regard the vacuum as the
ground state of all possible fields, and ground states in quantum field theory can
have very high energies indeed. Of course, it is the astonishing gap between
these theoretical energy densities and those which now seem to be observed
in cosmology which constitutes the cosmological-constant problem, perhaps the
single deepest unsolved puzzle in theoretical physics today.
Many authors have sought to address the crisis by finding a mechanism by
which the energy density ρv of the vacuum could decay with time. This would
be equivalent to a variable cosmological term since c2 = 8π Gρv from (2.34).
With such a mechanism in hand, the problem would be reduced to explaining
why the Universe is of intermediate age: old enough that has relaxed from
primordial values like those suggested by quantum field theory to the values which
we measure now, but young enough that ≡ ρv /ρcrit has not yet reached its
asymptotic value of unity.
Energy conservation requires that any decrease in the energy density of
the vacuum be made up by a corresponding increase somewhere else. In some
scenarios, vacuum energy goes into the kinetic energy of new forms of matter such
as scalar fields, which have yet to be observed in nature. In others it is channelled
instead into baryons, photons or neutrinos. Baryonic decays would produce equal
amounts of matter and antimatter, whose subsequent annihilation would flood
the Universe with γ -rays. Radiative decays would similarly pump photons into
intergalactic space, but are harder to constrain because they could, in principle,
involve any part of the electromagnetic spectrum. As we will see, however, robust
limits can be set on any such process under conservative assumptions.
We proceed in the remainder of the chapter to describe how cosmology
is modified in the presence of a decaying cosmological ‘constant’ and then
90
The variable cosmological ‘constant’ 91
specialize to the case of vacuum decay into radiation. Our objective is to assess
the impact of such a process on both the bolometric and spectral EBL intensity,
using the same formalism that was applied in chapters 2 and 3 to the light from
ordinary galaxies.
5.2 The variable cosmological ‘constant’

Einstein originally introduced the -term in 1917 as a constant of nature akin
to c or G, and it is worth beginning by asking how such a quantity can be
allowed to vary. Suppose that we take the covariant divergence of Einstein’s field
equations (2.1) and apply the Bianchi identities. The latter are a mathematical
Ê Ê
statement about Riemannian geometry and read ∇ ν ( µν − 12 gµν ) = 0. With
these conditions, and the fact that the metric tensor itself has a vanishing covariant
derivative (∇ ν gµν = 0), we find
∂µ =
8π G ν
c4
∇ Ìµν . (5.1)
Ì
Assuming that matter and energy (as contained in µν ) are conserved, it follows
that ∂µ = 0 and, hence, that = constant.
In variable- theories, one must therefore do one of three things: abandon
matter–energy conservation, modify general relativity or stretch the definition of
what is conserved. The first of these routes was explored in 1933 by Bronstein [1],
who sought to connect energy non-conservation with the observed fact that the
Universe is expanding rather than contracting. (Einstein’s equations actually
allow both types of solutions, and a convincing explanation for this cosmological
arrow of time has yet to be given.) Bronstein was executed in Stalin’s Soviet
Union a few years later and his work is not widely known [2].
Today, few physicists would be willing to sacrifice energy conservation
outright. Some, however, would be willing to modify general relativity, or to
consider new forms of matter and energy. Historically, these two approaches have
sometimes been seen as distinct, with one being a change to the ‘geometry of
nature’ while the other is concerned with the material content of the Universe.
The modern tendency, however, is to regard them as equivalent. This viewpoint
is best personified by Einstein, who in 1936 compared the left-hand (geometrical)
and right-hand (matter) sides of his field equations to ‘fine marble’ and ‘low-
grade wooden’ wings of the same house [3]. In a more complete theory, he
argued, matter fields of all kinds would be seen to be just as geometrical as the
gravitational one.
Let us see how this works in one of the oldest and simplest variable-
theories: a modification of general relativity in which the metric tensor gµν is
supplemented by a scalar field ϕ whose coupling to matter is determined by a
parameter ω. The idea for such a theory goes back to Jordan in 1949 [4], Fierz
in 1956 [5] and Brans and Dicke in 1961 [6]. In those days, new scalar fields
92 The vacuum
were not introduced into theoretical physics as routinely as they are today, and
all these authors sought to associate ϕ with a known quantity. Various lines of
argument (notably Mach’s principle) pointed to an identification with Newton’s
gravitational ‘constant’ such that G ∼ 1/ϕ. By 1968 it was appreciated that
and ω too would depend on ϕ in general [7]. The original Brans–Dicke theory
(with = 0) has thus been extended to generalized scalar–tensor theories in
which = (ϕ) [8], = (ϕ), ω = ω(ϕ) [9] and = (ϕ, ψ), ω = ω(ϕ)
where ψ ≡ ∂ µ ϕ∂µ ϕ [10]. Let us consider the last and most general of these cases,
for which the field equations are found to read:

ω(ϕ)
Ê µν −
1
2
Ê gµν + [∇µ (∂ν ϕ) − ϕgµν ] + 2
1
ϕ ϕ
1
∂µ ϕ∂ν ϕ − ψgµν
2
∂(ϕ, ψ)
− (ϕ, ψ)gµν + 2
∂ψ
8π
∂µ ϕ∂ν ϕ = − 4 µν
ϕc
Ì (5.2)
where ϕ ≡ ∇ µ (∂µ ϕ) is the D’Alembertian. These reduce to Einstein’s

equations (2.1) when ϕ = constant = 1/G.
If we now repeat the exercise on the previous page and take the covariant
derivative of the field equations (5.2) with the Bianchi identities, we obtain a
generalized version of the equation (5.1) faced by Bronstein:

∂µ ϕ
Ê +
ω(ϕ)
ψ−
ω(ϕ)
ϕ + (ϕ, ψ) + ϕ ∂(ϕ, ψ)
−
ψ dω(ϕ)
2 2ϕ 2 ϕ ∂ϕ 2ϕ dϕ

∂(ϕ, ψ) ∂(ϕ, ψ)
− 2ϕ ϕ
∂ψ
κ
− 2∂ ϕ∂κ ϕ
∂ψ
8π
c
Ì
= 4 ∇ ν µν . (5.3)
Ì
Now energy conservation (∇ ν µν = 0) no longer requires = constant. In fact,
it is generally incompatible with constant , unless an extra condition is imposed
on the terms inside the curly brackets in (5.3). (This cannot come from the wave
equation for ϕ, which merely confirms that the terms inside the curly brackets sum
to zero, in agreement with energy conservation.) Similar conclusions hold for
other scalar–tensor theories in which ϕ is no longer associated with G. Examples
include models with non-minimal couplings between ϕ and the curvature scalar
Ê [11], conformal rescalings of the metric tensor by functions of ϕ [12] and non-
zero potentials V (ϕ) for the scalar field [13–15]. (Theories of this last kind are
now known as quintessence scenarios [16].) For all such cases, the cosmological
‘constant’ becomes a dynamical variable.
In the modern approach to variable- cosmology, which goes back to
Zeldovich in 1968 [17], all extra terms of the kind just described—including
—are moved to the right-hand side of the field equations (5.2), leaving only
Ê Ê
the Einstein tensor ( µν − 12 gµν ) to make up the ‘geometrical’ left-hand side.
The cosmological term, along with scalar (or other) additional fields, are thus
effectively reinterpreted as new kinds of matter. The field equations (5.2) may
then be written
Ê µν −
1
2
Ê gµν = − 4 µν
ϕc
Ì
8π eff
+ (ϕ)gµν . (5.4)
The variable cosmological ‘constant’ 93
Here Ìµνeff is an effective energy–momentum tensor describing the sum of
ordinary matter plus whatever scalar (or other) fields have been added to the
theory. For generalized scalar–tensor theories as described above, this could be
ϕ ϕ
written as Ìµν
eff ≡ Ì
µν + Ìµν where Ìµν refers to ordinary matter and Ìµν to the
scalar field. For the case with = (ϕ) and ω = ω(ϕ), for instance, the latter
would be defined by (5.2) as

ω(ϕ)
Ìµν ≡ [∇µ (∂ν ϕ) − ϕgµν ] + 2 ∂µ ϕ∂ν ϕ − ψgµν .
ϕ 1 1
(5.5)
ϕ ϕ 2
The covariant derivative of the field equations (5.4) with the Bianchi identities
now gives
ν 8π
0=∇ Ì − (ϕ)gµν .
eff
(5.6)
ϕc4 µν
Equation (5.6) carries the same physical content as (5.3), but is more general in
form and can readily be extended to other theories. Physically, it says that energy
is conserved in variable- cosmology—where ‘energy’ is now understood to refer
to the energy of ordinary matter along with that in any additional fields which may
be present, and along with that in the vacuum, as represented by . In general,
the latter parameter can vary as it likes, so long as the conservation equation (5.6)
is satisfied.
It is natural to wonder whether the evolution of in these theories actually
helps with the cosmological ‘constant’ problem, in the sense that drops
from large primordial values to ones like those seen today without fine-tuning.
Behaviour of this kind was noted at least as early as 1977 [8] in the context of
models with = (ϕ) and ω = constant, which have solutions for ϕ(t) such that
∝ t −2 . In precursors to the modern quintessence scenarios, Barr [13] found
models in which ∝ t − at late times, while Peebles and Ratra [14] discussed a
theory in which ∝ R −m at early ones (here and m are powers). There is now
a rich literature on -decay laws of this kind (see [18] for a review). Their appeal
is easy to understand and can be illustrated with a simple dimensional argument
for the case in which ∝ R −2 [19]. Since already has dimensions of L −2 , the
proportionality factor in this case is a pure number (α, say) which is presumably
of order unity. Taking α ∼ 1 and identifying R with a suitable length scale in
cosmology (namely the Hubble distance c/H0), one finds that 0 ∼ H02/c2 . The
present vacuum density parameter is then ,0 ≡ 0 c2 /3H02 ∼ 1/3, close to the
values implied by current supernovae data (section 4.6). A natural choice for the
primordial value of , moreover, would have R ∼ Pl so that Pl ∼ α−2 Pl . This
leads to a ratio Pl /0 ∼ (c/H0Pl )2 ∼ 10122, which may be compared with the
values in table 4.1.
While this would seem to be a promising approach, two cautions must be
kept in mind. The first is theoretical. Insofar as the mechanisms discussed so
far are entirely classical, they do not address the underlying problem. For this,
one would also need to explain why net contributions to from the quantum
94 The vacuum
vacuum do not remain at the primordial level or how they are suppressed with
time. Polyakov [20] and Adler [21] in 1982 were the first to speculate explicitly
that such a suppression might come about if the ‘bare’ cosmological term implied
by quantum field theory were progressively screened by an ‘induced’ counterterm
of opposite sign, driving the effective value of (t) toward zero at late times.
Many theoretical adjustment mechanisms have now been identified as potential
sources of such a screening effect, beginning with a 1983 suggestion by Dolgov
[22] based on non-minimally-coupled scalar fields. Subsequent proposals have
involved scalar fields [23–25], fields of higher spin [26–28], quantum effects
during inflation [29–31] and other phenomena [32–34]. In most of these cases,
no analytic expression is found for in terms of time or other cosmological
parameters; the intent is merely to demonstrate that decay (and preferably near-
cancellation) of the cosmological term is possible in principle. None of these
mechanisms has been widely accepted as successful to date. In fact, there is a
general argument due to Weinberg to the effect that a successful mechanism based
on scalar fields would necessarily be so finely tuned as to be just as mysterious
as the original problem [35]. Similar concerns have been raised in the case of
vector and tensor-based proposals [36]. Nevertheless, the idea of the adjustment
mechanism remains feasible in principle and continues to attract more attention
than any other approach to the cosmological-constant problem.
The second caution is empirical. Observational data place increasingly
strong restrictions on the way in which can vary with time. Among the
most important are early-time bounds on the vacuum energy density ρ c2 =
c4 /8π G. The success of standard primordial nucleosynthesis theory implies
that ρ was smaller than ρr and ρm during the radiation-dominated era, and large-
scale structure formation could not have proceeded in the conventional way unless
ρ < ρm during the early matter-dominated era. Since ρr ∝ R −4 and ρm ∝ R −3
from (2.32), these requirements mean in practice that the vacuum energy density
must climb less steeply than R −3 in the past direction, if it is comparable to that
of matter or radiation at present [37, 38]. The variable- term must also satisfy
late-time bounds like those which have been placed on the cosmological constant
(section 4.6). Tests of this kind have been carried out using data on the age of
the Universe [39, 40], structure formation [41–43], galaxy number counts [44],
the CMB power spectrum [45, 46], gravitational lensing statistics [46–48] and
Type Ia supernovae [46, 49]. Some of these tests are less restrictive in the case
of a variable -term than they are for = constant, and this can open up new
regions of parameter space. Observation may even be compatible with some non-
singular models whose expansion originates in a hot, dense ‘big bounce’ rather
than a big bang [50], a possibility which can be ruled out on quite general grounds
if = constant.
A third group of limits comes from asking what the vacuum decays into.
In quintessence theories, vacuum energy is transferred to the kinetic energy of
a scalar field as it ‘rolls’ down a gradient toward the minimum of its potential.
This may have observable consequences if the scalar field is coupled strongly to
Energy density 95
ordinary matter, but is hard to constrain in general. A simpler situation is that

in which the vacuum decays into known particles such as baryons, photons or
neutrinos. The baryonic decay channel would produce excessive levels of γ -
ray background radiation due to matter–antimatter annihilation unless the energy
density of the vacuum component is less than 3 × 10−5 times that of matter [37].
This limit can be weakened if the decay process violates baryon number or if it
takes place in such a way that matter and antimatter are segregated on large scales,
but such conditions are hard to arrange in a natural way. The radiative decay
channel is more promising, but also faces a number of tests. The decay process
should meet certain criteria of thermodynamic stability [51] and adiabaticity [52].
The shape of the spectrum of decay photons offers another possibility. If this
differs from that of pre-existing background radiation, then distortions will arise.
Freese et al have argued on this basis that a vacuum decaying primarily into
low-energy photons could have a density no more than 4 × 10−4 times that of
radiation [37].
It may be, however, that vacuum-decay photons blend into the spectrum of
background radiation without distorting it. Figure 2.1 shows that the best place
to ‘hide’ large quantities of excess energy would be the microwave and infrared
regions where the energy density of background radiation is highest. Could
part of the CMB originate in a decaying vacuum? We know from the COBE
satellite that its spectrum is very nearly that of a perfect blackbody [53]. Freese
et al pointed out that vacuum-decay photons would be thermalized efficiently
by brehmsstrahlung and double-Compton scattering in the early Universe, and
might continue to assume a blackbody spectrum at later times if pre-existing
CMB photons played a role in ‘inducing’ the vacuum to decay [37]. Subsequent
work has shown that this would require a special combination of thermodynamical
parameters [54]. Such a possibility is important in practice, however, because it
leads to the most conservative limits on the theory. If the radiation produced by
vacuum decay does not distort the background, it will in any case contribute to
the latter’s absolute intensity. We can calculate the size of these contributions to
the background radiation using the methods that have been laid out in chapters 2
and 3.
5.3 Energy density

The first step in this problem is to solve the field equations and conservation
equations for the energy density of the decaying vacuum. We will do this in
the context of a general phenomenological model. This means that we retain the
field equations (5.4) and the conservation law (5.6) without specifying the form of
the effective energy–momentum tensor in terms of scalar (or other) fields. These
equations may be written
Êµν − 21 Êgµν = − 8πc4G (Ìµνeff − ρ c2 gµν ) (5.7)

96 The vacuum
0 = ∇ ν (Ìµν
eff
− ρ c2 gµν ). (5.8)
Here ρ c2 ≡ c4 /8π G from (2.34) and we have put back G in place of 1/ϕ.
Equations (5.7) and (5.8) have the same form as their counterparts (2.1) and
(2.29) in standard cosmology, the key difference being that the cosmological term
has migrated to the right-hand side and is no longer necessarily constant. Its
dynamical evolution is now governed by the conservation equations (5.8), which
require only that any change in ρ c2 gµν be offset by an equal and opposite change
in the energy–momentum tensor Ìµν eff .
While the latter is model-dependent in general, it is reasonable to assume

in the context of isotropic and homogeneous cosmology that its form is that of a
perfect fluid, as given by (2.26):
Ìµν
eff
= (ρeff + peff /c2 )Uµ Uν + peff gµν . (5.9)
Comparison of equations (5.8) and (5.9) shows that the conserved quantity in
(5.8) must then also have the form of a perfect-fluid energy–momentum tensor,
with density and pressure given by
ρ = ρeff + ρ p = peff − ρ c2 . (5.10)
The conservation law (5.8) may then be simplified at once by analogy with
equation (2.29):
1 d 3 d
3
[R (ρeff c2 + peff )] = ( peff − ρ c2 ). (5.11)
R dt dt
This reduces to the standard result (2.30) for the case of a constant cosmological
term, ρ = constant.
In this chapter, we will allow the cosmological term to contain both a
constant part and a time-varying part so that
ρ = ρc + ρv (t) ρc = constant. (5.12)
Let us assume, in addition, that the perfect fluid described by Ìµν eff consists of a
mixture of dustlike matter ( pm = 0) and radiation ( pr = 3 ρr c ):

1 2
ρeff = ρm + ρr peff = 13 ρr c2 . (5.13)
The conservation equation (5.11) then reduces to
1 d 4 1 d dρv
(R ρr ) + 3 (R 3 ρm ) + = 0. (5.14)
R 4 dt R dt dt
From this equation it is clear that one (or both) of the radiation and matter
densities can no longer obey the usual relations ρr ∝ R −4 and ρm ∝ R −3 in
Energy density 97
a theory with = constant. Any change in (or ρ ) must be accompanied by a

change in radiation and/or matter densities.
To go further, some simplifying assumptions must be made. Let us take to
begin with
d 3
(R ρm ) = 0. (5.15)
dt
This is equivalent to imposing conservation of particle number, as may be seen
by replacing ‘galaxies’ with ‘particles’ in equation (2.7). Such an assumption
is well justified during the matter-dominated era by the stringent constraints on
matter creation discussed in section 5.2. It is equally well justified during the
radiation-dominated era, when the matter density is small so that the ρm term is
of secondary importance compared to the other terms in (5.14) in any case.
In light of equations (5.14) and (5.15), the vacuum can exchange energy only
with radiation. As a model for this process, let us follow Pollock in 1980 [55] and
assume that it takes place in such a way that the energy density of the decaying
vacuum component remains proportional to that of radiation, ρv ∝ ρr . We adopt
the notation of Freese et al in 1987 and write the proportionality factor as x/(1−x)
with x the coupling parameter of the theory [37]. If this is allowed to take
(possibly different) constant values during the radiation- and matter-dominated
eras, then
ρv x r (t < teq )
x≡ = (5.16)
ρr + ρv x m (t teq ).
Here teq refers to the epoch of matter–radiation equality radiation when ρr =
ρm . Standard cosmology is recovered in the limits x r → 0 and x m → 0. The most
natural situation is that in which the value of x stays constant so that x r = x m .
However, since observational constraints on x are, in general, different for the
radiation- and matter-dominated eras, the most conservative limits on the theory
are obtained by letting x r and x m take different values. Physically, this would
correspond to a phase transition or sudden change in the expansion rate Ṙ/R of
the Universe at t = teq .
With the assumptions (5.15) and (5.16), the conservation equation (5.14)
reduces to
ρ˙v Ṙ
+ 4(1 − x) = 0 (5.17)
ρv R
where overdots denote derivatives with respect to time. This may be integrated to
give
ρv (R) = αv R −4(1−x) (5.18)
where αv = constant. The cosmological term is thus an inverse power-
law function of the scale factor R. This is a scenario that has been widely
studied, also in cases where vacuum energy density is not proportional to that
of radiation [18]. Equation (5.18) shows that the conserved quantity in this theory
has a form intermediate between that of ordinary radiation entropy (R 4 ρr ) and
particle number (R 3 ρm ) when 0 < x < 14 .
98 The vacuum
The fact that ρr ∝ ρv ∝ R −4(1−x) places an immediate upper limit of 14

on x (in both eras), since higher values would erase the dynamical distinction
1
between radiation and matter. With x 4 it then follows from (5.16) that
ρv 1
ρ
3 r . This is consistent with section 5.2 where we noted that a vacuum
energy component which climbs more steeply than R −3 in the past direction
cannot have an energy density greater than that of radiation at present. Freese
et al [37] set a stronger bound by showing that x 0.07 if the baryon-to-
photon ratio η is to be consistent with both primordial nucleosynthesis during
the radiation-dominated era and CMB observations at present. (This argument
assumes that x = x r = x m .) As a guideline in what follows, then, we will allow
x r and x m to take values between zero and 0.07 and consider, in addition, the
theoretical possibility that x m could increase to 0.25 in the matter-dominated era.
With ρm (R) specified by (5.15), ρr related to ρv by (5.16) and ρv (R) given
by (5.18), we can solve for all three components as functions of time if the scale
factor R(t) is known. This comes as usual from the field equations (5.7). Since
these are the same as equations (2.1) for standard cosmology, they lead to the
same result, equation (2.33):
2
Ṙ 8π G
= (ρm + ρr + ρv + ρc ). (5.19)
R 3
Here we have used equations (5.12) to replace ρ with ρv + ρc and (5.13) to
replace ρeff with ρm + ρr . We have also set k = 0 since observations indicate that
these components together make up very nearly the critical density (section 4.6).
Equation (5.19) can be solved analytically for the three cases which are of
greatest physical interest. We define these as follows:
(1) the radiation-dominated regime (t < teq , ρr + ρv ρm + ρc );
(2) the matter-dominated regime (t teq , ρr + ρv ρm , ρc = 0); and
(3) the vacuum-dominated regime (t teq , ρr + ρv ρm + ρc ).
The distinction between regimes 2 and 3 is important because we will find in
practice that the former describes models which are close to the EdS one, while
the latter is needed to model vacuum-dominated cosmologies like CDM or
BDM (table 3.1). The definitions of these terms should be amended slightly
for this chapter, since we now consider flat models containing not only matter
and a cosmological constant, but radiation and a decaying-vacuum component as
well. The densities of the latter two components, however, are at least four orders
of magnitude below that of matter at present. Thus models with ρc = 0, for
example, have m,0 = 1 to four-figure precision or better and are dynamically
indistinguishable from EdS during all but the first fraction (of order 10−4 or less)
of their lifetimes. For definiteness, we will use the terms ‘EdS’, ‘CDM’ and
‘BDM’ in this chapter to refer to flat models in which m,0 = 1, 0.3 and
0.03 respectively, with the remainder (if any) of the present critical density being
effectively made up by the constant-density component of the vacuum energy.
Energy density 99
It remains to solve equations (5.15), (5.16), (5.18) and (5.19) for the four
dynamical quantities R, ρm , ρr and ρv in terms of the constants ρm,0 , ρr,0 , x r and
x m . We relegate the details of this exercise to appendix B and summarize the
results here. The scale factor is found as



 t 1/2(1−xr)
 (t < teq )
t
R(t) ∝ 0 2/3 (5.20)


 Ëm (t)
 (t teq ) .
Ëm (t0 )
The vacuum density is given by
 αx r

 t −2 (t < teq )
(1 − x r )2

ρv (t) = xm (5.21)

 ρr (t) (t teq )
1 − xm
where α = 3/(32π G) = 4.47 × 105 g cm−2 s2 . For the density of radiation we

find 
 1 − xr

 ρv (t) (t < teq )
x
ρr (t) = r −8(1−xm)/3 (5.22)


 ρr,0
Ëm (t)
(t teq ) .
Ëm (t0 )
The matter density can be written
 −2 −3/2(1−xr)



Ë m (teq ) t
 (t < teq )
Ë (t )
ρm (t) = ρm,0 × m 0 −2
teq
(5.23)


 Ë m (t)
 (t teq ) .
Ëm (t0 )
Here we have applied ρm,0 = m,0 ρcrit,0 and ρr,0 = r,0 ρcrit,0 as boundary
conditions, with the values of m,0 and r,0 to be specified and ρcrit,0 given by
(2.36). The function Ëm (t) is defined as

t (m,0 = 1)
Ëm (t) ≡ sinh(t/τ ) (0 < m,0 < 1)
(5.24)
0

where τ0 ≡ 2/(3H0 1 − m,0 ) and H0 is given by (2.19). The age of the
Universe is
2/(3H0) (m,0 = 1)
t0 = (5.25)
τ0 sinh−1 χ0 (0 < m,0 < 1)

where χ0 ≡ (1 − m,0 )/m,0 . The second of these expressions comes
from (2.70). Corrections from the radiation-dominated era can be ignored since
t0 teq in all cases.
100 The vacuum
(a) (b)
−4
xr = xm = 10 : matter EdS: matter
1e−09 radiation 1e−09 radiation
vacuum vacuum
1e−12 xr = xm = 0.07: matter 1e−12 ΛCDM: matter
Densities ρv, ρr, ρm ( g cm−3 )
Densities ρv, ρr, ρm ( g cm−3 )

radiation radiation
1e−15 vacuum 1e−15 vacuum
ΛBDM: matter
1e−18 1e−18 radiation
vacuum
1e−21 1e−21
1e−24 1e−24
1e−27 1e−27
1e−30 1e−30
1e−33 1e−33
1e+08 1e+10 1e+12 1e+14 1e+16 1e+18 1e+08 1e+10 1e+12 1e+14 1e+16 1e+18
Time t ( s ) Time t ( s )
Figure 5.1. The densities of decaying vacuum energy (ρv ), radiation (ρr ) and matter (ρm )
as functions of time. The left-hand panel (a) shows the effects of changing the values of xr
and xm , assuming a model with m,0 = 0.3 (similar to CDM). The right-hand panel (b)
shows the effects of changing the cosmological model, assuming xr = xm = 10−4 . The
vertical lines indicate the epochs when the densities of matter and radiation are equal (teq ).
All curves assume h 0 = 0.75.
The parameter teq is obtained as in standard cosmology by setting ρr (teq ) =

ρm (teq ) in equations (5.22) and (5.23). This leads to
 3/2(1−4x m)
 t0 r,0


(m,0 = 1)
3/2(1−4xm)
teq = −1 r,0 (5.26)

 τ0 sinh χ0 (0 < m,0 < 1) .
m,0
The epoch of matter–radiation equality plays a crucial role in this chapter because
it is at about this time that the Universe became transparent to radiation (the two
events are not simultaneous but the difference between them is minor for our
purposes). Decay photons created before teq would simply have been thermalized
by the primordial plasma and eventually re-emitted as part of the CMB. It is the
decay photons emitted after this time which can contribute to the EBL, and whose
contributions we wish to calculate. The quantity teq is thus analogous to the galaxy
formation time tf in previous chapters.
The densities ρm (t), ρr (t) and ρv (t) are plotted as functions of time in
figure 5.1. The left-hand panel (a) shows the effects of varying the parameters
x r and x m within a given cosmological model (here, CDM). Raising the value
of x m (i.e. moving from bold to light curves) leads to a proportionate increase
in ρv and a modest drop in ρr . It also flattens the slope of both components.
This change in slope (relative to that of the matter component) pushes the epoch
Source regions and luminosity 101
of equality back toward the big bang (vertical lines). Such an effect could,
in principle, allow more time for structure to form during the early matter-
dominated era [37], although the ‘compression’ of the radiation-dominated era
rapidly becomes unrealistic for values of x m close to 14 . Thus figure 5.1(a) shows
that the value of teq is reduced by a factor of over 100 in going from a model with
x m = 10−4 to one with x m = 0.07. In the limit x m → 14 , the duration of the
radiation-dominated era in fact dwindles to nothing, as remarked earlier and as
shown explicitly by equations (5.26).
Figure 5.1(b) shows the effects of changes in cosmological model for fixed
values of x r and x m (here both set to 10−4 ). Moving from the matter-filled EdS
model toward vacuum-dominated ones such as CDM and BDM does three
things. The first is to increase the age (t0 ) of the Universe. This increases the
density of radiation at any given time, since the latter is fixed at present and climbs
at the same rate in the past direction. Based on our experience with the galactic
EBL in previous chapters, we may expect that this should lead to significantly
higher levels of background radiation when integrated over time. However, there
is a second effect in the present theory which acts in the opposite direction:
smaller values of m,0 imply higher values of teq as well as t0 , thus delaying
the onset of the matter-dominated era (vertical lines). As we will see, these
two changes all but cancel each other out as far as vacuum decay contributions
to the background are concerned. The third important consequence of vacuum-
dominated cosmologies is ‘late-time inflation’, the sharp increase in the expansion
rate at recent times (figure 4.2). This translates in figure 5.1(b) into a noticeable
‘droop’ in the densities of matter and radiation at the right-hand edge of the figure
for the BDM model in particular.
5.4 Source regions and luminosity

In order to proceed with the formalism we have developed in chapters 2 and 3,
we need to define discrete ‘sources’ of vacuum-decay radiation, analogous to the
galaxies of previous chapters. For this purpose we carve up the Universe into
hypothetical regions of arbitrary comoving volume V0 . The comoving number
density of these source regions is just
n(t) = n 0 = V0−1 = constant. (5.27)
These regions are introduced for convenience, and are not physically significant
since the vacuum decays uniformly throughout space. We therefore expect that
the parameter V0 will not appear in our final results.
The next step is to identify the ‘source luminosity’. There are at least two
ways to approach this question [56]. One could simply regard the source region as
a ball of physical volume V (t) = R̃ 3 (t)V0 filled with fluctuating vacuum energy.
As the density of this energy drops by −dρv during time dt, the ball loses energy
at a rate −dρv /dt. If some fraction β of this energy flux goes into photons, then
102 The vacuum
the luminosity of the ball is

L v (t) = −βc2 ρ˙v (t)V (t). (5.28)
This is the definition of vacuum luminosity which has been assumed implicitly
by several workers including Pavón [51], who investigated the thermodynamical
stability of the vacuum decay process by requiring that fluctuations in ρ˙v not grow
larger than the mean value of ρ˙v with time. For convenience we will refer to (5.28)
as the thermodynamical definition of vacuum luminosity (L th ).
A second approach is to treat this as a problem involving spherical symmetry
within general relativity. The assumption of spherical symmetry allows the
total mass–energy (Mc2 ) of a localized region of perfect fluid to be identified
unambiguously. Luminosity can then be related to the time rate of change of this
mass–energy. Assuming once again that the two are related by a factor β, one has
L v (t) = β Ṁ(t)c2 . (5.29)
Application of Einstein’s field equations leads to the following expression [57]
for the rate of change of mass–energy in terms of the pressure pv at the region’s
surface:
Ṁ(t)c2 = −4π pv(t)[r (t)]2ṙ (t) (5.30)
where r (t) = R̃(t)r0 is the region’s physical radius. Taking V = 43 πr 3 , applying
the vacuum equation of state pv = −ρv c2 and substituting (5.30) into (5.29), we
find that the latter can be written in the form
L v (t) = βc2ρv (t)V̇ (t). (5.31)
This is just as appealing dimensionally as equation (5.28), and shifts the emphasis
physically from fluctuations in the material content of the source region toward
changes in its geometry. We will refer to (5.31) for convenience as the relativistic
definition of vacuum luminosity (L rel ).
It is not obvious which of the two definitions (5.28) and (5.31) more correctly
describes vacuum luminosity; this is a conceptual issue. Before choosing between
them, let us inquire whether the two expressions might not be equivalent. We can
do this by taking the ratio
L th ρ˙v V 1 ρ˙v R
=− =− . (5.32)
L rel ρv V̇ 3 ρv Ṙ
Differentiating equations (5.20) and (5.21) with respect to time, we find
 2

 (m,0 = 1)
Ṙ 3t
= 2 t (5.33)
R 
 coth (0 < m,0 < 1)
3τ0 τ0
 −8

 (1 − x m ) (m,0 = 1)
ρ˙v 3t
= −8 t (5.34)
ρv 
 (1 − x m ) coth (0 < m,0 < 1).
3τ0 τ0
Source regions and luminosity 103
The ratio of L th to L rel is therefore a constant:
L th 4
= (1 − x m ). (5.35)
L rel 3
This takes numerical values between 43 (in the limit x m → 0 where standard
cosmology is recovered) and 1 (in the opposite limit where x m takes its maximum
theoretical value of 14 ). There is thus little difference between the two scenarios
in practice, at least where this model of vacuum decay is concerned. We will
proceed using the relativistic definition (5.31) which gives lower intensities and
hence more conservative limits on the theory. At the end of the chapter it will
be a small matter to obtain the corresponding intensity for the thermodynamical
case (5.28) by multiplying through by 43 (1 − x m ).
We now turn to the question of the branching ratio β, or fraction of vacuum
decay energy which goes into photons as opposed to other forms of radiation such
as massless neutrinos. This is model-dependent in general. If the vacuum-decay
radiation reaches equilibrium with that already present, however, then we may
reasonably set this equal to the ratio of photon-to-total radiation energy densities
in the CMB:
β = γ /r,0 . (5.36)
The density parameter γ of CMB photons is given in terms of their blackbody
temperature T by Stefan’s law. Using the COBE value Tcmb = 2.728 K [53], we
get
4σSB T 4
γ = 3 = 2.48 × 10−5 h −2
0 . (5.37)
c ρcrit,0
The total radiation density r,0 = γ + ν is harder to determine, since there
is little prospect of detecting the neutrino component directly. What is done in
standard cosmology is to calculate the size of neutrino contributions to r,0 under
the assumption of entropy conservation. With three massless neutrino species,
this leads to
7 Tν 4
r,0 = γ 1 + 3 × (5.38)
8 T
where Tν is the blackbody temperature of the relic neutrinos and the factor
of 7/8 arises from the fact that these particles obey Fermi rather than Bose–
Einstein statistics [58]. During the early stages of the radiation-dominated era,
neutrinos were in thermal equilibrium with photons so that Tν = T . They
dropped out of equilibrium, however, when the temperature of the expanding
fireball dropped below about kT ∼ 1 MeV (the energy scale of weak interactions).
Shortly thereafter, when the temperature dropped to kT ∼ m e c2 = 0.5 MeV,
electrons and positrons began to annihilate, transferring their entropy to the
remaining photons in the plasma. This raised the photon temperature by a factor
of (1 +2 × 78 = 114 )
1/3 relative to that of the neutrinos. In standard cosmology, the
104 The vacuum
ratio of Tν /T has remained at (4/11)1/3 down to the present day, so that (5.38)
gives
r,0 = 1.68γ = 4.17 × 10−5 h −2
0 . (5.39)
Using (5.36) for β, this would imply:
β = 1/1.68 = 0.595. (5.40)
We will take these as our ‘standard values’ of r,0 and β in what follows. They
are conservative ones, in the sense that most alternative lines of argument would
imply higher values of β. Birkel and Sarkar [38], for instance, have argued that
vacuum decay into radiation with x r = constant would be easier to reconcile
with processes such as electron–positron annihilation if the vacuum coupled
to photons but not neutrinos. This would complicate the theory, breaking the
radiation density ρr in (5.14) into a photon part ργ and a neutrino part with
different dependencies on R. One need not solve this equation, however, in
order to appreciate the main impact of such a modification. Decay into photons
alone would pump entropy into the photon component relative to the neutrino
component in an effectively ongoing version of the electron–positron annihilation
argument outlined above. The neutrino temperature Tν (and density ρν ) would
continue to be driven down relative to T (and ργ ) throughout the radiation-
dominated era and into the matter-dominated one. In the limit Tν /T → 0 one
sees from (5.36) and (5.38) that such a scenario would lead to
r,0 = γ = 2.48 × 10−5 h −2

0 β = 1. (5.41)
In other words, the present energy density of radiation would be lower, but it
would effectively all be in the form of photons. Insofar as the decrease in r,0
is precisely offset by the increase in β, these changes cancel each other out. The
drop in r,0 , however, has an added consequence which is not cancelled: it pushes
teq farther into the past, increasing the length of time over which the decaying
vacuum has been contributing to the background. This raises the latter’s intensity,
particularly at longer wavelengths. The effect can be significant, and we will
return to this possibility at the end of the chapter. For the most part, however, we
will stay with the values of r,0 and β given by equations (5.39) and (5.40).
Armed with a definition for vacuum luminosity, equation (5.31), and a value
for β, equation (5.40), we are in a position to calculate the luminosity of the
vacuum. Noting that V̇ = 3(R/R0 )3 ( Ṙ/R)V0 and substituting equations (5.20),
(5.21) and (5.33) into (5.31), we find that


 t −(5−8xm )/3


L v (t) = v,0 V0 × 0
t
(5.42)

 cosh(t/τ0 )
 sinh(t/τ0 ) −(5−8xm )/3
 .
cosh(t0 /τ0 ) sinh(t0 /τ0 )
Bolometric intensity 105
The first of these solutions corresponds to models with m,0 = 1 while the second
holds for the general case (0 < m,0 < 1). Both results reduce at the present time
t = t0 to
L v,0 = v,0 V0 (5.43)
where v,0 is the comoving luminosity density of the vacuum
9c2 H03r,0 βx m

v,0 =
8π G(1 − x m )

xm
= 4.1 × 10−30h 0 erg s−1 cm−3 . (5.44)
1 − xm
Numerically, we find for example that

3.1 × 10−31 h 0 erg s−1 cm−3 (x m = 0.07)
v,0 = 1.4 × 10−30 h 0 erg s−1 cm−3 (x m = 0.25) .
(5.45)
In principle, then, the comoving luminosity density of the decaying vacuum can
reach levels as high as 10 or even 50 times that of galaxies, as given by (2.24).
Raising the value of the branching ratio β to 1 instead of 0.595 does not affect
these results, since this must be accompanied by a proportionate drop in the value
of r,0 as argued earlier. The numbers in (5.45) do go up if one replaces the
relativistic definition (5.31) of vacuum luminosity with the thermodynamical one
(5.28) but the change is modest, raising v,0 by no more than a factor of 1.2 (for
x m = 0.07). The primary reason for the high luminosity of the decaying vacuum
lies in the fact that it converts nearly 60% of its energy density into photons. By
comparison, less than 1% of the rest energy of ordinary luminous matter has gone
into photons so far in the history of the Universe.
5.5 Bolometric intensity

We showed in chapter 2 that the bolometric intensity of an arbitrary distribution
of sources with comoving number density n(t) and luminosity L(t) could be
expressed as an integral over time by (2.14). Let us apply this result here to
regions of decaying vacuum energy, for which n v (t) and L v (t) are given by (5.27)
and (5.42) respectively. Putting these equations into (2.14) along with (5.20) for
the scale factor, we find that
 t −(1−8xm)/3


0 t

 dt
t
Q = cv,0 × eqt
t 0
(5.46)



0 cosh(t/τ0 ) sinh(t/τ0 ) −(1−8xm )/3
 dt.
teq cosh(t0 /τ0 ) sinh(t0 /τ0 )
The first of these integrals corresponds to models with m,0 = 1 while the second
holds for the general case (0 < m,0 < 1). The latter may be simplified
106 The vacuum
with a changeof variables to y ≡ [sinh(t/τ0 )]8xm /3 . Using the facts that

sinh(t0 /τ0 ) = (1 − m,0 )/m,0 and cosh(t0 /τ0 ) = 1/ m,0 along with the
definition (5.26) of teq , one can show that both of these integrals reduce to the
same formula,

r,0 4xm /(1−4xm)
Q = Qv 1 − . (5.47)
m,0
Here Q v is found with the help of (5.44) as
cv,0 9c3 H02r,0 β 0.0094 erg cm−2 s−1

Qv ≡ = = . (5.48)
4H0 x m 32π G(1 − x m ) (1 − x m )
There are several points to note about this result. First, it does not depend on
V0 , as expected. There is also no dependence on the uncertainty h 0 in Hubble’s
constant, since the two factors of h 0 in H02 are cancelled out by those in r,0 .
In the limit x m → 0 one sees that Q → 0 as expected. In the opposite limit
where x m → 14 , the vacuum reaches a maximum possible bolometric intensity of
Q → Q v = 0.013 erg cm−2 s−1 . This is 50 times the approximate bolometric
intensity due to galaxies, as given by (2.25).
The matter density m,0 enters only weakly into this result, and plays no role
at all in the limit x m → 14 . Based on our experience with the EBL due to galaxies,
we might have expected that Q would rise significantly in models with smaller
values of m,0 since these have longer ages, giving more time for the Universe to
fill up with light. What is happening here, however, is that the larger values of t0
are offset by larger values of teq (which follow from the fact that smaller values
of m,0 imply smaller ratios of m,0 /r,0 ). This removes contributions from
the early matter-dominated era and thereby reduces the value of Q. In the limit
x m → 14 these two effects cancel each other out. For smaller values of x m , the teq
effect proves to be the stronger of the two, and one finds an overall decrease in Q
for these cases. With x m = 0.07, for instance, the value of Q drops by 2% when
moving from the EdS model to CDM, and by another 6% when moving from
CDM to BDM.
5.6 Spectral energy distribution

To obtain limits on the parameter x m , we would like to calculate the spectral
intensity of the background due to vacuum decay, just as we did for galaxies in
chapter 3. For this we need to know the spectral energy distribution (SED) of the
decay photons. As discussed in section 5.2, theories in which the these photons
are distributed with a non-thermal spectrum can be strongly constrained by means
of distortions in the CMB. We therefore restrict ourselves here to the case of a
blackbody SED, as given by equation (3.22):
C(t)/λ5
Fv (λ, t) = (5.49)
exp[hc/kT (t)λ] − 1
Spectral energy distribution 107
where T (t) is the blackbody temperature. The function C(t) is found as usual by
normalization, equation (3.1). Changing integration variables from λ to ν = c/λ
for convenience, we find
−4
C(t) ∞ ν 3 dν C(t) h
L v (t) = 4 = 4 (4)ζ(4). (5.50)
c 0 exp[hν/kT (t)] − 1 c kT (t)
Inserting our result (5.42) for L v (t) and using the facts that (4) = 3! = 6 and
ζ (4) = π 4 /90, we then obtain for C(t):

 t −(5−8xm )/3
4  

15v,0 V0 hc t
C(t) = × 0
π4 kT (t) 

 cosh(t/τ0 ) sinh(t/τ0 ) −(5−8xm)/3
 .
(5.51)
Here the upper expression refers as usual to the EdS case (m,0 = 1), while
the lower applies to the general case (0 < m,0 < 1). Temperature T can
be determined by assuming thermal equilibrium between the vacuum-decay
photons and those already present. Stefan’s law then relates T (t) to the radiation
energy density ρr (t)c2 as follows:
4σSB
ρr (t)c2 = [T (t)]4 . (5.52)
c
Substituting equation (5.22) into this expression and expanding the Stefan–
Boltzmann constant, we find that



 t 2(1−xm )/3
 (m,0 = 1)
hc t
= λv × 0 (5.53)
kT (t) 

 sinh(t/τ0 ) 2(1−xm )/3
 (0 < m,0 < 1)
sinh(t0 /τ0 )
where the constant λv is given by
1/4 −1/4
8π 5 hc r,0 h 20
λv ≡ = 0.46 cm . (5.54)
15ρr,0 c2 4.17 × 10−5
This value of λv tells us that the peak of the observed spectrum of decay radiation
lies in the microwave region as expected, near that of the CMB (λcmb = 0.11 cm).
Putting (5.53) back into (5.51), we obtain



t
15λ4v v,0 V0  t0
C(t) = × (5.55)
π4 
 cosh(t/τ0 ) sinh(t/τ0 )
 .
These two expressions refer to models with m,0 = 1 and 0 < m,0 < 1
respectively. With C(t) thus defined, the SED (5.49) of vacuum decay is
completely specified.
108 The vacuum
5.7 The microwave background

We showed in chapter 3 that the spectral intensity of an arbitrary distribution
of sources with comoving number density n(t) and an SED F(λ, t) could be
expressed as an integral over time by (3.5). Using (5.20), (5.27), (5.49), (5.53)
and (5.55), this becomes
 1

 τ −1 dτ



 teq /t0 λv

 exp τ −2x m /3
−1

 λ0
t0 /τ0
Iλ (λ0 ) = Iv (λ0 ) × coth τ dτ (5.56)

  −2xm /3  .



 teq /τ0 λv m,0 sinh τ

 exp  −1

 λ0 1 − m,0
Here we have used integration variables τ ≡ t/t0 in the first case (for models with
m,0 = 1) and τ ≡ t/τ0 in the second (for models with 0 < m,0 < 1). The
dimensional content of both integrals is contained in the prefactor Iv (λ0 ), which
is given by
4 4
5v,0 λv xm λv
Iv (λ0 ) ≡ = 15 500 CUs . (5.57)
2π h H0 λ0
5 1 − xm λ0
We have divided through by photon energy hc/λ0 so as to express the results in
continuum units (CUs) as usual, where 1 CU ≡ 1 photon s−1 cm−2 Å−1 ster−1 .
We will use CUs throughout this book, for the sake of uniformity as well as
the fact that these units carry several advantages from the theoretical point of
view (section 3.3). The reader who consults the literature, however, will soon
find that each part of the electromagnetic spectrum has its own ‘dialect’ of
preferred units. In the microwave region it is most common to find background
intensities given in terms of the quantity ν Iν , which is the integral of flux per unit
frequency over frequency, and is usually expressed in units of nW m−2 ster−1 =
10−6 erg s−1 cm−2 ster−1 . To translate a given value of ν Iν (in these units) into
CUs, one need only multiply by a factor of 10−6 /(hc) = 50.34 erg−1 Å−1 . The
Jansky (Jy) is also often encountered, with 1 Jy = 10−23 erg s−1 cm−2 Hz−1 .
To convert a given value of ν Iν from Jy ster−1 into CUs, one multiplies by
10−23 / hλ = (1509 Hz erg−1)/λ with λ in Å.
Equation (5.56) gives the combined intensity of decay photons which have
been emitted at many wavelengths and redshifted by various amounts, but reach
us in a waveband centred on λ0 . The arbitrary volume V0 has dropped out of the
integral as expected and this result is also independent of the uncertainty h 0 in
Hubble’s constant since there is a factor of h 0 in both v,0 and H0. Results are
plotted in figure 5.2 over the waveband 0.01–1 cm, together with observational
detections of the background in this part of the spectrum. The most celebrated
of these is the COBE detection of the CMB [53] which we have shown as a
The microwave background 109
(a) (b)
xm = 0.25 xm = 0.25
xm = 0.060 xm = 0.015
xm = 0.010 xm = 0.003
xm = 0.001 −4
xm = 3×10
100000 xm = 1×10
−4 100000 xm = 3×10
−5
F96 F96
F98 F98
H98 H98
10000 L00 10000 L00
Iλ ( CUs )
Iλ ( CUs )
1000 1000
100 100
0.01 0.1 1 0.01 0.1 1

λ ( cm ) λ ( cm )
Figure 5.2. The spectral intensity of background radiation due to the decaying vacuum
for various values of xm , compared with observational data in the microwave region (bold
unbroken line) and far infrared (bold dotted line and points). For each value of xm there are
three curves representing cosmologies with m,0 = 1 (bold lines), m,0 = 0.3 (medium
lines) and m,0 = 0.03 (light lines). The left-hand panel (a) assumes L = L rel and
β = 0.595, while the right-hand panel (b) assumes L = L th and β = 1.
bold unbroken line (F96). The experimental uncertainties in this measurement

are far smaller than the thickness of the line. The other observational limits
shown in figure 5.2 have been obtained in the far infrared (FIR) region, also from
analysis of data from the COBE satellite. These are indicated with bold dotted
lines (F98 [59]) and open triangles (H98 [60] and L00 [61]).
Figure 5.2(a) shows the spectral intensity of background radiation from
vacuum decay under our standard assumptions, including the relativistic
definition (5.31) of vacuum luminosity and the values of r,0 and β given by
(5.39) and (5.40) respectively. Five groups of curves are shown, corresponding to
values of x m between 3 × 10−5 and the theoretical maximum of 0.25. For each
value of x m three curves are plotted: one each for the EdS, CDM and BDM
cosmologies. As previously noted in connection with the bolometric intensity Q,
the choice of cosmological model is less important in determining the background
due to vacuum decay than the background due to starlight from galaxies. In fact,
the intensities here are actually slightly lower in vacuum-dominated models. The
reason for this, as before, is that these models have smaller values of m,0 /r,0
and hence larger values of teq , reducing the size of contributions from the early
matter-dominated era when L v was large.
110 The vacuum
In figure 5.2(b), we have exchanged the relativistic definition of vacuum

luminosity for the thermodynamical one (5.28), and set β = 1 instead of 0.595.
As explained in section 5.4, the increase in β is partly offset by a drop in the
present radiation density r,0 . There is a net increase in intensity, however,
because smaller values of r,0 push teq back into the past, leading to additional
contributions from the early matter-dominated era. These particularly push up the
long-wavelength part of the spectrum in figure 5.2(b) relative to figure 5.2(a), as
seen most clearly for the case x m = 0.25. Overall, intensities in figure 5.2(b) are
higher than those in figure 5.2(a) by about a factor of four.
These figures tell us that the decaying-vacuum hypothesis is strongly
constrained by observations of the microwave background. The parameter x m
cannot be larger than 0.06 or the intensity of the decaying vacuum would exceed
that of the CMB itself under the most conservative assumptions, as represented by
figure 5.2(a). This limit tightens to x m 0.015 if different assumptions are made
about the luminosity of the vacuum, as shown by figure 5.2(b). These numbers are
comparable to the limit of x 0.07 obtained from entropy conservation under the
assumption that x = x r = x m [37]. And insofar as the CMB radiation is usually
attributed entirely to relic radiation from the big bang, the real limit on x m is
probably several orders of magnitude stronger than this.
With these upper bounds on x m , we can finally inquire about the potential
of the decaying vacuum as a dark-matter candidate. Since its density is given by
(5.16) as a fraction x/(1 − x) of that of radiation, we infer that its present density
parameter (v,0) satisfies:

xm 7 × 10−6 (a)
v,0 = r,0 (5.58)
1 − xm 1 × 10−6 (b) .
Here, (a) and (b) refer to the scenarios represented by figures 5.2(a) and (b),
with the corresponding values of r,0 as defined by equations (5.39) and (5.41)
respectively. We have assumed that h 0 0.6 as usual. It is clear from the
limits (5.58) that a decaying vacuum, at least of the kind we have considered
in this chapter, does not contribute significantly to the density of dark matter.
It should be recalled, however, that there are good reasons from quantum
theory for expecting some kind of instability for the vacuum in a Universe which
progressively cools. (Equivalently, there are good reasons for believing that the
cosmological ‘constant’ is not.) Our conclusion is that if the vacuum decays, it
either does so very slowly or in a manner that does not upset the isotropy of the
cosmic microwave background.
References
[1] Bronstein M 1933 Physikalische Zeitschrift der Sowjetunion 3 73
[2] Kragh H 1996 Cosmology and Controversy (Princeton, NJ: Princeton University
Press) p 36
The microwave background 111
[3] Einstein A 1936 J. Franklin Inst. 221 349

[4] Jordan P 1949 Nature 164 637
[5] Fierz M 1956 Helv. Phys. Acta 29 128
[6] Brans C and Dicke R H 1961 Phys. Rev. 124 925
[7] Bergmann P G 1968 Int. J. Theor. Phys. 1 25
[8] Endō M and Fukui T 1977 Gen. Rel. Grav. 8 833
[9] Barrow J D and Maeda K-I 1990 Nucl. Phys. B 341 294
[10] Fukui T and Overduin J M 2002 Int. J. Mod. Phys. D 11 669
[11] Madsen M S 1988 Class. Quantum Grav. 5 627
[12] Maeda K-I 1989 Phys. Rev. D 39 3159
[13] Barr S M 1987 Phys. Rev. D 36 1691
[14] Peebles P J E and Ratra B 1988 Astrophys. J. 325 L17
[15] Wetterich C 1988 Nucl. Phys. B 302 668
[16] Caldwell R R, Dave R and Steinhardt P J 1998 Phys. Rev. Lett. 80 1582
[17] Zeldovich Y B 1968 Uspekhi Fiz. Nauk 95 209
[18] Overduin J M and Cooperstock F I 1998 Phys. Rev. D 58 043506
[19] Chen W and Wu S 1990 Phys. Rev. D 41 695
[20] Polyakov A M 1982 Sov. Phys. Usp. 25 187
[21] Adler S L 1982 Rev. Mod. Phys. 54 729
[22] Dolgov A D 1983 The Very Early Universe ed G W Gibbons, S W Hawking and
S T C Siklos (Cambridge: Cambridge University Press) p 449
[23] Banks T 1985 Nucl. Phys. B 249 332
[24] Peccei R D, Solà J and Wetterich C 1987 Phys. Lett. B 195 183
[25] Frieman J A et al 1995 Phys. Rev. Lett. 75 2077
[26] Hawking S W 1984 Phys. Lett. B 134 403
[27] Brown J D and Teitelboim C 1987 Phys. Lett. B 195 177
[28] Dolgov A D 1997 Phys. Rev. 55 5881
[29] Mottola E 1985 Phys. Rev. D 31 754
[30] Tsamis N C and Woodard R P 1993 Phys. Lett. B 301 351
[31] Mukhanov V F, Abramo L R W and Brandenberger R H 1997 Phys. Rev. Lett. 78
1624
[32] Banks T 1988 Nucl. Phys. B 309 493
[33] Coleman S 1988 Nucl. Phys. B 310 643
[34] Guendelman E I and Kaganovich A B 1997 Phys. Rev. D 55 5970
[35] Weinberg S 1989 Rev. Mod. Phys. 61 1
[36] Dolgov A D 1998 The 4th Paris Cosmology Colloquium ed N Sanchez and
H J de Vega (Singapore: World Scientific)
[37] Freese K et al 1987 Nucl. Phys. B 287 797
[38] Birkel M and Sarkar S 1997 Astropart. Phys. 6 197
[39] Olson T S and Jordan T F 1987 Phys. Rev. D 35 3258
[40] Matyjasek J 1995 Phys. Rev. D 51 4154
[41] Ratra B and Peebles P J E 1988 Phys. Rev. D 37 3406
[42] Silveira V and Waga I 1994 Phys. Rev. D 50 4890
[43] Viana P T P and Liddle A R 1998 Phys. Rev. D 57 674
[44] Yoshii Y and Sato K 1992 Astrophys. J. 387 L7
[45] Sugiyama N and Sato K 1992 Astrophys. J. 387 439
[46] Silveira V and Waga I 1997 Phys. Rev. D 56 4625
[47] Ratra B and Quillen A 1992 Mon. Not. R. Astron. Soc. 259 738
112 The vacuum
[48] Bloomfield Torres L F and Waga I 1996 Mon. Not. R. Astron. Soc. 279 712
[49] Podariu S and Ratra B 2000 Astrophys. J. 532 109
[50] Overduin J M 1999 Astrophys. J. 517 L1
[51] Pavón D 1991 Phys. Rev. D 43 375
[52] Lima J A S 1996 Phys. Rev. D 54 2571
[53] Fixsen D J et al 1996 Astrophys. J. 473 576
[54] Hu W and Silk J 1993 Phys. Rev. D 48 485
[55] Pollock M D 1980 Mon. Not. R. Astron. Soc. 193 825
[56] Overduin J M, Wesson P S and Bowyer S 1993 Astrophys. J. 404 1
[57] Misner C W, Thorne K S and Wheeler J A 1973 Gravitation (San Francisco, CA:
Freeman) p 859
University Press) p 164
[59] Fixsen D J et al 1998 Astrophys. J. 508 128
[61] Lagache G et al 2000 Astron. Astrophys. 354 247
Chapter 6
Axions
6.1 Light axions

Axions are hypothetical particles whose existence would explain what is
otherwise a puzzling feature of quantum chromodynamics (QCD), the leading
theory of strong interactions. QCD contains a dimensionless free parameter ()
whose value must be ‘unnaturally’ small in order for the theory not to violate
a combination of charge conservation and mirror-symmetry known as charge
parity or CP. Upper limits on the electric dipole moment of the neutron currently
constrain the value of to be less than about 10−9 . The strong CP problem
is the question: ‘Why is so small?’ This is reminiscent of the cosmological-
constant problem which we encountered in chapter 5 (although less severe by
many orders of magnitude). Proposed solutions have similarly focused on making
, like , a dynamical variable whose value could have been driven toward zero
in the early Universe. In the most widely-accepted scenario, due to Peccei and
Quinn in 1977 [1], this is accomplished by the spontaneous breaking of a new
global symmetry (now called PQ symmetry) at energy scales f PQ . As shown by
Weinberg [2] and Wilczek [3] in 1978, the symmetry-breaking gives rise to a new
−1
particle which eventually acquires a rest energy m a c2 ∝ f PQ . This particle is the
axion (a).
Axions, if they exist, meet all the requirements of a successful cold dark-
matter or CDM candidate, as listed in section 4.4: they interact weakly with
the baryons, leptons and photons of the standard model; they are cold (i.e. non-
relativistic during the time when structure begins to form); and they are capable of
providing some or even all of the CDM density which is thought to be required,
0 º cdm º 0.6. A fourth property, and the one which is of most interest to
us here, is that axions decay generically into photon pairs. The importance of
this process depends on two things: the axion’s rest mass m a and its two-photon
coupling strength gaγ γ . Theoretical and experimental considerations restrict the
values of these two parameters but leave open the possibility that decaying axions
might contribute strongly to the extragalactic background light (EBL). Our goal
113
114 Axions
in this chapter will be to calculate the intensity of these contributions, just as we

did for those of the decaying vacuum in chapter 5. We will find that observations
of the EBL at infrared and optical wavelengths close off part of the parameter
space which is left open by other tests. Axions, if they are to make up the CDM,
must be either exceedingly light or exceedingly weakly coupled.
6.2 Rest mass

The PQ symmetry-breaking energy scale f PQ is not constrained by the theory
and reasonable values for this parameter are such that m a c2 might, in principle,
lie anywhere between 10−12 eV and 1 MeV [4]. This broad range of
theoretical possibilities has been narrowed down by an impressive combination
of cosmological, astrophysical and laboratory-based tests. In summarizing these,
it is useful to distinguish between axions with rest energies above and below
m a c2 ∼ 3 × 10−2 eV.
If the axion rest energy lies below this value, then most axions arise via
processes known as vacuum misalignment [5–7] and axionic string decay [8].
These are non-thermal mechanisms, meaning that the axions produced in this
way were never in thermal equilibrium with the primordial plasma. Their present
density would be at least [9]
−7/6
m a c2
a ≈ h −2
0 . (6.1)
4 × 10−6 eV
(This number is currently under debate, and may go up by an order of magnitude

or more if string effects play an important role [10].) If we require that axions
not provide too much CDM (cdm 0.6) then (6.1) implies a lower limit on the
axion rest energy:
m a c2 ² 7 × 10−6 . (6.2)
This neatly eliminates the lower third of the theoretically-allowed axion mass
window. Corresponding upper limits on m a in this range have come from
astrophysics. Prime among these is the fact that the weak couplings of axions
to baryons, leptons and photons allow them to stream freely from stellar cores,
carrying energy with them. More massive axions could, in principle, cool the
core of the Sun, alter the helium-burning phase in red-giant stars, and shorten the
duration of the neutrino burst from supernovae such as SN1987a. The last of
these effects is particularly sensitive and leads to the upper bound [11, 12]:
m a c2 º 6 × 10−3 eV. (6.3)
Axions with 10−5 º m a c2 º 10−2 thus remaincompatible with both

cosmological and astrophysical limits and could provide much or all of the
CDM. Moreover, it may be possible to detect these particles in the laboratory
Rest mass 115
Figure 6.1. The Feynman diagram corresponding to the decay of the axion (a) into two
photons (γ ) with coupling strength gaγ γ .
by enhancing their conversion into photons with strong magnetic fields, as shown
by Sikivie in 1983 [13]. This is now the basis for axion ongoing cavity detector
experiments in Japan [14] and the USA [15]. It has also been suggested that
the Universe itself might behave like a giant axion cavity detector, since it is
threaded by intergalactic magnetic fields. These could stimulate the reverse
process, converting photons into a different species of ‘ultralight’ axion and
possibly explaining the dimming of high-redshift supernovae without the need
for large amounts of vacuum energy [16].
Promising as they are, we will not consider low-mass axions (sometimes
known as ‘invisible axions’) further in this chapter. This is because they decay
too slowly to have a noticeable impact on the EBL. Axions decay into photon
pairs (a → γ + γ ) via a loop diagram, as illustrated in figure 6.1. The decay
lifetime of this process is [4]
τa = (6.8 × 1024 s)m −5

1 ζ
−2
. (6.4)
Here m 1 ≡ m a c2 /(1 eV) is the axion rest energy in units of eV and ζ is a constant
which is proportional to the coupling strength gaγ γ [17]. For our purposes, it
is sufficient to treat ζ as a free parameter which depends on the details of the
axion theory chosen. Its value has been normalized in equation (6.4) so that
ζ = 1 in the simplest grand unified theories (GUTs) of strong and electroweak
interactions. This could drop to ζ = 0.07 in other theories, however [18],
amounting to a strong suppression of the two-photon decay channel. In principle ζ
could even vanish altogether, corresponding to a radiatively stable axion, although
this would require an unlikely cancellation of terms. We will consider values
in the range 0.07 ζ 1 in what follows. For these values of ζ , and with
m 1 º 6 × 10−3 as given by (6.3), equation (6.4) shows that axions decay on
timescales τa ² 9 × 1035 s. This is so much longer than the age of the Universe
that such particles would truly be invisible.
We therefore shift our attention to axions with rest energies above m a c2 ∼
3 × 10−2 eV. Turner showed in 1987 [19] that the vast majority of these
would have arisen via thermal mechanisms such as Primakoff scattering and
photoproduction processes in the early Universe. Application of the Boltzmann
equation gives their present comoving number density as n a = (830/g∗F ) cm−3 ,
where g∗F counts the number of relativistic degrees of freedom left in the plasma
116 Axions
at the time when axions ‘froze out’ [4]. Ressell [17] has estimated the latter
quantity as g∗F = 15. The present density parameter a = n a m a /ρcrit,0 of
thermal axions is thus
a = 5.2 × 10−3h −2
0 m 1. (6.5)
Whether or not this is significant depends on the axion rest mass. The neutrino
burst from SN1987a now imposes a lower limit on m a rather than an upper limit as
before, because axions in this range are massive enough to interact with nucleons
in the supernova core and can no longer stream out freely (this is referred to as
the ‘trapped regime’). Only axions at the lower end of the mass-range are able
to carry away enough energy to interfere with the duration of the neutrino burst,
leading to the limit [20]
m a c2 2.2 eV. (6.6)
Laboratory upper limits on m a come from the fact that axions cannot carry away
too much ‘missing energy’ in processes such as the decay of the kaon [21].
Astrophysical constraints (similar to those mentioned already) are considerably
stronger. These now depend critically on whether axions couple only to hadrons
at tree-level, or to leptons as well. The former kind are known as KSVZ axions
after Kim [22] and Shifman, Vainshtein and Zakharov [23]; while the latter take
the name DFSZ axions after Zhitnitsky [24] and Dine, Fischler and Srednicki
[25]. The extra lepton coupling of DFSZ axions allows them to carry so much
energy out of the cores of red-giant stars that helium ignition is disrupted unless
m a c2 9 × 10−3 eV [26]. This closes the high-mass window on DFSZ axions,
which can consequently exist only with the ‘invisible’ masses discussed earlier.
For hadronic (KSVZ) axions, the duration of the helium-burning phase in red
giants imposes a weaker bound [27]:
m a c2 0.7ζ −1 eV. (6.7)
This translates into an upper limit m a c2 10 eV for the simplest axion models
with ζ 0.07. And irrespective of the value of ζ , it has been argued that axions
with m a c2 10 eV can be ruled out because they would interact strongly enough
with baryons to produce a detectable signal in existing Čerenkov detectors [28].
In the hadronic case, then, there remains a window of opportunity for multi-
eV axions with 2 m1 10. Equation (6.5) shows that these particles
would contribute a total density of about 0.03 a 0.15, where we take
0.6 h 0 0.9 as usual. Axions of this kind would not be able to provide the
required density of dark matter in the CDM model (m,0 = 0.3). They would,
however, suffice in low-density models midway between CDM and BDM
(section 3.3). Since such models are compatible with current observational data
(chapter 4), it is worth proceeding to see whether multi-eV axions can be further
constrained by their contributions to the EBL.
Axion halos 117
6.3 Axion halos

Thermal axions are not as cold as their non-thermal cousins, but will still be
found primarily inside gravitational potential wells such as those of galaxies
and galaxy clusters [19]. We need not be too specific about the fraction which
have settled into galaxies as opposed to larger systems, because we will be
concerned primarily with their combined contributions to the diffuse background.
(Distribution could become an issue if extinction due to dust or gas played a
strong role inside the bound regions, but this is not likely to be important for
the photon energies under consideration here.) These axion halos provide us
with a convenient starting-point as cosmological sources of axion-decay radiation,
analogous to the galaxies and vacuum source regions of previous chapters. Let
us consider to begin with the possibility that axions are cold enough that their
fractional contribution (Mh ) to the total mass of each halo (Mtot ) is the same as
their fractional contribution to the cosmological matter density:
Mh a a
= = . (6.8)
Mtot m,0 a + bar
Here we have made the minimal assumption that all the non-baryonic dark matter
is provided by axions. (As noted earlier, this effectively restricts us to strongly
vacuum-dominated cosmological models.) The mass of axions in each halo is,
from (6.8):

bar −1
Mh = Mtot 1 + . (6.9)
a
If these regions are distributed with a mean comoving number density n 0 , then
the cosmological density of bound axions is

n 0 Mh n 0 Mtot bar −1
a,bound = = 1+ . (6.10)
ρcrit,0 ρcrit,0 a
Setting this equal to a as given by (6.5) fixes the total mass:

a ρcrit,0 bar
Mtot = 1+ . (6.11)
n0 a
The comoving number density of galaxies at z = 0 is commonly taken as [29]
n 0 = 0.010h 30 Mpc−3 . (6.12)
Using this together with (6.5) for a , and setting bar ≈ 0.016h −2
0 from
section 4.3, we find from (6.11) that

 −3
 9 × 10 M h 0 (m 1 = 3)
11
−3
Mtot = 1 × 10 M h 0
12 (m 1 = 5) (6.13)

 2 × 1012 M h −3 (m = 8).
0 1
118 Axions
Let us compare these numbers with recent dynamical data on the mass of the
Milky Way using the motions of galactic satellites. These assume a Jaffe profile
[30] for the halo density:
vc2 rj2
ρtot (r ) = (6.14)
4π Gr 2 (r + rj )2
where vc is the circular velocity, rj the Jaffe radius and r the radial distance
from the centre of the Galaxy. The data imply that vc = 220 ± 30 km s−1 and
rj = 180 ± 60 kpc [31]. Integrating over r from zero to infinity gives
vc2rj
Mtot = = (2 ± 1) × 1012 M . (6.15)
G
This is consistent with (6.13) for a wide range of values of m 1 and h 0 . So axions
of this type could, in principle, make up all the dark matter which is required on
galactic scales.
Putting (6.11) into (6.9) we obtain for the mass of the axion halos:
a ρcrit,0
Mh = . (6.16)
n0
This could also have been derived as the mass of a region of space of comoving
volume V0 = n −1 0 filled with homogeneously-distributed axions of mean density
ρa = a ρcrit,0 . (This is the approach that we adopted in defining vacuum regions
in chapter 5.)
To obtain the halo luminosity, we add up the rest energies of all the decaying
axions in the halo and divide by the decay lifetime:
Mh c2
Lh = . (6.17)
τa
Substituting (6.4) and (6.16) into this equation, we find that
L h = (3.8 × 1040 erg s−1 )h −3

0 ζ m1
2 6
(6.18)

 −3 2
 7 × 10 L h 0 ζ (m 1 = 3)
9
= 2 × 1011 L h −3
0 ζ
2 (m 1 = 5)

 3 × 1012 L h −3 ζ 2 (m = 8).
0 1
These numbers should be compared to the luminosities of the galaxies themselves,

which are of order L 0 = 0 /n 0 = 2 × 1010h −2 0 L , where we have used (2.24)
for 0 . The proposed axion halos are, in principle, capable of outshining their
host galaxies unless the axion is either light (m 1 º 3) or weakly-coupled (ζ < 1).
Intensity 119
6.4 Intensity
Substituting the halo comoving number density n 0 and luminosity L h into
equation (2.20), we find for the bolometric intensity of the decaying axions:
zf
dz
Q = Qa . (6.19)
0 (1 + z)2 H̃ (z)
Here the dimensional content of the integral is contained in the prefactor Q a which
takes the following numerical values:
a ρcrit,0 c3
Qa = = (1.2 × 10−7 erg s−1 cm−2 )h −3
0 ζ m1
2 6
(6.20)
H0 τa

 −5 −1 −2 −1 2 (m = 3)
 9 × 10 erg s cm )h 0 ζ 1
= 2 × 10−3 erg s−1 cm−2 )h −1 ζ 2 (m = 5)
1


0
3 × 10−2 erg s−1 cm−2 )h −10 ζ 2 (m = 8).
1
There are three things to note about this quantity. First, it is comparable in
magnitude to the observed EBL due to galaxies, Q ∗ ≈ 3 × 10−4 erg s−1 cm−2
(chapter 2). Second, unlike Q ∗ for galaxies or Q v for decaying vacuum energy,
Q a depends explicitly on the uncertainty h 0 in Hubble’s constant. Physically, this
reflects the fact that the axion density ρa = a ρcrit,0 in the numerator of (6.20)
comes to us from the Boltzmann equation and is independent of h 0 , whereas the
density of luminous matter such as that in galaxies is inferred from its luminosity
density 0 (which is proportional to h 0 , thus cancelling the h 0 -dependence in
H0). The third thing to note about Q a is that it is independent of n 0 . This is
because the collective contribution of decaying axions to the diffuse background
is determined by their mean density a and does not depend on how they are
distributed in space.
To evaluate (6.19) we need to specify the cosmological model. Let us
assume that the Universe is spatially flat, as is increasingly suggested by the data
(chapter 4). Hubble’s parameter (2.40) then reduces to
H̃ (z) = [m,0 (1 + z)3 + 1 − m,0 ]1/2 (6.21)
where m,0 = a + bar . Putting this into (6.19) along with (6.20) for Q a , we
obtain the plots of Q(m 1 ) shown in figure 6.2 for ζ = 1. The three bold lines in
this plot show the range of intensities obtained by varying h 0 and bar h 20 within
the ranges 0.6 h 0 0.9 and 0.011 bar h 20 0.021 respectively. We have
set z f = 30, since axions were presumably decaying long before they became
bound to galaxies. (Results are insensitive to this choice, rising by less than 2%
as z f → 1000 and dropping by less than 1% for z f = 6.) The axion-decay
background is faintest for the largest values of h 0 , as expected from the fact that
Q a ∝ h −1
0 . This is partly offset, however, by the fact that larger values of h 0 also
lead to a drop in m,0 , extending the age of the Universe and hence the length
120 Axions
0.1
Ωbarh20 = 0.011, h0 = 0.6
Ωbarh20 = 0.016, h0 = 0.75
Ωbarh20 = 0.021, h0 = 0.9
0.01
EdS
Q ( erg s−1 cm−2 )
Q∗
0.001
0.0001
1e−05
2 3 4 5 6 7 8 9 10
ma c2 ( eV )
Figure 6.2. The bolometric intensity Q of the background radiation from axions as a
function of their rest mass m a . The faint dash-dotted line shows the equivalent intensity
in an EdS model (in which axions alone cannot provide the required CDM). The dotted
horizontal line indicates the approximate bolometric intensity (Q ∗ ) of the observed EBL.
of time over which axions have been contributing to the background. (Smaller
values of bar raise the intensity slightly for the same reason.) Overall, figure 6.2
confirms that axions with ζ = 1 and m a c2 3.5 eV produce a background
brighter than that from the galaxies themselves.
6.5 The infrared and optical backgrounds
To compare our predictions with observational data, it is necessary to calculate

the intensity of axionic contributions to the EBL as a function of wavelength.
The first step, as usual, is to specify the spectral energy distribution or SED
of the decay photons in the rest frame. Each axion decays into two photons of
energy 12 m a c2 (figure 6.1), so that the decay photons are emitted at or near a peak
wavelength
2hc 24 800 Å
λa = = . (6.22)
m a c2 m1
Since 2 º m 1 º 10, the value of this parameter tells us that we will be most
interested in the infrared and optical bands (roughly 4000–40 000 Å). We can
The infrared and optical backgrounds 121
model the decay spectrum with a Gaussian SED as in (3.19) :

Lh 1 λ − λa 2
F(λ) = √ exp − . (6.23)
2πσλ 2 σλ
For the standard deviation of the curve, we can use the velocity dispersion
vc of the bound axions [32]. This is 220 km s−1 for the Milky Way, giving
σλ ≈ 40 Å/m 1 where we have used σλ = 2(vc /c)λa (section 3.4). For axions
bound in galaxy clusters, vc rises to as much as 1300 km s−1 [17], implying that
σλ ≈ 220 Å/m 1 . For convenience we parametrize σλ in terms of a dimensionless
quantity σ50 ≡ σλ /(50 Å/m 1 ) so that
σλ = (50 Å/m 1 )σ50 . (6.24)
With the SED F(λ) thus specified along with Hubble’s parameter (6.21), the
spectral intensity of the background radiation produced by axion decays is given
by (3.6) as

1 λ0 /(1 + z) − λa 2
zf exp − 2 σλ
dz
Iλ (λ0 ) = Ia . (6.25)
0 (1 + z) [m,0 (1 + z) + 1 − m,0 ]
3 3 1/2
The dimensional prefactor in this case reads

a ρcrit,0 c2 λ0
Ia = √
32π 3 h H0τa σλ

λ0
= (95 CUs)h −1 2 7 −1
0 ζ m 1 σ50 . (6.26)
24 800 Å
Here we have divided through by the photon energy hc/λ0 to put results into
continuum units or CUs as usual (section 3.3). The source number density in
(3.6) cancels out the factor of 1/n 0 in source luminosity (6.17) so that results
are independent of axion distribution, as expected. Evaluating equation (6.25)
over 2000 Å λ0 20 000 Å with ζ = 1 and z f = 30, we obtain the plots
of Iλ (λ0 ) in figure 6.3. Three groups of curves are shown, corresponding to
m a c2 = 3, 5 and 8 eV. For each value of m a there are four curves; these assume
(h 0 , bar h 20 ) = (0.6, 0.011), (0.75, 0.016) and (0.9, 0.021) respectively, with the
fourth (faint dash-dotted) curve representing the equivalent intensity in an EdS
Universe (as in figure 6.2). Also plotted in figure 6.3 are many of the reported
observational constraints on EBL intensity in this waveband. Most have been
encountered already in chapter 3. They include data from the OAO-2 satellite
(LW76 [33]), several ground-based telescope observations (SS78 [34], D79 [35],
BK86 [36]), the Pioneer 10 spacecraft (T83 [37]), sounding rockets (J84 [38], T88
[39]), the Space Shuttle-borne Hopkins UVX experiment (M90 [40]), combined
122 Axions
ma c2 = 8 eV
2
ma c = 5 eV
2
100000 ma c = 3 eV
EdS
LW76
SS78
D79
T83
10000 J84
Iλ ( CUs )
BK86
T88
M90
B98
H98
1000 WR00
C01
10000 100000
λ0 ( Å )
Figure 6.3. The spectral intensity Iλ of the background radiation from decaying axions
as a function of observed wavelength λ0 . The four curves for each value of m a (labelled)
correspond to upper, median and lower limits on h 0 and bar together with the equivalent
intensity for the EdS model (as in figure 6.2). Also shown are observational upper limits
(full symbols and bold lines) and reported detections (empty symbols) over this waveband.
HST/Las Campanas telescope observations (B98 [41]) and the DIRBE instrument
aboard the COBE satellite (H98 [42], WR00 [43], C01 [44]).
Figure 6.3 shows that 8 eV axions with ζ = 1 would produce a hundred
times more background light at ∼3000 Å than is actually seen. The background
from 5 eV axions would exceed observed levels by a factor of ten at ∼5000 Å.
(This, we can note in passing, would colour the night sky green.) Only axions
with m a c2 3 eV are compatible with observation if ζ = 1. These results
are significantly higher than those obtained assuming an EdS cosmology [17, 45]
especially at wavelengths longward of the peak. This simply reflects the fact
that the background in a low-m,0 , high-,0 Universe like that considered here
receives many more contributions from sources at high redshift.
To obtain more detailed constraints, we can instruct a computer to evaluate
the integral (6.25) at more finely-spaced intervals in m a . Since Iλ ∝ ζ −2 , the
value of ζ required to reduce the minimum predicted axion intensity Ith (m a , λ0 )
below a given observational upper limit Iobs (λ0 ) in figure 6.3 is

Iobs (λ0 )
ζ (m a , λ0 ) . (6.27)
Ith (m a , λ0 )
10
Ωm,0 = Ωa + Ωbar (axions + baryons only)
Ωm,0 = 1 (EdS)
9
8
ma c2 ( eV )
3
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
ζ
Figure 6.4. The upper limits on the value of m a c2 as a function of the coupling strength ζ
(or vice versa). These are derived by requiring that the minimum predicted axion intensity
(as plotted in figure 6.3) be less than or equal to the observational upper limits on the
intensity of the EBL.
The upper limit on ζ (for each value of m a ) is then the smallest such value of
ζ (m a , λ0 ); i.e. that which brings Ith (m a , λ0 ) down to Iobs (λ0 ) or less for each
wavelength λ0 . From this procedure we obtain a function which can be regarded
as an upper limit on the axion rest mass m a as a function of ζ (or vice versa).
Results are plotted in figure 6.4 (bold full line). This curve tells us that even in
models where the axion–photon coupling is strongly suppressed and ζ = 0.07,
the axion cannot be more massive than
m a c2 8.0 eV (ζ = 0.07). (6.28)
For the simplest axion models with ζ = 1, this tightens to
m a c2 3.7 eV (ζ = 1). (6.29)
As expected, these limits are stronger than those which would be obtained in an
EdS model (faint dotted line in figure 6.4). This is a small effect, however, because
the strongest constraints tend to come from the region near the peak wavelength
(λa ), whereas the difference between matter- and vacuum-dominated models is
most pronounced at wavelengths longward of the peak where the majority of the
radiation originates at high redshift. Figure 6.4 shows that cosmology has the
most effect over the range 0.1 º ζ º 0.4, where upper limits on m a c2 weakened
124 Axions
by about 10% in the EdS model relative to one in which all the CDM is provided
by axions of the kind we have discussed here.
Combining equations (6.6) and (6.29) we conclude that axions in the
simplest models are confined to a slender range of viable rest masses:
2.2 eV º m a c2 3.7 eV. (6.30)
Considerations of background radiation thus complement the red-giant

bound (6.7) and close off most, if not all of the multi-eV window for thermal
axions. The range of values (6.30) can be further narrowed by looking for the
enhanced signal which might be expected to emanate from concentrations of
bound axions associated with galaxies and clusters of galaxies, as first suggested
by Kephart and Weiler in 1987 [32]. The most thorough search programme along
these lines was that carried out in 1991 by Ressell [17], who found no evidence
of such a signal from three selected clusters, further tightening the upper limit on
the multi-eV axion window to 3.2 eV in the simplest axion models. Constraints
obtained in this way for non-thermal axions would be considerably weaker, as
noted by several workers [32, 46]. This has no effect on our results, however,
since non-thermal axions are vastly outnumbered by thermal ones over the range
of rest masses we have considered here.
Let us turn finally to the question of how much dark matter can be provided
by axions of this type. With rest energies given by (6.30), equation (6.5) shows
that
0.014 a 0.053. (6.31)
Here we have taken 0.6 h 0 0.9 as usual. This is comparable to the density of
baryonic matter (section 4.3), but falls well short of most expectations for CDM
density (section 4.4).
Our main conclusions, then, are as follows: thermal axions in the multi-eV
window remain (only just) viable at the lightest end of the range of possible rest-
masses given by equation (6.30). They may also exist with slightly higher rest-
masses, up to the limit given by equation (6.28), but only in certain axion theories
where their couplings to photons are weak. In either of these two scenarios, their
contributions to the density of dark matter in the Universe are so feeble as to
remove much of their motivation as CDM candidates. If they are to provide a
significant portion of the dark matter, then axions must have rest masses in the
‘invisible’ range where they do not contribute significantly to the light of the night
sky.
References
[1] Peccei R and Quinn H 1977 Phys. Rev. Lett. 38 1440
[2] Weinberg S 1978 Phys. Rev. Lett. 40 223
[3] Wilczek F 1978 Phys. Rev. Lett. 40 279
[4] Kolb E W and Turner M S 1990 The Early Universe (Reading, MA: Addison-Wesley)
pp 401–27
[5] Preskill J, Wise M and Wilczek F 1983 Phys. Lett. B 120 127
[6] Abbott L and Sikivie P 1983 Phys. Lett. B 120 133
[7] Dine M and Fischler W 1983 Phys. Lett. B 120 137
[8] Davis R 1986 Phys. Lett. B 180 255
[9] Sikivie P 2000 Beyond the Desert 1999 ed H V Klapdor-Kleingrothaus and
I V Krivosheina (Oxford: Institute of Physics Press) p 547
[10] Battye R A and Shellard E P S 2000 Beyond the Desert 1999 ed H V Klapdor-
Kleingrothaus and I V Krivosheina (Oxford: Institute of Physics Press) p 565
[11] Janka H-T et al 1996 Phys. Rev. Lett. 76 2621
[12] Keil W et al 1997 Phys. Rev. D 56 2419
[13] Sikivie P 1983 Phys. Lett. 51 1415
[14] Yamamoto K et al 2000 Dark Matter in Astro- and Particle Physics ed H V Klapdor-
Kleingrothaus (Heidelberg: Springer) p 638
[15] Asztalos S J and Kinion D 2000 Dark Matter in Astro- and Particle Physics ed
H V Klapdor-Kleingrothaus (Heidelberg: Springer) p 630
[16] Csáki C, Kaloper N and Terning J 2002 Phys. Rev. Lett. 88 161302
[17] Ressell M T 1991 Phys. Rev. D 44 3001
[18] Kaplan D B 1985 Nucl. Phys. B 260 215
[19] Turner M S 1987 Phys. Rev. Lett. 59 2489
[20] Turner M S 1988 Phys. Rev. Lett. 60 1797
[21] Kim J-E 1987 Phys. Rep. 150 1
[22] Kim J E 1979 Phys. Rev. Lett. 43 103
[23] Shifman M A, Vainshtein A I and Zakharov V I 1980 Nucl. Phys. B 166 493
[24] Zhitnitsky A R 1980 Sov. J. Nucl. Phys. 31 260
[25] Dine M, Fischler W and Srednicki M 1981 Phys. Lett. B 104 199
[26] Raffelt G and Weiss A 1995 Phys. Rev. D 51 1495
[27] Raffelt G G and Dearborn D S P 1987 Phys. Rev. D 36 2211
[28] Engel J, Seckel D and Hayes A C 1990 Phys. Rev. D 65 960
[30] Jaffe W 1983 Mon. Not. R. Astron. Soc. 202 995
[32] Kephart T W and Weiler T J 1987 Phys. Rev. Lett. 58 171
[34] Spinrad H and Stone R P S 1978 Astrophys. J. 226 609
[36] Boughn S P and Kuhn J R 1986 Astrophys. J. 309 33
[37] Toller G N 1983 Astrophys. J. 266 L79
[38] Jakobsen P et al 1984 Astron. Astrophys. 139 481
[39] Tennyson P D et al 1988 Astrophys. J. 330 435
[40] Murthy J et al 1990 Astron. Astrophys. 231 187
[43] Wright E L and Reese E D 2000 Astrophys. J. 545 43
126 Axions
[44] Cambrésy L et al 2001 Astrophys. J. 555 563

[45] Overduin J M and Wesson P S 1993 Astrophys. J. 414 449
[46] Lodenquai J F and Dixit V V 1987 Phys. Lett. B 194 350
Chapter 7
Neutrinos
7.1 The decaying-neutrino hypothesis

Experiments now indicate that neutrinos possess non-zero rest mass and make
up at least part of the dark matter. If different neutrino species have different
rest masses, then heavier species can decay into lighter ones plus a photon.
The neutrino number density, moreover, is high enough that their decay photons
might be observable, as first appreciated by Cowsik in 1977 [1] and de Rujula
and Glashow in 1980 [2]. The strength of the expected signal depends on the
way in which neutrino rest masses are incorporated into the standard model of
particle physics. In minimal extensions of this model, radiative neutrino decays
are characterized by lifetimes on the order of 1029 yr or more [3]. This is so
much longer than the age of the Universe that neutrinos are effectively stable
and would not produce a detectable signal. In other theories such as those
involving supersymmetry, however, their decay lifetime can drop to 1015 yr [4].
This is within five orders of magnitude of the age of the Universe and opens up
the possibility of significant contributions to the extragalactic background light
(EBL).
Decay photons from neutrinos with lifetimes this short are also interesting
because of their potential for resolving a number of longstanding astrophysical
puzzles related to the ionization of hydrogen and nitrogen in the interstellar and
intergalactic medium [5, 6]. As was first pointed out by Melott et al in 1988
[7], these point in particular to a neutrino which decays on timescales of order
τν ∼ 1024 s and has a rest energy m ν ∼ 30 eV. The latter value fits awkwardly
with current thinking on large-scale structure formation in the early Universe
(section 4.5). Neutrinos of this kind could help with so many other problems,
however, that the prospect of their existence bears further scrutiny. Sciama and
others have been led on this basis to develop a detailed scenario, now known as the
decaying-neutrino hypothesis [8–10]. In this, the rest energy and decay lifetime
of the massive τ neutrino are specified as
m ντ c2 = 28.9 ± 1.1 eV τν = (2 ± 1) × 1023 s. (7.1)
127
128 Neutrinos
The τ neutrino decays into a µ neutrino plus a photon (figure 7.1). Assuming that
m ντ m νµ , conservation of energy and momentum require this photon to have
an energy E γ = 12 m ντ c2 = 14.4 ± 0.5 eV.
The concreteness of this proposal has made it eminently testable. Some of
the strongest bounds come from searches for the line emission near 14 eV that
would be expected from concentrations of decaying dark matter in clusters of
galaxies. No such signal has been seen in the direction of the galaxy cluster
surrounding the quasar 3C 263 [11] or in the direction of the rich cluster Abell 665
which was observed using the Hopkins Ultraviolet Telescope in 1991 [12]. It
may be, however, that absorption plays a stronger role than expected along the
lines of sight to these clusters or that most of their dark matter is in another
form [4, 13]. A potentially more robust test of the decaying-neutrino hypothesis
comes from the diffuse background light. This has been looked at in a number of
studies [14–20]. The task is a challenging one for several reasons. Photons whose
energies lie near 14 eV are strongly absorbed by both dust and neutral hydrogen,
and the distribution of these quantities in intergalactic space is uncertain. It is also
notoriously difficult, perhaps more so in this part of the spectrum than any other,
to distinguish between those parts of the background which are truly extragalactic
and those which are due to a complex mixture of competing foreground signals
[21, 22].
Here we reconsider the problem with the help of the formalism developed
in preceding chapters, adapting it to allow for absorption by dust and neutral
hydrogen. Despite all the uncertainties, we will come to a firm conclusion:
existing observations are sufficient to rule out the decaying-neutrino hypothesis,
unless either the rest mass or decay lifetime of the neutrino lie outside the bounds
specified by the theory.
7.2 Bound neutrinos

Neutrinos, the original hot dark-matter candidates, are likely to be found
predominantly outside gravitational potential wells such as those of individual
galaxies. However, some at least will have tended to collect into galactic
dark-matter halos since the epoch of galaxy formation. These neutrino halos
comprise our sources of background radiation in this section. Their comoving
number density is just that of the galaxies themselves, given by (6.12) as n 0 =
0.010h 30 Mpc−3 . The decay photon wavelengths at emission can be taken as
distributed normally about the peak wavelength corresponding to E γ :
hc
λν = = 860 ± 30 Å. (7.2)
Eγ
This lies in the extreme ultraviolet (EUV) portion of the spectrum, although
the redshifted tail of the observed photon spectrum will stretch across the far
ultraviolet (FUV) and near ultraviolet (NUV) bands as well. [For definiteness,
Bound neutrinos 129
Figure 7.1. Feynman diagrams corresponding to the decay of a massive neutrino (ν1 ) into
a second, lighter neutrino species (ν2 ) together with a photon (γ ). The process is mediated
by charged leptons ( ) and the W boson (W).
we take these wavebands to extend over 100–912 Å (EUV), 912–2000 Å (FUV)

and 2000–4000 Å (NUV). It should, however, be noted that universal standards
have yet to be established in this area and individual conventions vary.] The
spectral energy distribution (SED) of the decaying-neutrino halos is then given
by equation (3.19) :

Lh 1 λ − λγ 2
F(λ) = √ exp − (7.3)
2πσλ 2 σλ
where L h , the halo luminosity, has yet to be determined. For the standard
deviation σλ we can follow the same procedure as in the preceding chapter and
use the velocity dispersion of the bound neutrinos, giving σλ = 2λν vc /c. Let us
parametrize this for convenience using the range of uncertainty given by (7.2) for
the value of λν , so that σ30 ≡ σλ /(30 Å).
The halo luminosity is just the ratio of the number of decaying neutrinos
(Nτ ) to their decay lifetime (τν ), multiplied by the energy of each decay photon
(E γ ). Because the latter is just above the hydrogen-ionizing energy of 13.6 eV,
we also need to multiply the result by an efficiency factor (between zero and
one), to reflect the fact that some of the decay photons are absorbed by neutral
hydrogen in their host galaxy before they can leave the halo and contribute to its
luminosity. Altogether, then:
Nτ E γ Mh c2
Lh = = . (7.4)
τν 2τν
Here we have expressed Nτ as the number of neutrinos with rest mass m ντ =
2E γ /c2 per halo mass Mh .
To calculate the mass of the halo, let us follow reasoning similar to that
adopted for axion halos in section 6.3 and assume that the ratio of baryonic-to-
total mass in the halo is comparable to the ratio of baryonic-to-total matter density
in the Universe at large:
Mtot − Mh Mbar bar

= = . (7.5)
Mtot Mtot bar + ν
130 Neutrinos
Here we have made the economical assumption that there are no other
contributions to the matter density, apart from those of baryons and massive
neutrinos. It follows from equation (7.5) that

bar −1
Mh = Mtot 1 + . (7.6)
ν
We take Mtot = (2 ± 1) × 1012 M following equation (6.15). For bar we use the
value (0.016 ± 0.005)h −20 quoted in section 4.3. And to calculate ν we put the
neutrino rest mass m ντ into equation (4.7), giving
ν = (0.31 ± 0.01)h −2
0 . (7.7)
Inserting these values of Mtot , bar and ν into (7.6), we obtain
Mh = (0.95 ± 0.01)Mtot = (1.9 ± 0.9) × 1012 M . (7.8)
This leaves a baryonic mass Mbar = Mtot − Mh ≈ 1×1011 M , in good agreement

with the observed sum of contributions from stars in the disc, bulge and halo of
our own Galaxy, together with the matter making up the interstellar medium.
The neutrino density (7.7), when combined with that of baryons, leads to a
total present-day matter density of
m,0 = bar + ν = (0.32 ± 0.01)h −2

0 . (7.9)
As pointed out by Sciama [9], massive neutrinos are thus consistent with a critical-
density EdS Universe (m,0 = 1) if
h 0 = 0.57 ± 0.01. (7.10)
This is just below the range of values which many workers now consider
observationally viable for Hubble’s constant (section 4.3). But it is a striking fact
that the same neutrino rest mass which resolves several unrelated astrophysical
problems also implies a reasonable expansion rate in the simplest cosmological
model. In the interests of testing the decaying-neutrino hypothesis in a
self-consistent way, we will follow Sciama in adopting the narrow range of
values (7.10) for this chapter only.
7.3 Luminosity
To evaluate the halo luminosity (7.4), it remains to find the fraction of decay
photons which escape from the halo. A derivation of is given in appendix C,
based on the hydrogen photoionization cross section with respect to 14 eV
photons, and taking into account the distribution of decaying neutrinos relative
to that of neutral hydrogen in the galactic disc. Here we show that the results can
Luminosity 131
be simply understood once it is appreciated that the photoionization cross section

and hydrogen density are such that effectively all of the decay photons striking
the disc are absorbed. The probability of absorption for a single decay photon is
then proportional to the solid angle subtended by the galactic disc, as seen from
the point where the photon is released.
We model the distribution of τ neutrinos (and their decay photons) in the
halo with a flattened ellipsoidal profile which has been advocated in the context
of the decaying-neutrino scenario by Salucci and Sciama [23]. This has
ρν (r, z) = 4n m ντ ν (r, θ ) (7.11)
with −2

ν (r, θ ) ≡ 1 + (r/r ) sin θ + (r/ h) cos θ
2 2 2 2 .
Here r and θ are spherical coordinates, n = 5 × 107 cm−3 is the number

density of halo neutrinos in the vicinity of the Sun, r = 8 kpc is the distance of
the Sun from the centre of the Galaxy, and h = 3 kpc is the scale height of the
halo. Although this function has essentially been constructed to account for the
ionization structure of the Milky Way, it agrees reasonably well with dark-matter
halo distributions which have derived on strictly dynamical grounds [24].
Defining x ≡ r/r , one can use (7.11) to express the mass Mh of the halo in
terms of the halo radius rh as
π/2 x max (rh ,θ)
Mh (rh ) = Mν ν (x, θ )x 2 sin θ dx dθ (7.12)
θ=0 x=0
where
3
Mν ≡ 16πn m ντ r = 9.8 × 1011 M

x max (rh , θ ) = (rh /r )/ sin2 θ + (r / h)2 cos2 θ.
Outside x > x max , we assume that the halo density drops off exponentially and can
be ignored. Using (7.12) it can be shown that halos whose masses Mh are given
by (7.8) have scale radii rh = (70 ± 25) kpc. This is consistent with evidence
from the motion of galactic satellites [25].
We now put ourselves in the position of a decay photon released at cylindrical
coordinates (yν , z ν ) inside the halo, as shown in figure 7.2(a). It may be seen that
the disc of the Galaxy presents an approximately elliptical figure with maximum
angular width 2α and angular height 2β, where

α = tan−1 (rd2 − d 2 )/[(yν − d)2 + z ν2 ]

1 −1 yν + rd −1 yν − rd
β= tan − tan . (7.13)
2 zν zν
132 Neutrinos
zν
α
β βθ
α
rd yν rh
a φ
b r
(a) (b)
Figure 7.2. Left-hand side (a): absorption of decay photons inside the halo. For neutrinos
decaying at (yν , z ν ), the probability that decay photons will be absorbed inside the halo
(radius rh ) is essentially the same as the probability that they will strike the galactic disc
(radius rd , shaded). Right-hand side (b): an ellipse of semi-major axis a, semi-minor axis
b and radial arm r (φ) subtends corresponding angles of α, β and θ(φ) respectively.

Here d = [(yν2 + z ν2 + rd2 ) − (yν2 + z ν2 + rd2 )2 − 4yν2rd2 ]/2yν , yν = r sin θ ,
z ν = r cos θ and r = r x. In spherical coordinates centred on the photon, the
solid angle subtended by an ellipse is
2π θ(φ) 2π
e = dφ sin θ dθ = [1 − cos θ (φ)] dφ. (7.14)
0 0 0
Here θ (φ) is the angle subtended by a radial arm of the ellipse as depicted in
figure 7.2(b). The cosine of this angle can be expressed in terms of α and β by
−1/2
tan2 α tan2 β
cos θ (φ) = 1 + . (7.15)
tan2 α sin2 φ + tan2 β cos2 φ
The single-point probability that a photon released at (x, θ ) will escape from the
halo is then
e = 1 − e
(α, β)
. (7.16)
4π
For a given halo size rh and disc size rd , we obtain a good approximation to
by averaging e over all locations (x, θ ) in the halo and weighting by the
neutrino number density ν . Trial and error shows that, for the range of halo
sizes considered here, a disc radius of rd = 36 kpc leads to values which are
within 1% of those found in appendix C. The latter read:
0.63 (r = 45 kpc)
h
= 0.77 (rh = 70 kpc) (7.17)
0.84 (rh = 95 kpc).
Free-streaming neutrinos 133
As expected, the escape fraction of decay photons goes up as the scale size of the
halo increases relative to that of the disc. As rh rd one gets → 1, while a
small halo with rh º rd leads to ≈ 0.5.
With the decay lifetime τν , halo mass Mh and efficiency factor all known,
equation (7.4) gives for the luminosity of the halo:
L h = (6.5 × 1042 erg s−1 ) fh f τ−1 . (7.18)
Here we have introduced two dimensionless constants f h and f τ in order to
parametrize the uncertainties in Mh and τν . For the ranges of values given earlier,
these take the values f h = 1.0 ± 0.6 and f τ = 1.0 ± 0.5 respectively. Setting
f h = fτ = 1 gives a halo luminosity of about 2 × 109 L , or less than 5% of the
optical luminosity of the Milky Way (L 0 = 2 × 1010h −2 0 L ), with h 0 as specified
by equation (7.10).
The combined bolometric intensity of decay photons from all the bound
neutrinos out to a redshift z f is given by (2.21) as usual:
zf
dz
Q bound = Q h (7.19)
0 (1 + z)2 H̃ (z)
where
cn 0 L h
Qh ≡ = (2.0 × 10−5 erg s−1 cm−2 )h 20 f h f τ−1 .
H0
The h 0 -dependence in this quantity comes from the fact that we have so far
considered only neutrinos in galaxy halos, whose number density n 0 goes as
h 30 . Since we follow Sciama in adopting the EdS cosmology in this chapter, the
Hubble expansion rate (2.40) is
H̃ (z) = (1 + z)3/2. (7.20)
Putting this into (7.19), we find
Q bound ≈ 25 Q h = (8.2 × 10−6 erg s−1 cm−2 )h 20 fh f τ−1 . (7.21)
(The approximation is good to better than 1% if z f 8.) Here we have neglected
absorption between the galaxies, an issue we will return to shortly. Despite
their size, dark-matter halos in the decaying-neutrino hypothesis are not very
bright. Their combined intensity is about 1% of that of the EBL due to galaxies,
Q ∗ ≈ 3 × 10−4 erg s−1 cm−2 (section 2.3). The main reason for this is the
long neutrino decay lifetime, five orders of magnitude longer than the age of the
galaxies. The ultraviolet decay photons have not had time to contribute as much
to the EBL as their counterparts in the optical part of the spectrum.
7.4 Free-streaming neutrinos

The cosmological density of decaying τ -neutrinos in dark-matter halos is small:
ν,bound = n 0 Mh /ρcrit = (0.068 ± 0.032)h 0. With h 0 as given by (7.10), this
134 Neutrinos
amounts to less than 6% of the total neutrino density, equation (7.7). Therefore,
as expected for hot dark-matter particles, the bulk of the EBL contributions in
the decaying-neutrino scenario come from neutrinos which are distributed on
larger scales. We will refer to these collectively as free-streaming neutrinos,
though some of them may actually be associated with larger scale systems such
as clusters of galaxies. (The distinction is not critical for our purposes, since
we are concerned with the summed contributions of all such neutrinos to the
diffuse background.) Their cosmological density is found using (7.7) as ν,free =
ν − ν,bound = 0.30h −2 0 f f , where the dimensionless constant f f parametrizes
the uncertainties in this quantity and takes the value f f = 1.00 ± 0.05.
To identify ‘sources’ of radiation in this section we follow the same
procedure as with vacuum regions (section 5.4) and axions (section 6.3) and
divide up the Universe into regions of comoving volume V0 = n −1 0 . The mass
of each region is
Mf = ν,free ρcrit,0 V0 = ν,free ρcrit,0 /n 0 . (7.22)
The luminosity of these sources has the same form as equation (7.4) except that
we put Mh → Mf and drop the efficiency factor since the density of intergalactic
hydrogen is too low to absorb a significant fraction of the decay photons within
each region. Thus,
ν,free ρcrit,0 c2
Lf = . (7.23)
2n 0 τν
With the values already mentioned for ν,free and τν , and with ρcrit,0 and n 0 given
by (2.36) and (6.12) respectively, equation (7.23) implies a comoving luminosity
density due to free-streaming neutrinos of
f = n 0 L f = (1.2 × 10−32 erg s−1 cm−3 ) ff fτ−1 . (7.24)
This is high: 0.5h −1
0 times the luminosity density of the Universe, as given by
equation (2.24). Based on this, we may anticipate strong constraints on the
decaying-neutrino hypothesis from observation. Similar conclusions follow from
the bolometric intensity due to free-streaming neutrinos. Replacing L h with L f in
(7.19), we find
2cn 0 L f
Q free = = (1.2 × 10−4 erg s−1 cm−2 )h −1 −1
0 ff f τ . (7.25)
5H0
This is of the same order of magnitude as Q ∗ and goes as h −1 2
0 rather than h 0 .
Taking into account the uncertainties in h 0 , fh , f f and f τ , the bolometric intensity
of bound and free-streaming neutrinos together is
Q = Q bound + Q free = (0.33 ± 0.17)Q ∗ . (7.26)
In principle, then, these particles are capable of producing a background as bright
as that from the galaxies themselves, equation (2.49). The vast majority of their
light comes from free-streaming neutrinos. These are more numerous than their
halo-bound counterparts, and are not appreciably affected by absorption at source.
Intergalactic absorption 135
7.5 Intergalactic absorption

To constrain neutrinos in a more quantitative way, it is necessary to determine
their EBL contributions as a function of wavelength. In the absence of absorption,
this would be done as in previous chapters by putting the source luminosity (which
may be either L h for the galaxy halos or L f for the free-streaming neutrinos) into
the SED (7.3) and substituting the latter into equation (3.6). This gives
zf
−9/2 1 λ0 /(1 + z) − λν 2
Iλ (λ0 ) = Iν (1 + z) exp − dz. (7.27)
0 2 σλ
Here we have used (7.20) for H̃ (z). The numerical value of the prefactor Iν is
given with the help of (7.18) for bound neutrinos and (7.23) for free-streaming
ones as follows:

cn 0 Lh
Iν = √ × (7.28)
32π H0σλ
3 L f

−1
(940 CUs)h 20 f h f τ−1 σ30 (λ0 /λν ) (bound)
= −1 −1 −1
(5280 CUs)h 0 f f f τ σ30 (λ0 /λν ) (free) .
Now, however, we must take into account the fact that decay photons (from both
bound and free-streaming neutrinos) encounter significant amounts of absorbing
material as they travel through the intergalactic medium (IGM) on the way to
our detectors. The wavelength of neutrino decay photons, λν = 860 ± 30 Å, is
just shortward of the Lyman-α line at 912 Å, which means that these photons are
absorbed almost as strongly as they can be by neutral hydrogen (this is, in fact, one
of the prime motivations of the theory). It is also very close to the waveband of
peak extinction by dust. The simplest way to handle both these types of absorption
is to include an opacity term τ (λ0 , z) inside the argument of the exponential, so
that (7.27) is modified to read
zf
−9/2 1 λ0 /(1 + z) − λν 2
Iλ (λ0 ) = Iν (1 + z) exp − − τ (λ0 , z) dz.
0 2 σλ
(7.29)
The optical depth τ (λ0 , z) can be broken into separate terms corresponding to
hydrogen gas and dust along the line of sight:
τ (λ0 , z) = τgas (λ0 , z) + τdust (λ0 , z). (7.30)
Our best information about both of these quantities comes from high-redshift
quasar spectra. The fact that these can be seen at all already puts limits on the
degree of attenuation due to intervening matter. And the shape of the observed
spectra can give us clues about the effects of the absorbing medium in specific
wavebands.
136 Neutrinos
The gas component contains fewer uncertainties and is better understood

at present. Zuo and Phinney [26] have developed a formalism to describe the
absorption due to randomly distributed clouds such as quasar absorption-line
systems and normalized this to the number of Lyman-limit systems at z = 3. We
use their model 1, which gives the highest absorption below λ0 º 2000 Å and
is thus conservative for our purposes. Assuming an EdS cosmology, the optical
depth at λ0 due to neutral hydrogen out to a redshift z is given by
 3/2

 λ0
 τZP
 ln(1 + z) (λ0 λL )
 λL
τgas (λ0 , z) =
 λ0 3/2 1+z (7.31)

 τZP ln [λL < λ0 < λL (1 + z)]

 λ λ0 /λL
[λ0 λL (1 + z)]
L
0
where λL = 912 Å and τZP = 2.0.

Dust is a more complicated and potentially more important issue, and
we pause to discuss this critically before presenting our model. The simplest
possibility, and the one which should be most effective in obscuring a diffuse
signal like that considered here, would be for the dust to be spread uniformly
through intergalactic space. A quantitative estimate of opacity due to a uniform
dusty intergalactic medium has, in fact, been suggested [27] but is regarded as
an extreme upper limit because it would lead to excessive reddening of quasar
spectra [28]. Subsequent discussions have tended to treat intergalactic dust as
clumpy [29], with significant debate about the extent to which such clumps would
redden and/or hide background quasars, possibly helping to explain the observed
‘turnoff’ in quasar population at around z ∼ 3 [30–33]. Most of these models
assume an EdS cosmology. The effects of dust extinction could be enhanced if
m,0 < 1 and/or ,0 > 0 [32], but we ignore this possibility here because
neutrinos (not vacuum energy) are assumed to make up the critical density in the
decaying-neutrino scenario.
We will use a formalism due to Fall and Pei [34] in which dust is associated
with damped Lyα absorbers whose numbers and density profiles are sufficient to
obscure a portion of the light reaching us from z ∼ 3, but not to account fully for
the turnoff in quasar population. Obscuration is calculated based on the column
density of hydrogen in these systems, together with estimates of the dust-to-gas
ratio, and is normalized to the observed quasar luminosity function. The resulting
mean optical depth at λ0 out to redshift z is

z τFP (z )(1 + z ) λ0
τdust (λ0 , z) = ξ dz . (7.32)
0 (1 + m,0 z )1/2 1 + z
Here ξ(λ) is the extinction of light by dust at wavelength λ relative to that in the
B-band (4400 Å). If τFP (z) = constant and ξ(λ) ∝ λ−1 , then τdust is proportional
to λ−1 −1
0 [(1 + z) − 1] or λ0 [(1 + z)
3 2.5 − 1], depending on cosmology [27, 29].
Intergalactic absorption 137
In the more general treatment of Fall and Pei [34], τFP (z) is parametrized as a
function of redshift so that
τFP (z) = τFP (0)(1 + z)δ (7.33)
where τFP (0) and δ are adjustable parameters. Assuming an EdS cosmology
(m,0 = 1), the observational data are consistent with lower limits of τ∗ (0) =
0.005, δ = 0.275 (model A); best-fit values of τ∗ (0) = 0.016, δ = 1.240 (model
B); or upper limits of τ∗ (0) = 0.050, δ = 2.063 (model C). We will use all three
models in what follows.
To calculate the extinction ξ(λ) in the 300–2000 Å range, we use numerical
Mie scattering routines in conjunction with various dust populations. In
performing these calculations, we tacitly assume that intergalactic and interstellar
dust are similar in nature, which is a reasonable assumption that is, of course, very
difficult to test. Many people have constructed dust-grain models that reproduce
the average extinction curve for the diffuse interstellar medium (DISM) at λ >
912 Å [35] but there have been fewer studies at shorter wavelengths. One such
study is that of Martin and Rouleau [36], who extended earlier silicate/graphite
synthetic extinction curves due to Draine and Lee [37] assuming:
(1) two populations of homogeneous spherical dust grains composed of graphite

and silicates respectively;
(2) a power-law size distribution of the form a −3.5 where a is the grain radius;
(3) a range of grain radii from 50–2500 Å; and
(4) solar abundances of carbon and silicon relative to hydrogen [38].
The last of these assumptions is questionable in light of new work which

shows that heavy elements may be far less abundant in the DISM than they are
in the Sun. Snow and Witt [39], for example, report interstellar abundances of
214 × 10−6 /H and 18.6 × 10−6 /H for carbon and silicon respectively. This cuts
earlier values in half and actually makes it difficult for a simple silicate/graphite
model to reproduce the observed DISM extinction curve. We therefore derive
new dust-extinction curves based on the revised abundances. In the interests
of obtaining conservative bounds on the decaying-neutrino hypothesis, we also
consider a range of modified grain populations, looking, in particular, for those
which provide optimal extinction efficiency in the FUV without drifting too far
from the average DISM curve in the optical and NUV bands. We describe the
general characteristics of these models here and show the resulting extinction
curves in figure 7.3; more details are found in [20]).
Our population 1 grain model (figure 7.3, dash-dotted line) assumes the
standard grain model employed by other workers, but uses the new, lower
abundance numbers together with dielectric functions due to Draine [40]. The
shape of the extinction curve provides a reasonable fit to observation at longer
wavelengths (reproducing, for example, the absorption bump at 2175 Å); but
its magnitude is too low, confirming the inadequacies of the old dust model.
138 Neutrinos
Population 4 (fluffy silicates and graphite nanoparticles)

20 Population 3 (graphite nanoparticles)
Population 2 (fluffy silicates)
Population 1 (standard grains)
Population 0 (intermediate)
15 Neutrino line
ξ(λ)
10
200 500 800 1100 1400 1700 2000

λ(Å)
Figure 7.3. The FUV extinction (relative to that in the B-band) produced by five different
dust-grain populations. Standard grains (population 1) produce the least extinction,
while PAH-like carbon nanoparticles (population 3) produce the most extinction near the
decaying neutrino line at 860 Å (vertical line).
Extinction in the vicinity of 860 Å is also weak (with a peak value of τmax ∼
1.0 × 10−21 cm2 H−1 at 770 Å), so that this model is able to ‘hide’ very little of
the light from decaying neutrinos. Insofar as it almost certainly underestimates
the true extent of extinction by dust, this grain model provides a good lower limit
on absorption in the context of the decaying-neutrino hypothesis.
The silicate component of our population 2 grain model is modified along
the lines of the ‘fluffy silicate’ model which has been suggested as a resolution of
the heavy-element abundance crisis in the DISM [41]. We replace the standard
silicates of population 1 by silicate grains with a 45% void fraction, assuming a
silicon abundance of 32.5×10−6/H [20]. We also decrease the size of the graphite
grains (a = 50–250 Å) and reduce the carbon depletion to 60% to provide a better
match to the DISM curve. This mixture provides a better match to the interstellar
data at optical wavelengths, and also shows significantly more FUV extinction
than population 1, with a peak of τmax ∼ 2.0 × 10−21 cm2 H−1 at 750 Å. Results
are shown in figure 7.3 as a short-dashed line.
For population 3, we retain the standard silicates of population 1 but
modify the graphite component as an approximation to the polycyclic aromatic
hydrocarbon (PAH) nanostructures which have recently been proposed as carriers
of the 2175 Å absorption bump [42]. PAH nanostructures are thought to consist
of stacks of molecules such as coronene (C24 H12 ), circumcoronene (C54 H18 ) and
The ultraviolet background 139
larger species in various states of edge hydrogenation. They have been linked
to the 3.4 µm absorption feature in the DISM [43] as well as the extended
red emission in nebular environments [44]. With sizes in the range 7–30 Å,
these structures are much smaller than the canonical graphite grains. Their
dielectric functions, however, go over to that of graphite in the high-frequency
limit [42]. So as an approximation to these particles, we use spherical graphite
grains with extremely small radii (3–150 Å). This greatly increases extinction near
the neutrino-decay peak, giving τmax ∼ 3.9 × 10−21 cm2 H−1 at 730 Å. Results
are shown in figure 7.3 as a dotted line.
Our population 4 grain model, finally, combines the distinctive features of
populations 2 and 3. It has a graphite component made up of nanoparticles (as
in population 3) and a fluffy silicate component with a 45% porosity (like that
of population 2). Results (plotted as a long-dashed line in figure 7.3) are close to
those obtained with population 3. This is because extinction in the FUV waveband
is dominated by small-particle contributions, so that silicates (whatever their void
fraction) are of secondary importance. Absolute extinction rises slightly, with a
peak of τmax ∼ 4.1 × 10−21 cm2 H−1 at 720 Å. The quantity which is of most
interest to us, extinction relative to that in the B-band, drops slightly longward of
800 Å. Therefore it is the population 3 grains which provide us with the highest
value of ξ(λ0 ) near 860 Å, and hence a conservative upper limit on dust absorption
in the context of the decaying-neutrino scenario. Neither the population 3 nor the
population 4 grains fit the average DISM curve as well as those of population 2,
because the Mie scattering formalism cannot accurately reproduce the behaviour
of nanoparticles near the 2175 Å resonance. Their high levels of extinction in the
FUV region, however, suit these grain models for our purpose, which is to set the
strongest possible limits on the decaying-neutrino hypothesis.
7.6 The ultraviolet background
We are now ready to specify the total optical depth (7.30) and hence to evaluate
the intensity integral (7.29). We will do this using three combinations of the
dust models just described, with a view to establishing lower and upper bounds
on the EBL intensity predicted by the theory. A minimum-absorption model is
obtained by combining Fall and Pei’s model A with the extinction curve of the
population 1 (standard) dust grains. At the other end of the spectrum, model C
of Fall and Pei together with the population 3 (nanoparticle) grains provides the
most conservative maximum-absorption model (for λ0 800 Å). Finally, as an
intermediate model, we combine model B of Fall and Pei with the ‘middle-of-the-
road’ extinction curve labelled as population 0 in figure 7.3.
The resulting predictions for the spectral intensity of the FUV background
due to decaying neutrinos are plotted in figure 7.4 (light lines) and compared
with observational limits (bold lines and points). The curves in the bottom half
of this figure refer to EBL contributions from bound neutrinos only, while those
140 Neutrinos
No absorption by intergalactic gas or dust

All neutrinos (maximum)
(nominal)
1000 (minimum)
Bound only (maximum)
E00 (nominal)
(minimum)
K98 P79 LW76
A79 Z82
Fe81 Fi89
We83 Wr92
Iλ ( CUs )
H86
100 WP94
Ma91 E01
MB89
Mu99
10
800 1000 1200 1400 1600 1800 2000 2200

λ0 ( Å )
Figure 7.4. The spectral intensity Iλ of background radiation from decaying neutrinos as
a function of observed wavelength λ0 (light curves), plotted together with observational
upper limits on EBL intensity in the far ultraviolet (points and bold curves). The bottom
four theoretical curves refer to bound neutrinos only, while the top four refer to bound
and free-streaming neutrinos together. The minimum predicted signals consistent with
the theory are combined with the highest possible extinction in the intergalactic medium
and vice versa. The faint dotted lines show the signal that would be seen (in the
maximum-intensity case) if there were no intergalactic extinction at all.
in the top half correspond to contributions from both bound and free-streaming
neutrinos together.
We begin our discussion with the bound neutrinos. The key results are the
three widely-spaced curves in the lower half of the figure, with peak intensities
of about 6, 20 and 80 CUs at λ0 ≈ 900 Å. These are obtained by letting h 0 and
f h take their minimum, nominal and maximum values respectively in (7.29), with
the reverse order applying to f τ . Simultaneously we have adopted the maximum,
intermediate and minimum-absorption models for intergalactic dust, as described
earlier. Thus the highest-intensity model is paired with the lowest possible dust
extinction and vice versa. (The faint dotted line appended to the highest-intensity
curve is included for comparison purposes and shows the effects of neglecting
absorption by dust and gas altogether.) These curves should be seen as extreme
upper and lower bounds on the theoretical intensity of EBL contributions from
decaying neutrinos in galaxy halos.
They are best compared with an experimental measurement by Martin and

Bowyer in 1989 [45], labelled ‘MB89’ in figure 7.4. These authors used data from
a rocket-borne imaging camera to search for small-scale fluctuations in the FUV
EBL, and deduced from this that the combined light of external galaxies (and
their associated halos) reaches the Milky Way with an intensity of 16–52 CUs
over 1350–1900 Å. There is now some doubt as to whether this was really an
extragalactic signal, and indeed whether it is feasible to detect such a signal at all,
given the brightness and fluctuations of the galactic foreground in this waveband
[46]. Viable or not, however, it is of interest to see what a detection of this order
would mean for the decaying-neutrino hypothesis. Figure 7.4 shows that it would
constrain the theory only weakly. The expected signal in this waveband lies below
20 CUs in even the most optimistic scenario where signal strength is highest and
absorption is weakest. In the nominal ‘best-fit’ scenario this drops to less than
7 CUs. As noted already (section 7.3), the low intensity of the background light
from decaying neutrinos in galactic halos (as compared to that from the galaxies
themselves) is due primarily to the long neutrino decay lifetime. In order to place
significant constraints on the theory, one needs the stronger signal which comes
from free-streaming, as well as bound neutrinos. This, in turn, requires limits on
the intensity of the total background rather than that associated with fluctuations.
The curves in the upper half of figure 7.4 (with peak intensities of about 300,
700 and 2000 CUs at λ0 ≈ 900 Å) represent the combined EBL contributions
from all decaying neutrinos. These are obtained by letting f h and f f take their
minimum, nominal and maximum values respectively in (7.29), with the reverse
order applying to f τ as well as h 0 (the latter change being due to the fact that the
dominant free-streaming contribution goes as h −1 2
0 rather than h 0 ). Simultaneously
we have adopted the maximum-, intermediate- and minimum-absorption models
for intergalactic dust, as before. Intensity is greatly reduced in the maximum-
absorption case (unbroken line): by 11% at 900 Å, 53% at 1400 Å and 86% at
1900 Å. The bulk of this reduction is due to dust, especially at longer wavelengths
where most of the light originates at high redshifts. Comparable reduction factors
in the intermediate-absorption case (short-dashed line) are 9% at 900 Å, 28% at
1400 Å and 45% at 1900 Å. In the minimum-absorption case (long-dashed line),
dust becomes less important than gas at shorter wavelengths and the intensity is
reduced by a total of 9% at 900 Å, 21% at 1400 Å and 31% at 1900 Å. These
three curves cover the full range of theoretical EBL intensities in the context of
the decaying-neutrino scenario. For comparison we show also the intensity that
would be observed in the maximum-intensity case if there were no absorption by
intergalactic dust or neutral hydrogen at all (faint dotted line).
The most conservative constraints on the theory are obtained by comparing
the minimum-intensity theoretical EBL contributions (unbroken line) with
observational upper limits on total EBL intensity. Most of the limits which have
been reported to date over the FUV waveband are represented in figure 7.4, and we
describe these briefly here. They can be usefully divided into two groups: those
above and below the Lyman α-line at 1216 Å. At the longest wavelengths are two
142 Neutrinos
more points from the analysis of OAO-2 satellite data by Lillie and Witt ([47];
labelled ‘LW76’ in figure 7.4) which we have already encountered in chapter 3.
Nearby is an upper limit from the Russian Prognoz satellite by Zvereva et al
([48]; ‘Z82’). Considerably stronger broadband limits have come from rocket
experiments by Paresce et al ([49]; ‘P79’), Anderson et al ([50]; ‘A79’) and
Feldman et al ([51]; ‘Fe81’), as well as an analysis of data from the Solrad-11
spacecraft by Weller ([52]; ‘We83’).
A number of studies have proceeded by establishing a correlation between
background intensity and the column density of neutral hydrogen inside the
Milky Way, and then extrapolating this out to zero column density to obtain the
presumed extragalactic component. Martin et al [53] applied this method to data
taken by the Berkeley UVX experiment, setting an upper limit of 110 CUs on
the intensity of any unidentified EBL contributions over 1400–1900 Å (‘Ma91’).
The correlation method is subject to uncertainties involving the true extent of
scattering by dust, as well as absorption by ionized and molecular hydrogen at
high galactic latitudes. Henry [22] and Henry and Murthy [55] approach these
issues differently and raise the upper limit on background intensity to 400 CUs
over 1216–3200 Å. A good indication of the complexity of the problem is found
at 1500 Å, where Fix et al [56] used data from the DE-1 satellite to identify an
isotropic background flux of 530 ± 80 CUs (‘Fi89’), the highest value reported
so far. The same data were subsequently reanalysed by Wright [54] who found a
much lower best-fit value of 45 CUs, with a conservative upper limit of 500 CUs
(‘Wr92’). The former would rule out the decaying-neutrino hypothesis, while the
latter does not constrain it at all. A third treatment of the same data has led to an
intermediate result of 300 ± 80 CUs ([57]; ‘WP94’).
Limits on the FUV background shortward of Lyα have been even more
controversial. Several studies have been based on data from the Voyager 2
ultraviolet spectrograph, beginning with that of Holberg [58], who obtained limits
between 100 and 200 CUs over 500–1100 Å (labelled ‘H86’ in figure 7.4). A
reanalysis of the data over 912–1100 Å by Murthy et al [59] led to similar
numbers. In a subsequent reanalysis, however, Murthy et al [60] tightened this
bound to 30 CUs over the same waveband (‘Mu99’). The statistical validity of
these results has been vigorously debated [61, 62], with a second group asserting
that the original data do not justify a limit smaller than 570 CUs (‘E00’). Of
these Voyager-based limits, the strongest (‘Mu99’) is wholly incompatible with
the decaying-neutrino hypothesis, while the weakest (‘E00’) puts only a modest
constraint on the theory. Two new experiments have yielded results midway
between these extremes: the DUVE orbital spectrometer [63] and the EURD
spectrograph aboard the Spanish MINISAT 01 [64]. Upper limits on continuum
emission from the former instrument are 310 CUs over 980–1020 Å and 440 CUs
over 1030–1060 Å (‘K98’), while the latter has produced upper bounds of
280 CUs at 920 Å and 450 CUs at 1000 Å (‘E01’).
What do these observational data imply for the decaying-neutrino
hypothesis? Longward of Lyα, figure 7.4 shows that they span very nearly the
same parameter space as the minimum and maximum-intensity predictions of the

theory (unbroken and long-dashed lines). Most stringent are Weller’s Solrad-11
result (‘We83’) and the correlation-method constraint of Martin et al (‘Ma91’).
Taken on their own, these data constrain the decaying-neutrino hypothesis rather
severely but do not rule it out. Absorption (by dust in particular) plays a critical
role in reducing the strength of the signal.
Shortward of Lyα, most of the signal originates nearby and intergalactic
absorption is far less important. Ambiguity here comes rather from the spread
in reported limits which, in turn, reflects the formidable experimental challenges
in this part of the spectrum. Nevertheless it is clear that both the Voyager-
based limits of Holberg (‘H86’) and Murthy et al (‘Mu99’), as well as the new
EURD measurement at 920 Å (‘E01’) are incompatible with the theory. These
upper bounds are violated by the weakest predicted signal, which assumes the
strongest possible extinction (unbroken line). The easiest way to reconcile theory
with observation is to increase the neutrino decay lifetime. If we require that
Ith < Iobs , then the previously mentioned EURD measurement (‘E01’) implies
a lower bound of τν > 3 × 1023 s. This rises to (5 ± 3) × 1023 s and
(26 ± 10) × 1023 s for the Voyager limits (‘H86’ and ‘Mu99’ respectively).
All these numbers lie outside the range of lifetimes required in the decaying-
neutrino scenario, τν = (2 ± 1) × 1023 s. The DUVE constraint (‘K98’) is more
forgiving but still pushes the theory to the edge of its available parameter space.
Taken together, these data may safely be said to exclude the decaying-neutrino
hypothesis. This conclusion is in accord with current thinking on the value of
Hubble’s constant (section 4.3) and structure formation (section 4.5), as well as
more detailed analysis of the EURD data [65].
These limits would be weakened (by a factor of up to nearly one-third) if
the value of Hubble’s constant h 0 were allowed to exceed 0.57 ± 0.01, since
the dominant free-streaming contributions to Iλ (λ0 ) go as h −10 . (This would be
only partly offset by the fact that the bound ones are proportional to h 20 .) A higher
expansion rate would, however, exacerbate problems with structure formation and
the age of the Universe, the more so because the dark matter in this theory is
hot. It would also mean sacrificing the critical density of neutrinos. Another
possibility would be to consider neutrinos of lower rest mass. This would,
however, entail a proportionate drop in the energy of the decay photons. The
latter would, in fact, have to drop below the Lyman or hydrogen-ionizing limit,
thus removing the whole motivation for the proposed neutrinos in the first place.
Similar considerations apply to neutrinos with longer decay lifetimes.
Our conclusions, then, are as follows. Neutrinos with rest masses and
decay lifetimes as specified by the decaying-neutrino scenario produce levels
of ultraviolet background radiation very close to and, in several cases, above
experimental upper limits on the intensity of the EBL. At wavelengths longer than
1200 Å, where intergalactic absorption is most effective, the theory is marginally
compatible with observation—if one adopts the upper limits on dust density
consistent with quasar obscuration; and if the dust grains are extremely small. At
144 Neutrinos
wavelengths in the range 900–1200 Å, predicted intensities are either comparable
to or higher than those actually seen. Thus, while there is now good experimental
evidence that some of the dark matter is provided by massive neutrinos, the light
of the night sky tells us that these particles cannot have the rest masses and decay
lifetimes attributed to them in the decaying-neutrino hypothesis.
References
[1] Cowsik R 1977 Phys. Rev. Lett. 39 784
[2] de Rujula A and Glashow S L 1980 Phys. Rev. Lett. 45 942
[3] Pal P B and Wolfenstein L 1982 Phys. Rev. D 25 766
[4] Bowyer S et al 1995 Phys. Rev. D 52 3214
[5] Melott A L and Sciama D W 1981 Phys. Rev. Lett. 46 1369
[6] Sciama D W and Melott A L 1982 Phys. Rev. D 25 2214
[7] Melott A L, McKay D W and Ralston J P 1988 Astrophys. J. 324 L43
[8] Sciama D W 1990 Astrophys. J. 364 549
[9] Sciama D W 1993 Modern Cosmology and the Dark Matter Problem (Cambridge:
Cambridge University Press)
[10] Sciama D W 1997 Mon. Not. R. Astron. Soc. 289 945
[11] Fabian A, Naylor T and Sciama D 1991 Mon. Not. R. Astron. Soc. 249 21
[12] Davidsen A F et al 1991 Nature 351 128
[13] Sciama D W 1993 Pub. Astron. Soc. Pac. 105 102
[14] Stecker F W 1980 Phys. Rev. Lett. 45 1460
[15] Kimble R, Bowyer S and Jakobsen P 1981 Phys. Rev. Lett. 46 80
[16] Sciama D W 1991 The Early Observable Universe from Diffuse Backgrounds ed
B Rocca-Volmerange, J M Deharveng and J Van Tran Thanh (Paris: Edition
Frontières) p 127
[17] Overduin J M, Wesson P S and Bowyer S 1993 Astrophys. J. 404 460
[18] Dodelson S and Jubas J M 1994 Mon. Not. R. Astron. Soc. 266 886
[20] Overduin J M, Seahra S S, Duley W W and Wesson P S 1999 Astron. Astrophys. 349
317
[21] Bowyer S 1991 Ann. Rev. Astron. Astrophys. 29 59
[22] Henry R C 1991 Ann. Rev. Astron. Astrophys. 29 89
[23] Salucci P and Sciama D W 1990 Mon. Not. R. Astron. Soc. 244 9P
[24] Gates E I, Gyuk G and Turner M S 1995 Astrophys. J. 449 L123
[26] Zuo L and Phinney E S 1993 Astrophys. J. 418 28
[27] Ostriker J P and Cowie L L 1981 Astrophys. J. 243 L127
[28] Wright E L 1981 Astrophys. J. 250 1
[29] Ostriker J P and Heisler J 1984 Astrophys. J. 278 1
[31] Wright E L and Malkan M A 1987 Bull. Am. Astron. Soc. 19 699
[32] Heisler J and Ostriker J P 1988 Astrophys. J. 332 543
[34] Fall S M and Pei Y C 1993 Astrophys. J. 402 479
[35] Mathis J S 1990 Ann. Rev. Astron. Astrophys. 28 37
[36] Martin P G and Rouleau F 1991 Extreme Ultraviolet Astronomy ed R F Malina and
S Bowyer (New York: Pergamon) p 341
[37] Draine B T and Lee H M 1984 Astrophys. J. 285 89
[38] Meyer J-P 1979 Les Elements et leurs Isotopes dan L’univers (Université de Liège)
p 153
[39] Snow T P and Witt A N 1996 Astrophys. J. 468 L65
[40] Draine B T 1995 http://www.astro.princeton.edu/∼draine/dust/
[41] Mathis J S 1996 Astrophys. J. 472 643
[42] Duley W W and Seahra S 1998 Astrophys. J. 507 874
[43] Duley W W and Seahra S S 1999 Astrophys. J. 522 L129
[44] Seahra S S and Duley W W 1999 Astrophys. J. 520 719
[45] Martin C and Bowyer S 1989 Astrophys. J. 338 677
[46] Sasseen T P, Lampton M, Bowyer S and Wu X 1995 Astrophys. J. 447 630
[48] Zvereva A M et al 1982 Astron. Astrophys. 116 312
[49] Paresce F, Margon B, Bowyer S and Lampton M 1979 Astrophys. J. 230 304
[50] Anderson R C et al 1979 Astrophys. J. 234 415
[51] Feldman P D, Brune W H and Henry R C 1991 Astrophys. J. 249 L51
[52] Weller C S 1983 Astrophys. J. 268 899
[53] Martin C, Hurwitz M and Bowyer S 1991 Astrophys. J. 379 549
[55] Henry R C and Murthy J 1993 Astrophys. J. 418 L17
[56] Fix J D, Craven J D and Frank L A 1989 Astrophys. J. 345 203
[57] Witt A N and Petersohn J K 1994 The First Symposium on the Infrared Cirrus
and Diffuse Interstellar Clouds (Astronomical Society of the Pacific Conference
Series, Volume 58) ed R M Cutri and W B Latter (San Francisco, CA: ASP) p 91
[58] Holberg J B 1986 Astrophys. J. 311 969
[59] Murthy J, Henry R C and Holberg J B 1991 Astrophys. J. 383 198
[60] Murthy J et al 1999 Astrophys. J. 522 904
[61] Edelstein J, Bowyer S and Lampton M 2000 Astrophys. J. 539 187
[62] Murthy J et al 2001 Astrophys. J. 557 L47
[63] Korpela E J, Bowyer S and Edelstein J 1998 Astrophys. J. 495 317
[64] Edelstein J et al 2001 Astrophys. Space Sci. 276 177
[65] Bowyer S et al 1999 Astrophys. J. 526 10
Chapter 8
Supersymmetric weakly interacting

particles
8.1 The lightest supersymmetric particle

Weakly interacting massive particles (WIMPs) are hypothetical particles whose
interaction strengths are comparable to those of neutrinos, but whose rest masses
exceed those of the baryons. The most widely-discussed examples arise in the
context of supersymmetry (SUSY), which is motivated quite independently of the
dark-matter problem as a theoretical framework for many attempts to unify the
forces of nature. SUSY predicts that, for every known fermion in the standard
model, there exists a new bosonic ‘superpartner’ and vice versa (more than
doubling the number of fundamental degrees of freedom in the simplest models;
see [1] for a review). These superpartners were recognized as potential dark-
matter candidates in the early 1980s by Cabibbo et al [2], Pagels and Primack [3],
Weinberg [4] and others [5–9], with the generic term ‘WIMP’ being coined
shortly thereafter [10].
There is, as yet, no firm experimental evidence for SUSY WIMPs. This
means that their rest energies, if they exist, lie beyond the range currently probed
by accelerators (and, in particular, beyond the rest energies of their standard-
model counterparts). Supersymmetry is, therefore, not an exact symmetry of
nature. The masses of the superpartners, like that of the axion (chapter 6),
are thought to have been generated by a symmetry-breaking event in the early
Universe. Subsequently, as the temperature of the expanding fireball dropped
below their rest energies, heavier species would have dropped out of equilibrium
and begun to disappear by pair annihilation, leaving progressively lighter ones
behind. Eventually, only one massive superpartner would have remained: the
lightest supersymmetric particle (LSP). It is this particle which plays the role
of the WIMP in SUSY theories. Calculations using the Boltzmann equation show
that the collective density of relic LSPs today lies within one or two orders of
magnitude of the suspected cold dark-matter (CDM) density across much of the
parameter space of the simplest SUSY theories [11].
146
Neutralinos 147
There are at least three ways in which SUSY WIMPs could contribute to
the extragalactic background light (EBL). The first is by pair annihilation to
photons. This occurs in even the simplest or minimal SUSY model (MSSM) but
it is a very slow process because it takes place via intermediate loops of charged
particles such as leptons and quarks and their antiparticles. Processes involving
intermediate bound states contribute as well, but these are also suppressed by loop
diagrams and are equally suppressed. (Each vertex in such a diagram weakens
the interaction by one factor of the coupling parameter.) The reason for the
stability of the LSP in the MSSM is an additional new symmetry of nature,
known as R-parity, which is necessary (among other things) to protect the proton
from decaying via intermediate SUSY states. There are also non-minimal SUSY
theories which do not conserve R-parity (and in which the proton can decay).
In these theories, LSPs can contribute to the EBL in two more ways: by direct
decay into photons via loop diagrams; and also indirectly via tree-level decays
to secondary particles which then scatter off pre-existing background photons to
produce a signal.
In this chapter, we will estimate the strength of EBL contributions from
all three kinds of processes (annihilations, one-loop and tree-level decays). The
strongest signals come from decaying LSPs in non-minimal SUSY theories, and
we will find that these particles are strongly constrained by data on background
radiation in the x-ray and γ -ray bands. While our results depend on a number of
theoretical parameters, they imply in general that the LSP, if it decays, must do so
on timescales which are considerably longer than the lifetime of the Universe.
8.2 Neutralinos
The first step is to choose an LSP from the lineup of SUSY candidate particles.
Early workers variously identified this as the photino γ̃ [2], the gravitino g̃ [3],
the sneutrino ν̃ [7] or the selectron ẽ [8]. (SUSY superpartners are denoted by a
tilde and take the same names as their standard-model counterparts, with a prefix
‘s’ for superpartners of fermions and a suffix ‘ino’ for those of bosons.) Many of
these contenders were eliminated by Ellis et al, who showed in 1984 that the LSP
is, in fact, most likely to be a neutralino χ̃, a linear superposition of the photino
γ̃ , the zino Z̃ and two neutral higgsinos h̃ 01 and h̃ 02 [9]. (These are the SUSY
spin- 12 counterparts of the photon, Z 0 and Higgs bosons respectively.) There
are four neutralinos, each a mass eigenstate made up of (in general) different
amounts of photino, zino, etc, although in special cases a neutralino could be
‘mostly photino’, say, or ‘pure zino’. The LSP is, by definition, the lightest
such eigenstate. Accelerator searches place a lower limit on its rest energy which
currently stands at m χ̃ c2 > 46 GeV [12].
In minimal SUSY, the density of neutralinos drops only by way of the (slow)
pair-annihilation process, and it is quite possible for these particles to ‘overclose’
the Universe if their rest energy is too high. This does not change the geometry of
148 Supersymmetric weakly interacting particles
the Universe but rather speeds up its expansion rate, whose square is proportional
to the total matter density from equation (2.33). In such a situation, the Universe
would have reached its present size in too short a time. Lower bounds on the
age of the Universe thus impose an upper bound on the neutralino rest energy
which has been set at m χ̃ c2 º 3200 GeV [13]. Detailed exploration of the
parameter space of minimal SUSY theory shows that this upper limit tightens
in most cases to m χ̃ c2 º 600 GeV [14]. Much recent work is focused on a
slimmed-down version of the MSSM known as the constrained minimal SUSY
model (CMSSM), in which all existing experimental bounds and cosmological
requirements are comfortably met by neutralinos with rest energies in the range
90 GeV º m χ̃ c2 º 400 GeV [15].
Even in its constrained minimal version, SUSY theory contains at least
five adjustable input parameters, making the neutralino a considerably harder
proposition to test than the axion or the massive neutrino. Fortunately, there
are several other ways (besides accelerator searches) to look for these particles.
Because their rest energies are above the temperature at which they decoupled
from the primordial fireball, WIMPs have non-relativistic velocities and ought
to be found predominantly in gravitational potential wells like those of our
own galaxy. They will occasionally scatter against target nuclei in terrestrial
laboratories as the Earth follows the Sun around the Milky Way. Efforts to
detect such scattering events are hampered by both low event rates and the
relatively small amounts of energy deposited. But they could be given away by
the directional dependence of nuclear recoils, or by the seasonal modulation of
the event rate due to the Earth’s motion around the Sun. Both effects are now the
basis for direct detection experiments around the world [16]. Great excitement
followed the announcement by the DAMA team in 2000 of a WIMP signature
with rest energy m χ̃ c2 = 59+17
−14 GeV [17]. Two other groups (CDMS [18] and
EDELWEISS [19]) have, however, been unable to reproduce this result; and it
is generally regarded as a false start but one which is stimulating a great deal of
productive follow-up work.
A second, indirect search strategy is to look for annihilation byproducts
from neutralinos which have collected inside massive bodies. Most attention has
been directed at the possibility of detecting antiprotons from the galactic halo [20]
or neutrinos from the Sun [21] or Earth [22]. The heat generated in the cores of gas
giants like Jupiter or Uranus has also been considered as a potential annihilation
signature [23]. The main challenge in each case lies in separating the signal
from the background noise. Neutrinos from WIMP annihilations in the Sun or
Earth provide perhaps the best chance for a detection because backgrounds for
these cases are small and relatively well understood. In the case of the Earth,
one looks for a flux of neutrino-induced muons which can be distinguished from
the atmospheric background by the fact that they are travelling straight up. The
AMANDA team, whose detectors are buried deep in Antarctic ice, is one of
several which have reported preliminary results in experiments of this kind [24].
After various filters were applied to the first year’s worth of raw data, 15 candidate
Pair annihilation 149
Figure 8.1. Some Feynman diagrams corresponding to the annihilation of two neutralinos
(χ̃), producing a pair of photons (γ ). The process is mediated by fermions (f) and their
supersymmetric counterparts, the sfermions (f̃).
events remained. This was consistent with the expected statistical background of
17 events, leaving only an upper limit on the size of a possible WIMP population
inside the Earth at present.
8.3 Pair annihilation

Pair annihilation into photons (figure 8.1) provides a complementary indirect
search technique. For the range of neutralino rest energies under consideration
here (50 GeV º m χ̃ c2 º 1000 GeV), the photons so produced lie in the
γ -ray portion of the spectrum. Many workers, beginning with Sciama [8] and
Silk and Srednicki [20], have studied the possibility of γ -rays from SUSY
WIMP annihilations in the halo of the Milky Way, which gives the strongest
signal. Prognoses for detection have ranged from very optimistic [25] to
very pessimistic [26]; converging gradually to the conclusion that neutralino-
annihilation contributions would be at or somewhat below the level of the galactic
background, and possibly distinguishable from it by their spectral shape [27–29].
Uncertainties are considerable, arising not only from weak signal-to-noise levels
but also from factors such as the assumed distribution of WIMPs in the halo
and scattering processes along the line of sight. Recent studies have focused
on possible enhancements of the signal such as might occur in the presence of a
high-density core [30], a flattened halo [31], a very extended singular halo [32]
or a massive black hole at the galactic centre [33]. It has also been pointed out
that the large effective area of atmospheric Čerenkov detectors could compensate
for the weaker signal from neutralinos at the upper end of the mass range [34].
This, however, appears unlikely to raise the expected signal above the galactic
background [35]. Other authors have carried the search away from the halo of
the Milky Way toward systems such as dwarf spheroidal galaxies [36], the Large
Magellanic Cloud [37] and the giant elliptical galaxy M87 in Virgo [38].
The possibility of neutralino-annihilation contributions to the diffuse
extragalactic background, rather than the signal from localized concentrations
of dark matter, has received far less attention. First to apply the problem to SUSY
WIMPs were Cabibbo et al [2] who assumed a WIMP rest energy (10–30 eV)
which we now know is far too low. Like the decaying neutrino (chapter 7), this
would have produced a background in the ultraviolet. It is excluded, however,
by an argument (sometimes known as the Lee–Weinberg bound) which restricts
WIMPs to rest energies above 2 GeV [6]. EBL contributions from SUSY
WIMPs in this range were first estimated by Silk and Srednicki [20]. Their
conclusion, and those of most workers who have followed them [39–41], is that
neutralino annihilations would be responsible for no more than a small fraction
of the observed γ -ray background. Here we review this argument, reversing our
usual procedure and attempting to set a reasonably conservative upper limit on
neutralino contributions to the EBL.
As usual, we take for our sources of background radiation the galactic dark-
matter halos with comoving number density n 0 . Each neutralino pair annihilates
into two monochromatic photons of energy E γ ≈ m χ̃ c2 [29]. We model this with
the Gaussian SED (3.19)

L h,ann 1 λ − λann 2
F(λ) = √ exp − . (8.1)
2πσλ 2 σλ
Photon wavelengths are distributed normally about the peak wavelength:
λann = hc/m χ̃ c2 = (1.2 × 10−6 Å)m −1

10 (8.2)
where m 10 ≡ m χ̃ c2 /(10 GeV) is the neutralino rest energy in units of 10 GeV.

The standard deviation σγ can be related to the velocity dispersion of bound dark-
matter particles as in previous chapters, so that σλ = 2(vc /c)λann . With vc ∼
220 km s−1 and m 10 ∼ 1 this is of order ∼10−9 Å. For convenience we specify
this from now on in terms of a dimensionless parameter σ9 ≡ σλ /(10−9 Å).
The luminosity L h,ann of the halo is given by
L h,ann = 2m χ̃ c2 σ vγ γ (8.3)

2
where σ vγ γ is the annihilation cross section, ≡ m −2 χ̃ ρχ̃ (r )4πr 2 dr and
ρχ̃ (r ) is the mass density distribution of neutralinos in the halo [32]. For the
cross section we turn to the particle-physics literature. Berezinsky et al [42] have
calculated this quantity as σ vγ γ ≈ aγ γ (for non-relativistic neutralinos) where
2 c 3 α 4 (m 4
χ̃ c )
2 2
Z 11
aγ γ = (45 + 48y + 139y 2)2 . (8.4)
36 π(m f˜ c 2 )4 sin θW
Here α is the fine structure constant, m f˜ is the mass of an intermediate sfermion,

y ≡ (Z 12/Z 11 ) tan θW , θW is the weak mixing angle and Z i j are elements of
the real orthogonal matrix which diagonalizes the neutralino mass matrix. In
particular, the ‘pure photino’ case is specified by Z 11 = sin θW , y = 1 and
the ‘pure zino’ by Z 11 = cos θW , y = − tan2 θW . Collecting these expressions

together and parametrizing the sfermion rest energy by m̃ 10 ≡ m f˜ /10 GeV, we
obtain:
σ vγ γ = (8 × 10−27 cm3 s−1 ) f χ m 210 m̃ −4
10 . (8.5)
Here fχ (= 1 for photinos, 0.4 for zinos) is a dimensionless quantity whose value
parametrizes the makeup of the neutralino.
Since we attempt in this section to set an upper limit on EBL contributions
from neutralino annihilations, we take fχ ≈ 1 (the photino case). In the same
spirit, we would like to use lower limits for the sfermion mass m̃ 10 . It is important
to estimate this quantity accurately since the cross section goes as m̃ −4 10 . Giudice
and Griest [43] have made a detailed study of photino annihilations and find a
lower limit on m̃ 10 as a function of m 10 , assuming that photinos provide at least
0.025h −2
0 of the critical density. Over the range 0.1 m 10 4, this lower
limit is empirically well fit by a function of the form m̃ 10 ≈ 4m 10 0.3
. If this holds
over our broader range of masses, then we obtain an upper limit on the neutralino
annihilation cross section of σ vγ γ º (3×10−29 cm3 s−1 )m 10 0.8
. This expression
gives results which are about an order of magnitude higher than the cross sections
quoted by Gao et al [40] for photinos with m 10 ≈ 1.
For the WIMP density distribution ρχ (r ) we adopt the simple and widely-
used isothermal model [29]:

a 2 + r
2
ρχ (r ) = ρ . (8.6)
a2 + r 2
Here ρ = 5 × 10−25 g cm−3 is the approximate dark-matter density in the Solar

vicinity, assuming a spherical halo [44], r = 8 kpc is the distance of the Sun
from the galactic centre [45] and a = (2–20) kpc is a core radius. To fix this latter
parameter, we can integrate (8.6) over volume to obtain total halo mass Mh (r )
inside radius r :
a r
Mh (r ) = 4πρ0 r (a 2 + r
2
) 1− tan−1 . (8.7)
r a
Observations of the motions of galactic satellites imply that the total mass inside
50 kpc is about 5 × 1011 M [46]. This in (8.7) implies a = 9 kpc, which we
consequently adopt. The maximum extent of the halo is not well constrained
observationally but can be specified if we suppose that Mh = (2 ± 1) × 1012 M
as in (6.15). Equation (8.7) then gives a radius rh = (170 ± 80) kpc. The
cosmological density of WIMPs in galactic dark-matter halos adds up to h =
n 0 Mh /ρcrit,0 = (0.07 ± 0.04)h 0.
If there are no other sources of CDM, then the total matter density is
m,0 = h +bar ≈ 0.1h 0 and the observed flatness of the Universe (section 4.6)
implies a strongly vacuum-dominated cosmology. While we use this as a lower
limit on WIMP contributions to the dark matter in subsequent sections, it is quite
possible that CDM also exists in larger-scale regions such as galaxy clusters. To
take this into account in a general way, we define a cosmological enhancement

factor f c ≡ (m,0 − bar )/h representing the added contributions from
WIMPs outside galactic halos (or perhaps in halos which extend far enough to
fill the space between galaxies [47]). This takes the value f c = 1 for the most
conservative case just described but rises to f c = (4 ± 2)h −1
0 in the CDM model
with m,0 = 0.3, and (14 ± 7)h −1 0 in the EdS model with m,0 = 1.
With ρχ̃ (r ) known, we are in a position to calculate the quantity , which
becomes:

2πρ 2 (a 2 + r 2 )2 (rh /a)
= 2 tan−1 rh − . (8.8)
m χ̃ a a 1 + (rh /a)2
Using the values for ρ , r and a specified before and setting rh = 250 kpc to
get an upper limit, we find that (5 × 1065 cm−3 )m −210 . Putting this result
together with the cross section (8.5) into (8.3), we obtain:
L h,ann (1 × 1038 erg s−1 ) fχ m 10 m̃ −4

10 . (8.9)
Inserting Giudice and Griest’s [43] lower limit on the sfermion mass m̃ 10 (as
empirically fit before), we find that (8.9) gives an upper limit on halo luminosity
of L h,ann (5 × 1035 erg s−1 ) fχ m −0.2
10 . Higher estimates can be found in the
literature [48] but these assume a singular halo whose density drops off as only
ρχ̃ (r ) ∝ r −1.8 and extends out to a very large halo radius rh = 4.2h −1
0 Mpc. For
a standard isothermal distribution of the form (8.6), our results confirm that halo
luminosity due to neutralino annihilations alone is very low, amounting to less
than 10−8 times the total bolometric luminosity of the Milky Way.
The combined bolometric intensity of neutralino annihilations between
redshift z f and the present is given by substituting the comoving number density
n 0 and luminosity L h,ann into equation (2.20) to give
zf
dz
Q = Q χ̃ ,ann (8.10)
0 (1 + z) [m,0 (1 + z) + (1 − m,0 )]
2 3 1/2
where Q χ̃,ann = (cn 0 L h,ann f c )/H0 and we have assumed spatial flatness. With
values for all these parameters as specified earlier, we find

 −12 erg s−1 cm−2 )h 2 f m −0.2 (
 (1 × 10 0 χ 10 m,0 = 0.1h 0 )
Q = (3 × 10−12 erg s−1 cm−2 )h 0 fχ m −0.2 ( m,0 = 0.3) (8.11)

 −11 −1 −2
10
−0.2
(1 × 10 erg s cm )h 0 fχ m 10 (m,0 = 1).
Here we have set z f = 30 (larger values do not substantially increase the value
of Q) and used values of f c = 1, 4h −1 −1
0 and 20h 0 respectively. The effects of
the cosmological enhancement factor f c are partially offset in (8.10) by the fact
that a Universe with higher matter density m,0 is younger, and hence contains
less background light in general. Even the highest value of Q given in (8.11) is
0.1
0.01
Iλ ( CUs )
0.001
N80
0.0001
SAS−2 (TF82)
COMPTEL (K96)
1e−05 EGRET (S98)
mχ c2 = 1000 GeV
mχ c2 = 300 GeV
1e−06
mχ c2 = 100 GeV
mχ c2 = 30 GeV
1e−07 mχ c2 = 10 GeV
1e−08 1e−07 1e−06 1e−05 0.0001 0.001 0.01
λ0 ( Å )
Figure 8.2. The spectral intensity of the diffuse γ -ray background due to neutralino
annihilations (lower left), compared with observational limits from high-altitude balloon
experiments (N80), as the SAS-2 spacecraft and the COMPTEL and EGRET instruments
aboard the Compton Gamma-Ray Observatory. The three plotted curves for each value of
m χ̃ c2 depend on the total density of neutralinos: galaxy halos only (m,0 = 0.1h 0 ; bold
lines), CDM model (m,0 = 0.3; medium lines) or EdS model (m,0 = 1; light lines).
negligible in comparison to the approximate intensity (2.25) of the EBL due to

ordinary galaxies.
The total spectral intensity of annihilating neutralinos is found by
substituting the SED (8.1) into (3.6) to give

1 λ0 /(1 + z) − λann 2
zf exp − 2 σλ
dz
Iλ (λ0 ) = Iχ̃ ,ann . (8.12)
0 (1 + z) [m,0 (1 + z) + 1 − m,0 ]
3 3 1/2
For a typical neutralino with m 10 ≈ 10 the annihilation spectrum peaks near

λ0 ≈ 10−7 Å. The dimensional prefactor reads as

n 0 L h,ann f c λ0
Iχ̃ ,ann = √
32π 3 h H0 σλ

0.8 −1 λ0
= (0.0002 CUs)h 0 f χ f c m 10 σ9
2
. (8.13)
10−7 Å
Figure 8.3. Feynman diagrams corresponding to one-loop decays of the neutralino (χ̃)
into a neutrino (here ντ ) and a photon (γ ). The process can be mediated by the W boson
and a τ muon, or by the τ and its supersymmetric counterpart (τ̃ ).
Here we have divided through by the photon energy hc/λ0 to put results into
continuum units or CUs as usual (section 3.3). Equation (8.12) gives the
combined intensity of radiation from neutralino annihilations, emitted at various
wavelengths and redshifted by various amounts but observed at wavelength λ0 .
Results are plotted in figure 8.2 together with observational constraints. We defer
discussion of this plot (and the data) to section 8.7, where they are compared with
the other WIMP results for this chapter.
8.4 One-loop decays

Figure 8.2 confirms that neutralino annihilations do not produce enough
background radiation to lead to useful constraints on these particles. We therefore
turn in the rest of this chapter to non-minimal SUSY theories in which R-parity is
not necessarily conserved and the LSP (in this case the neutralino) can decay. The
cosmological consequences of R-parity breaking have been reviewed by Bouquet
and Salati [49]. There is only one direct decay mode into photons:
χ̃ → ν + γ . (8.14)
Feynman diagrams for this process are shown in figure 8.3. Because these decays
occur via loop diagrams, they are again subdominant. We consider theories in
which R-parity breaking occurs spontaneously. This introduces a scalar sneutrino
with a non-zero vacuum expectation value vR ≡ ν̃τ R , as discussed by Masiero
and Valle [50]. Neutralino decays into photons could be detectable if m χ̃ and vR
are large [51].
The photons produced in this way are again monochromatic, with E γ =
1 2
2 χ̃ . In fact the SED here is the same as (8.1) except that the peak wavelength
m c
is doubled, λloop = 2hc/m χ̃ c2 = (2.5 × 10−6 Å) m −1 10 . The only parameter that
needs to be recalculated is the halo luminosity L h . For one-loop neutralino decays
of lifetime τχ̃ , this takes the form:
Nχ̃ bγ E γ bγ Mh c2
L h,loop = = . (8.15)
τχ̃ 2τχ̃
One-loop decays 155
Here Nχ̃ = Mh /m χ̃ is the number of neutralinos in the halo and bγ is the

branching ratio or fraction of neutralinos that decay into photons. This is
estimated by Berezinsky et al [51] as
bγ ≈ 10−9 f R2 m 210 (8.16)
where the new parameter f R ≡ vR /(100 GeV). The requirement that SUSY
WIMPs do not carry too much energy out of stellar cores implies that f R is of
order ten or more [50]. We take f R > 1 as a lower limit.
For halo mass we adopt Mh = (2 ± 1) × 1012 M as usual, with rh =
(170 ± 80) kpc from the discussion following (8.7). As in the previous section,
we parametrize our lack of certainty about the distribution of neutralinos on
larger scales with the cosmological enhancement factor f c . Collecting these
results together and expressing the decay lifetime in dimensionless form as
f τ ≡ τχ̃ /(1 Gyr), we obtain for the luminosity of one-loop neutralino decays
in the halo:
L h,loop = (6 × 1040 erg s−1 )m 210 f R2 f τ−1 . (8.17)
With m 10 ∼ f R ∼ f τ ∼ 1, equation (8.17) gives L h,loop ∼ 2 × 107 L . This is
considerably brighter than the halo luminosity due to neutralino annihilations in
minimal SUSY models but still amounts to less than 10−3 times the bolometric
luminosity of the Milky Way.
The combined bolometric intensity is found as in the previous section, but
with L h,ann in (8.10) replaced by L h,loop so that
 −7 −1 −2 2 2 2 −1
 (1 × 10 erg s cm )h 0 m 10 f R fτ (m,0 = 0.1h 0 )
Q = (4 × 10 erg s cm−2 )h 0 m 210 f R2 fτ−1
−7 −1 (m,0 = 0.3) (8.18)

(2 × 10−6 erg s−1 cm−2 )h 0 m 210 f R2 fτ−1 (m,0 = 1).
This is again small. However, we see that massive (m 10 10) neutralinos which
provide close to the critical density (m,0 ∼ 1) and decay on timescales of order
1 Gyr or less ( f τ 1) could, in principle, rival the intensity of the conventional
EBL.
To obtain more quantitative constraints, we turn to spectral intensity. This
is given by equation (8.12) as before, except that the dimensional prefactor Iχ̃ ,ann
must be replaced by

n 0 L h,loop f c λ0
Iχ̃ ,loop = √
32π 3 h H0 σλ

λ0
= (30 CUs)h 20 m 310 fR2 f τ−1 f c σ9−1 . (8.19)
10−7 Å
Results are plotted in figure 8.4 for neutralino rest energies 1 m 10 100.
While their bolometric intensity is low, these particles are capable of significant
EBL contributions in narrow portions of the γ -ray background. To keep the
0.1
Iλ ( CUs )
N80
0.01 SAS−2 (TF82)
COMPTEL (K96)
EGRET (S98)
mχ c2 = 1000 GeV
mχ c2 = 300 GeV
0.001 2
mχ c = 100 GeV
2
mχ c = 30 GeV
mχ c2 = 10 GeV
1e−08 1e−07 1e−06 1e−05 0.0001 0.001
λ0 ( Å )
Figure 8.4. The spectral intensity of the diffuse γ -ray background due to neutralino
one-loop decays (lower left), compared with observational upper limits from high-altitude
balloon experiments (filled dots), SAS-2, EGRET and COMPTEL. The three plotted
curves for each value of m χ̃ c2 correspond to models with m,0 = 0.1h 0 (heavy lines),
m,0 = 0.3 (medium lines) and m,0 = 1 (light lines). For clarity we have assumed
decay lifetimes in each case such that the highest theoretical intensities lie just under the
observational constraints.
diagram from becoming too cluttered, we have assumed values of f τ such that
the highest predicted intensity in each case stays just below the EGRET limits.
Numerically, this corresponds to lower bounds on the decay lifetime τχ̃ between
100 Gyr (for m χ̃ c2 = 10 GeV) and 105 Gyr (for m χ̃ c2 = 300 GeV). For
rest energies at the upper end of this range, these limits are probably optimistic
because the decay photons are energetic enough to undergo pair production on
cosmic microwave background (CMB) photons. Some of the decay photons,
in other words, do not reach us from cosmological distances but are instead
reprocessed into lower-energy radiation along the way. As we show in the next
section, however, stronger limits arise from a different process in any case. We
defer further discussion of figure 8.4 (including the observational data and the
limits on neutralino lifetime that follow from them) to section 8.7.
Tree-level decays 157
Figure 8.5. The Feynman diagrams corresponding to tree-level decays of the neutralino
(χ̃) into a neutrino (ν) and a lepton–antilepton pair (here, the electron and positron). The
process can be mediated by the W or Z boson.
8.5 Tree-level decays

The dominant decay processes for the LSP neutralino in non-minimal SUSY
(assuming spontaneously broken R-parity) are tree-level decays to leptons and
neutrinos, as follows:
χ̃ → + + − + ν . (8.20)
Of particular interest is the case in which = e; Feynman diagrams for this decay
are shown in figure 8.5. Although these processes do not contribute directly to the
EBL, they do so indirectly, because the high-energy electrons undergo inverse
Compton scattering (ICS) off the CMB photons via e + γcmb → e + γ . This
gives rise to a flux of high-energy photons which can be at least as important
as those from the direct (one-loop) neutralino decays considered in the previous
section [52].
The spectrum of photons produced in this way depends on the rest energy
of the original neutralino. We consider first the case m 10 º 10, which is more
or less pure ICS. The input (‘zero-generation’) electrons are monoenergetic, but
after multiple scatterings they are distributed as E −2 [53]. From this the spectrum
of outgoing photons can be calculated [54] and shown to take the following form:

−3/2 (E
Nics (E) ∝ E E max ) (8.21)
0 (E > E max )
where 2
4 Ee
E max = E cmb .
3 m e c2
Here E e = 13 m χ̃ c2 = (3.3 GeV)m 10 is the energy of the input electrons,
m e is their rest mass and E cmb = 2.7kTcmb is the mean energy of the CMB
photons. Using m e c2 = 0.51 MeV and Tcmb = 2.7 K, and allowing for
decays at arbitrary redshift z after Berezinsky [54], we obtain the expression
E max (z) = (36 keV)m 210 (1 + z)−1 .
The halo SED may be determined as a function of wavelength by setting

∞ dλ = E N(E) dE where E = hc/λ. Normalizing the spectrum so that
F(λ)
0 F(λ) dλ = L h,tree then gives:

L h,tree λγ λ−3/2 (λ λγ )
Fics (λ) = × (8.22)
2 0 (λ < λγ )
where λγ = hc/E max = (0.34 Å)m −2 10 (1 + z) and L h,tree is the halo luminosity
due to tree-level decays.
In the case of more massive neutralinos with m 10 10, the situation is
complicated by the fact that outgoing photons become energetic enough to initiate
pair production via γ + γcmb → e+ + e− . This injects new electrons into the ICS
process, resulting in electromagnetic cascades. For particles which decay at high
redshifts (z 100), other processes such as photon–photon scattering must also
be taken into account [55]. Cascades on non-CMB background photons may also
be important [56]. A full treatment of these effects requires detailed numerical
analysis as carried out for example in [57]. Here we simplify the problem by
assuming that the LSP is stable enough to survive into the late matter-dominated
(or vacuum-dominated) era. The primary effect of cascades is to steepen the decay
spectrum at high energies, so that [54]

 E −3/2 (E E x )
Ncasc (E) ∝ E −2 (E x < E E c ) (8.23)

0 (E > E c )
where
1 E0 2
Ex = E cmb (1 + z)−1 E c = E 0 (1 + z)−1 .
3 m e c2
Here E 0 is a minimum absorption energy. We adopt the numerical expressions
E x = (1.8 × 103 GeV)(1 + z)−1 and E c = (4.5 × 104 GeV)(1 + z)−1
after Protheroe et al [58]. Employing the relation F(λ) dλ = E N(E) dE and
normalizing as before, we find:
√
L h,tree  λx λ−3/2 (λ λx )
Fcasc (λ) = × λ−1 (λx > λ λc ) (8.24)
[2 + ln(λx /λc )] 
0 (λ < λc )
where the new parameters are λx = hc/E x = (7 × 10−9 Å)(1 + z) and

λc = hc/E c = (3 × 10−10 Å)(1 + z).
The luminosity L h,tree is given by
Nχ̃ be E e 2be Mh c2
L h,tree = = . (8.25)
τχ̃ 3τχ̃
Here be is now the branching ratio for all processes of the form χ̃ → e + all, and
E e = 23 m χ̃ c2 is the total energy lost to the electrons. We assume that all of this
Tree-level decays 159
eventually finds its way into the EBL. Berezinsky et al [51] supply the following
branching ratio:
be ≈ 10−6 f χ f R2 m 210 . (8.26)
Here f χ parametrizes the composition of the neutralino, taking the value 0.4
for the pure higgsino case. With the halo mass specified by (6.15) and f τ ≡
τχ̃ /(1 Gyr) as usual, we obtain:
L h,tree = (8 × 1043 erg s−1 )m 210 f χ f R2 fτ−1 . (8.27)
This is approximately four orders of magnitude higher than the halo luminosity
due to one-loop decays, and provides for the first time the possibility of significant
EBL contributions. With all adjustable parameters taking values of order unity, we
find that L h,tree ∼ 2 × 1010 L , which is comparable to the bolometric luminosity
of the Milky Way.
The combined bolometric intensity of all neutralino halos is computed as in
the previous two sections. Replacing L h,loop in (8.10) with L h,tree leads to the
values
 −4 −1 −2 2 2 2 −1 (
 (2 × 10 erg s cm )h 0 m 10 f χ f R fτ m,0 = 0.1h 0 )
Q = (5 × 10−4 erg s−1 cm−2 )h 0 m 210 f χ f R2 fτ−1 (m,0 = 0.3) (8.28)
 −3 −1 −2 −1
(2 × 10 erg s cm )h 0 m 10 f χ f R fτ
2 2 (m,0 = 1).
These are of the same order as or higher than the bolometric intensity of the EBL
from ordinary galaxies, equation (2.25).
To obtain the spectral intensity, we substitute the SEDs Fics (λ) and Fcasc (λ)
into equation (3.6). The results can be written:
zf
(z) dz
Iλ (λ0 ) = Iχ̃,tree (8.29)
0 (1 + z)[ m,0 (1 + z)3 + (1 − m,0 )]1/2
where the quantities Iχ̃ ,tree and (z) are defined as follows. For neutralino rest
energies m 10 º 10 (ICS):
1/2
n 0 L h,tree f c 0.34 Å
Iχ̃ ,tree =
8πh H0m 10 λ0
−1/2
λ0
= (300 CUs)h 20 m 10 fχ fR2 f τ−1 fc (8.30)
Å

1 [λ0 λγ (1 + z)]
(z) =
0 [λ0 < λγ (1 + z)].
Conversely, for m 10 10 (cascades):

1/2
n 0 L h,tree f c 7 × 10−9 Å
Iχ̃ ,tree =
4πh H0[2 + ln(λx /λc )] λ0
SAS−2 (TF82)
G92
COMPTEL (K96)
(upper limits)
EGRET (S98)
1 Fit to all (G99)
2
mχ c > 100 GeV
2
mχ c = 100 GeV
2
mχ c = 30 GeV
2
mχ c = 10 GeV
Iλ ( CUs )
0.1
0.01
1e−08 1e−07 1e−06 1e−05 0.0001 0.001 0.01 0.1 1 10

λ0 ( Å )
Figure 8.6. The spectral intensity of the diffuse γ -ray and x-ray backgrounds due
to neutralino tree-level decays, compared with observational upper limits from SAS-2,
EGRET and COMPTEL in the γ -ray region, and from Gruber’s fits to the experimental
data in the x-ray region (G92,G99). The three plotted curves for each value of m χ̃ c2
correspond to models with m,0 = 0.1h 0 (bold lines), m,0 = 0.3 (medium lines) and
m,0 = 1 (light lines). For clarity we have assumed decay lifetimes in each case such that
the highest theoretical intensities lie just under the observational constraints.
−1/2
λ0
= (0.02 CUs)h 20 m 210 fχ fR2 f τ−1 f c (8.31)
Å


 1 [λ 0 λ x (1 + z)]
λ0
(z) = [λx (1 + z) > λ0 λc (1 + z)]

 λx (1 + z)
0 [λ0 < λc (1 + z)].
Numerical integration of equation (8.29) leads to the plots shown in figure 8.6.
Cascades (like the pair annihilations we have considered already) dominate the
γ -ray part of the spectrum. The ICS process, however, is most important at
lower energies, in the x-ray region. We discuss the observational limits and the
constraints that can be drawn from them in more detail in section 8.7.
Gravitinos 161
8.6 Gravitinos
Gravitinos (g̃) are the SUSY spin- 32 counterparts of gravitons. Although often
discussed along with neutralinos, they are not favoured as dark-matter candidates,
at least in the simplest SUSY theories. The reason for this, often called the
gravitino problem [5], boils down to the fact that they interact too weakly, not
only with other particles but with themselves as well. Hence they annihilate too
slowly and survive long enough to ‘overclose’ the Universe unless some other
way is found to reduce their numbers. Decays are one possibility, but not if the
gravitino is a stable LSP. Gravitino decay products must also not be allowed
to interfere with processes such as primordial nucleosynthesis [4]. Inflation,
followed by a judicious period of reheating, can thin out their numbers to almost
any desired level. But the reheat temperature TR must satisfy kTR º 1012 GeV
or gravitinos will once again become too numerous [9]. Related arguments
based on entropy production, primordial nucleosynthesis and the CMB power
spectrum force this number down to kTR º (109 –1010) GeV [59] or even
kTR º (106–109 ) GeV [60]. These temperatures are incompatible with the
generation of baryon asymmetry in the Universe, a process which is usually
taken to require kTR ∼ 1014 GeV or higher [61].
Recent developments are, however, beginning to loosen the baryogenesis
requirement [62], and there are alternative models in which baryon asymmetry
is generated at energies as low as ∼10 TeV [63] or even 10 MeV–1 GeV [64].
With this in mind we include a brief look at gravitinos here. There are two
possibilities: (1) If the gravitino is not the LSP, then it decays early in the
history of the Universe, well before the onset of the matter-dominated era. In
the model of Dimopoulos et al [65], for example, the gravitino decays both
radiatively and hadronically and is, in fact, ‘long-lived for its mass’ with a lifetime
of τg̃ = (2 − 9) × 105 s. Particles of this kind have important consequences
for nucleosynthesis, and might affect the shape of the CMB, if τg̃ were to exceed
∼107 s. However, they are irrelevant as far as the EBL is concerned. We therefore
restrict our attention to the case (2), in which the gravitino is the LSP. Moreover,
in light of the results we have already obtained for the neutralino, we disregard
annihilations and consider only models in which the LSP can decay.
The decay mode depends on the specific mechanism of R-parity violation.
We follow Berezinsky [66] and concentrate on dominant tree-level processes. In
particular we consider the decay:
g̃ → e+ + all (8.32)
followed by ICS off the CMB, as in the previous section on neutralinos. The
spectrum of photons produced by this process is identical to that in section 8.5,
except that the monoenergetic electrons have energy E e = 12 m g̃ c2 = (5 GeV)m 10
[66], where m g̃ c2 is the rest energy of the gravitino and m 10 ≡ m g̃ c2 /(10 GeV)
as before. This, in turn, implies that E max = (81 keV)m 210 (1 + z)−1 and
λγ = hc/E max = (0.15 Å)m −210 (1 + z). The values of λx and λc are unchanged.
The SED comprises equations (8.22) for ICS and (8.24) for cascades, as
before. Only the halo luminosity needs to be recalculated. This is similar to
equation (8.25) for neutralinos, except that the factor of 23 becomes 12 , and the
branching ratio can be estimated at [66]
α 2
be ∼ = 5 × 10−6 . (8.33)
π
Using our standard value for the halo mass Mh , and parametrizing the gravitino
decay lifetime by f τ ≡ τg̃ /(1 Gyr) as before, we obtain the following halo
luminosity due to gravitino decays:
L h,grav = (3 × 1044 erg s−1 ) f τ−1 . (8.34)
This is higher than the luminosity due to neutralino decays and exceeds the
luminosity of the Milky Way by several times if f τ ∼ 1.
The bolometric intensity of all gravitino halos is computed exactly as before.
Replacing L h,tree in (8.10) with L h,grav , we find:

 (7 × 10−4 erg s−1 cm−2 )h 20 f τ−1 (m,0 = 0.1h 0)
Q = (2 × 10−3 erg s−1 cm−2 )h 0 f τ−1 (m,0 = 0.3) (8.35)

(8 × 10−3 erg s−1 cm−2 )h 0 f τ−1 (m,0 = 1).
It is clear that gravitinos must decay on timescales longer than the lifetime of the
Universe ( f τ 16), or they would produce a background brighter than that of the
galaxies.
The spectral intensity is the same as before, equation (8.29), but with the
new numbers for λγ and L h . This results in
zf
(z) dz
Iλ (λ0 ) = Ig̃ (8.36)
0 (1 + z)[ m,0 (1 + z)3 + (1 − m,0 )]1/2
where the prefactor Ig̃ is defined as follows. For m 10 º 10 (ICS):
1/2
n 0 L h,grav f c 0.15 Å
Ig̃ =
8πh H0m 10 λ0
−1/2
λ0
= (800 CUs)h 20 m −1 −1
10 f τ f c . (8.37)
Å
Conversely, for m 10 ² 10 (cascades):
1/2
n 0 L h,grav f c 7 × 10−9 Å
Ig̃ =
4πh H0[2 + ln(λx /λc )] λ0
−1/2
λ0
= (0.06 CUs)h 20 f τ−1 f c . (8.38)
Å
The x-ray and γ -ray backgrounds 163
SAS−2 (TF82)
G92
COMPTEL (K96)
(upper limits)
EGRET (S98)
1 Fit to all (G99)
2
mg c > 100 GeV
2
mg c = 100 GeV
2
mg c = 30 GeV
2
mg c = 10 GeV
Iλ ( CUs )
0.1
0.01
1e−08 1e−07 1e−06 1e−05 0.0001 0.001 0.01 0.1 1 10

λ0 ( Å )
Figure 8.7. The spectral intensity of the diffuse γ -ray and x-ray backgrounds due
to gravitino tree-level decays, compared with observational upper limits. These come
from SAS-2, COMPTEL and EGRET in the γ -ray region, and from Gruber’s fits to the
experimental data in the x-ray region. The three plotted curves for each value of m g̃ c2
correspond to models with m,0 = 0.1h 0 (bold lines), m,0 = 0.3 (medium lines) and
m,0 = 1 (light lines). For clarity we have assumed decay lifetimes in each case such that
the highest theoretical intensities lie just under the observational constraints.
The function (z) has the same form as in equations (8.30) and (8.31) and does
not need to be redefined (requiring only the new value for the cut-off wavelength
λγ ). Because the branching ratio be in (8.33) is independent of the gravitino
rest mass, m 10 appears in these results only through λγ . Thus the ICS part of
the spectrum goes as m −110 while the cascade part does not depend on m 10 at
all. As with neutralinos, cascades dominate the γ -ray part of the spectrum and
the ICS process is most important in the x-ray region. Numerical integration of
equation (8.36) leads to the results plotted in figure 8.7. We proceed to discuss
these next, beginning with the observational data.
8.7 The x-ray and γ -ray backgrounds

The experimental situation as regards EBL intensity in the x-ray and γ -ray regions
is clearer than that in the optical and ultraviolet. Firm detections (as opposed
to upper limits) have been made in both bands and these are consistent with
expectations based on known astrophysical sources. The constraints that we

derive here are thus conservative ones, in the sense that the EBL flux which could
plausibly be due to decaying WIMPs is almost certainly smaller than the levels
actually measured.
We have not used any data on the lowest-energy, or soft x-ray background,
which lies just beyond the ultraviolet and extends over approximately 0.1–3 keV
or 4–100 Å. The hard x-ray background (3–800 keV or 0.02–4 Å) is, however,
crucial in constraining the decays of low-mass neutralinos and gravitinos via the
ICS process, as can be seen in figures 8.6 and 8.7. (The high-energy spectrum is
conventionally divided into wavebands whose precise definition, however, varies
from author to author. The definitions we use here coincide roughly with energy
ranges in which different detection techniques must be used.) We have included
two compilations of observations in the hard x-ray band, both by Gruber [67, 68].
The first (labelled ‘G92’ in figures 8.6 and 8.7) is an empirical fit to various pre-
1992 measurements, including those from the Kosmos and Apollo spacecraft,
HEAO-1 and balloon experiments. In plotting the uncertainty for this curve, we
have used the fact that the relative scatter of the data increases logarithmically
from 2% at 3 keV to 60% at 3 MeV. The second compilation (‘G99’) is a revision
of this fit in light of new data at higher energies, and has been extended deep
into the γ -ray region. The prominent peak in the range 3–300 keV (0.04–4 Å) is
widely attributed to integrated light from active galactic nuclei (AGN) [69].
In the low-energy γ -ray region (0.8–30 MeV or 0.0004–0.02 Å) we have
used results from the COMPTEL instrument on the Compton Gamma-Ray
Observatory (CGRO), which was operational from 1990–2000 [70]. Four data
points are plotted in figures 8.2, 8.4, 8.6 and 8.7, and two more (upper limits
only) appear at low energies in figures 8.6 and 8.7. These experimental results,
which interpolate smoothly between the backgrounds at both lower and higher
energies, played a key role in the demise of the ‘MeV bump’ (visible in figures 8.6
and 8.7 as a significant upturn in Gruber’s fit to the pre-1992 data from about
0.002–0.02 Å). This apparent feature in the background had previously attracted
a great deal of attention from theoretical cosmologists as a possible signature of
new physics. Figures 8.6 and 8.7 suggest that it could also have been interpreted
as evidence for a long-lived non-minimal SUSY WIMP with a rest energy near
100 GeV. The MeV bump is, however, no longer believed to be real, as the new
fit (‘G99’) makes clear. Most of the background in this region is now suspected
to be due to Type Ia supernovae (SNIa) [71].
There are a number of good measurements in the high-energy γ -ray band
(30 MeV–30 GeV or 4 × 10−7 –4 × 10−4 Å). We have plotted two of these: one
from the SAS-2 satellite which flew in 1972–73 [72], and one from the EGRET
instrument which was part of the CGRO mission along with COMPTEL [73]. As
may be seen in figures 8.2, 8.4, 8.6 and 8.7, the new results essentially confirm
the old ones and extend them to energies as high as 120 GeV (λ0 = 10−7 Å),
with error bars which have been reduced in size by a factor of about ten. Most of
this extragalactic background is thought to arise from unresolved blazars, highly
variable AGN whose relativistic jets are pointed in nearly our direction [74].
Finally, we have made use of some observations in the very high-energy
(VHE) γ -ray region (30 GeV–30 TeV or 4 × 10−10 –4 × 10−7 Å). Because the
extragalactic component of the background has not yet been measured beyond
120 GeV, we have fallen back on measurements of total γ -ray flux, obtained by
Nishimura et al in 1980 [75] using a series of high-altitude balloon experiments.
These are shown as filled dots (‘N80’) in figures 8.2 and 8.4. They constitute a
robust upper limit on EBL flux, since the majority of the observed signals are due
to cosmic-ray interactions in the atmosphere of the Earth.
Some comments are in order here about notation and units in this part of
the spectrum. For experimental reasons, measurements are often expressed in
terms of integral flux E IE (>E 0 ) or number of particles with energies above
E 0 . This presents no difficulties in the high-energy γ -ray region where the
differential spectrum is well approximated with a single power-law component,
IE (E 0 ) = I∗ (E 0 /E ∗ )−α . The conversion to integral form is then given by
∞ 1−α
E ∗ I∗ E0
E IE (>E 0 ) = IE (E) dE = . (8.39)
E0 α−1 E∗
The spectrum in either case can be specified by its index α, together with the
values of E ∗ and I∗ (or E 0 and E IE the integral case). Thus the final SAS-2 results
were reported as α = 2.35+0.4−0.3 with E IE = (5.5 ± 1.3) × 10
−5 s−1 cm−2 ster−1
for E 0 = 100 MeV [72]. The EGRET spectrum is instead specified by

α = 2.10 ± 0.03 with I∗ = (7.32 ± 0.34) × 10−9 s−1 cm−2 ster−1 MeV−1 for
E ∗ = 451 MeV [73]. To convert a differential flux in these units to Iλ in CUs, we
multiply by E 0 /λ0 = E 02 / hc = 80.66E 02 where E 0 is photon energy in MeV.
We now discuss our results, beginning with the neutralino annihilation fluxes
plotted in figure 8.2. These are at least three orders of magnitude fainter than
the background detected by EGRET [73] (and four orders of magnitude below
the upper limit set by the data of Nishimura et al [75] at shorter wavelengths).
This agrees with previous studies assuming a critical density of neutralinos
[20, 40]. Figure 8.2 shows that EBL contributions would drop by another order of
magnitude in the favoured scenario with m,0 ≈ 0.3, and by another if neutralinos
are confined to galaxy halos (m,0 ≈ 0.1h 0 ). Unfortunately, the same stability
that makes minimal SUSY WIMPs so compelling as dark-matter candidates also
makes them hard to detect. As discussed in section 8.3, prospects for observing
these particles can improve substantially if one looks for the enhanced flux from
high-density regions like the Galactic centre.
Figure 8.4 shows the EBL contributions from one-loop neutralino decays in
non-minimal SUSY. We have put h 0 = 0.75, z f = 30 and f R = 1. Depending
on their decay lifetime (here parametrized by f τ ), these particles are capable,
in principle, of producing a background comparable to (or even in excess of)
the EGRET limits. The plots in figure 8.4 correspond to the smallest values
of f τ which are consistent with the data for m 10 = 1, 3, 10, 30 and 100.
1e+08
1e+07
1e+06
100000
τχ or τg ( Gyr )
10000
1000
χ̃ (loop): Ωm,0 = 1
100 Ωm,0 = 0.3
Ωm,0 = 0.1 h0
10 χ
˜ (tree): Ωm,0 = 1
Ωm,0 = 0.3
1 Ωm,0 = 0.1 h0
g˜ (tree): Ωm,0 = 1
Ωm,0 = 0.3
Ωm,0 = 0.1 h0
10 100 1000
2 2
mχ c or mg c ( GeV )
Figure 8.8. The lower limits on WIMP decay lifetime derived from observations of
the x-ray and γ -ray backgrounds. Neutralino bounds are shown for both one-loop
decays (dotted lines) and tree-level decays (dashed lines). For gravitinos we show only
the tree-level constraints (unbroken lines). For each process there are three curves
corresponding to models with m,0 = 1 (light lines), m,0 = 0.3 (medium lines) and
m,0 = 0.1h 0 (bold lines).
Following the same procedure here as we did for axions in chapter 6, we can
repeat this calculation over more finely-spaced intervals in neutralino rest mass,
obtaining a lower limit on decay lifetime τχ̃ as a function of m χ̃ . Results are
shown in figure 8.8 (dotted lines). The lower limits obtained in this way range
from 4 Gyr for the lightest neutralinos (assumed to be confined to galaxy halos
with a total matter density of m,0 = 0.1h 0 ) to 70 000 Gyr for the heaviest (if
these provide enough CDM to put m,0 = 1). A typical intermediate limit (for
m χ̃ c2 = 100 GeV and m,0 = 0.3) is τχ̃ > 2000 Gyr.
Figure 8.6 is a plot of EBL flux from indirect neutralino decays via the tree-
level, ICS and cascade processes described in section 8.5. These provide us with
our strongest constraints on non-minimal SUSY WIMPs. We have set h 0 = 0.75,
z f = 30 and f χ = f R = 1, and assumed values of f τ such that the highest
predicted intensities lie just under observational limits, as before. Contributions
from neutralinos at the light end of the mass range are constrained by x-ray data,
while those at the heavy end are checked by the EGRET measurements. Both
the shape and absolute intensity of the ICS spectra depend on the neutralino rest
mass. The cascade spectra, however, depend on m 10 through intensity alone

(via the prefactor Iχ̃ ,tree ). When normalized to the observational upper limit,
all curves for m 10 > 10 therefore overlap. Normalizing across the full range
of neutralino rest masses (as for one-loop decays), we obtain the lower bounds
on lifetime τχ̃ plotted in figure 8.8 (dashed lines). These range from 60 Gyr to
6 × 107 Gyr, depending on the values of m χ̃ c2 and m,0 . A typical result (for
m χ̃ c2 = 100 GeV and m,0 = 0.3) is τχ̃ > 70 000 Gyr.
Figure 8.7, finally, shows the EBL contributions from tree-level gravitino
decays. These follow the same pattern as the neutralino decays, and need little
additional comment. Requiring that the predicted signal not exceed the x-ray
and γ -ray observations, we obtain the lower limits on decay lifetime plotted in
figure 8.8 (unbroken lines). These curves are seen to be flatter for gravitinos than
they are for neutralinos. This is a consequence of the fact that the branching
ratio (8.33) is independent of m 10 . Our lower limits on τg̃ range from 200 to
20 000 Gyr, with a typical value (for m g̃ c2 = 100 GeV and m,0 = 0.3) of
τχ̃ > 4000 Gyr.
These results agree with estimates in the literature [51, 66] and confirm that,
whether it is a neutralino or gravitino, the LSP in non-minimal SUSY theories must
be very nearly stable . Neither particle can be excluded as a dark-matter candidate
on this basis; the LSP may after all be perfectly stable, which is the simplest
(minimal) case. However, an ‘almost-stable’ LSP is difficult to understand
in the context of non-minimal SUSY theory, because it requires that R-parity
conservation be violated, but violated at improbably low levels. Equivalently, it
introduces a very small dimensionless quantity into the theory, analogous to the
parameter associated with the strong-CP problem (chapter 6). To this extent,
then, our constraints on decay lifetime suggest that the SUSY WIMP either exists
in the context of minimal SUSY theory, or not at all.
Let us sum up our findings in this chapter. We have considered neutralinos
and gravitinos, either of which could be the LSP and hence make up the ‘missing
mass’ in SUSY theories. Both particles remain viable as dark-matter candidates
but their properties are constrained by ever-improving data on the intensity of the
extragalactic x-ray and γ -ray backgrounds. In the minimal SUSY model, where
R-parity is strictly conserved, neutralinos annihilate so slowly that no useful limits
can be set at this time. (This would be even more true of gravitinos.) In the
broader class of non-minimal SUSY theories, however, R-parity conservation is
no longer guaranteed and the LSP (whether neutralino or gravitino) can decay.
We have shown that any such decay must occur on timescales longer than 102–
108 Gyr (for neutralinos) or 102 –104 Gyr (for gravitinos), depending on their rest
masses and various theoretical input parameters. For either case it is clear that
these particles must be almost perfectly stable. This, in turn, implies that R-parity
must be broken very gently, if it is broken at all.
References
[1] Jungman G, Kamionkowski M and Griest K 1996 Phys. Rep. 267 195
[2] Cabibbo N, Farrar G R and Maiani L 1981 Phys. Lett. B 105 155
[3] Pagels H and Primack J R 1982 Phys. Rev. Lett. 48 223
[4] Weinberg S 1982 Phys. Rev. Lett. 48 1303
[5] Ellis J, Linde A D and Nanopoulos D V 1982 Phys. Lett. B 118 59
[6] Goldberg H 1983 Phys. Rev. Lett. 50 1419
[7] Ibáñez L E 1984 Phys. Lett. B 137 160
[8] Sciama D W 1984 Phys. Lett. B 137 169
[9] Ellis J et al 1984 Nucl. Phys. B 238 453
[10] Steigman G and Turner M S 1985 Nucl. Phys. B 253 375
[11] Ellis J and Olive K A 2001 Phys. Lett. B 514 114
[12] Ellis J, Falk T, Ganis G, Olive K A 2000 Phys. Rev. D 62 075010
[13] Griest K, Kamionkowski M and Turner M S 1990 Phys. Rev. D 41 3565
[14] Ellis J, Falk T and Olive K A 1998 Phys. Lett. B 444 367
[15] Roszkowski L, de Austri R R and Nihei T 2001 J. High Energy Phys. 08 024
[16] Baudis L and Klapdor-Kleingrothaus H V 2000 Beyond the Desert 1999 ed
H V Klapdor-Kleingrothaus and I V Krivosheina (Oxford: Institute of Physics
Press) p 881
[17] Belli P et al 2000 Beyond the Desert 1999 ed H V Klapdor-Kleingrothaus and
I V Krivosheina (Oxford: Institute of Physics Press) p 869
[18] Abusaidi R et al 2000 Phys. Rev. Lett. 84 5699
[19] Benoit A et al 2001 Phys. Lett. B 513 15
[20] Silk J and Srednicki M 1984 Phys. Rev. Lett. 53 624
[21] Silk J, Olive K and Srednicki M 1985 Phys. Rev. Lett. 55 257
[22] Freese K 1986 Phys. Lett. B 167 295
[23] Krauss L M, Srednicki M and Wilczek F 1986 Phys. Rev. D 33 2079
[24] Bai X et al 2001 Dark Matter in Astro- and Particle Physics ed H V Klapdor-
Kleingrothaus (Heidelberg: Springer) p 699
[25] Silk J and Bloemen H 1987 Astrophys. J. 313 L47
[26] Stecker F W 1988 Phys. Lett. B 201 529
[27] Rudaz S 1989 Phys. Rev. D 39 3549
[28] Stecker F W and Tylka A J 1989 Astrophys. J. 343
[29] Bouquet A, Salati P and Silk J 1989 Phys. Rev. D 40 3168
[30] Bergström L 1989 Nucl. Phys. B 325 647
[31] Freese K and Silk J 1989 Phys. Rev. D 40 3828
[32] Berezinsky V, Bottino A and Mignola G 1994 Phys. Lett. B 325 136
[33] Gondolo P and Silk J 1999 Phys. Rev. Lett. 83 1719
[34] Urban M et al 1992 Phys. Lett. B 293 149
[35] Chardonnet P et al 1995 Astrophys. J. 454 774
[36] Lake G 1990 Nature 346 39
[37] Gondolo P 1994 Nucl. Phys. Proc. Suppl. B B35 148
[38] Baltz E A et al 1999 Phys. Rev. D 61 023514
[39] Cline D B and Gao Y-T 1990 Astron. Astrophys. 231 L23
[40] Gao Y-T, Stecker F W and Cline D B 1991 Astron. Astrophys. 249 1
[42] Berezinsky V S, Bottino A and de Alfaro V 1992 Phys. Lett. B 274 122
[43] Giudice G F and Griest K 1989 Phys. Rev. D 40 2549

[44] Gates E I, Gyuk G and Turner M S 1995 Astrophys. J. 449 L123
[45] Metzger M R, Caldwell J A R and Schechter P L 1998 Astron. J. 115 635
[48] Berezinsky V S, Gurevich A V and Zybin K P 1992 Phys. Lett. B 294 221
[49] Bouquet A and Salati P 1987 Nucl. Phys. B 284 557
[50] Masiero A and Valle J W F 1990 Phys. Lett. B 251 273
[51] Berezinsky V, Masiero A and Valle J W F 1991 Phys. Lett. B 266 382
[52] Barbieri R and Berezinsky V 1988 Phys. Lett. B 205 559
[53] Blumenthal G R and Gould R J 1970 Rev. Mod. Phys. 42 237
[54] Berezinsky V S 1992 Nucl. Phys. B 380 478
[55] Svensson R and Zdziarski A A 1990 Astrophys. J. 349 415
[56] Coppi P S and Aharonian F A 1997 Astrophys. J. 487 L9
[57] Kribs G D and Rothstein I Z 1997 Phys. Rev. D 55 4435
[58] Protheroe R J, Stanev T and Berezinsky V S 1995 Phys. Rev. D 51 4134
[59] Ellis J, Kim E and Nanopoulos D V 1984 Phys. Lett. B 145 181
[60] Kawasaki M and Moroi T 1995 Prog. Theor. Phys. 93 879
[61] Kolb E W and Turner M S 1990 The Early Universe (Reading: Addison-Wesley)
p 157
[62] Giudice G F, Kolb E W and Riotto A 2001 Phys. Rev. D 64 023508
[63] Kuzmin V A, Rubakov V A and Shaposhnikov M E 1985 Phys. Lett. B 155 36
[64] Dimopoulos S and Hall L J 1987 Phys. Lett. B 196 135
[65] Dimopoulos S et al 1988 Astrophys. J. 330 545
[66] Berezinsky V S 1991 Phys. Lett. B 261 71
[67] Gruber D E 1992 The X-Ray Background ed X Barcon and A C Fabian (Cambridge:
Cambridge University Press) p 44
[68] Gruber D E 1999 Astrophys. J. 520 124
[69] Zdziarski A A 1996 Mon. Not. R. Astron. Soc. 281 L9
[70] Kappadath S C et al 1996 Astron. Astrophys. Suppl. Ser. 120 619
[71] The L-S, Leising M D and Clayton D D 1993 Astrophys. J. 403 32
[72] Thompson D J and Fichtel C E 1982 Astron. Astrophys. 109 352
[73] Sreekumar P et al 1998 Astrophys. J. 494 523
[74] McNaron-Brown K et al 1995 Astrophys. J. 451 575
[75] Nishimura J et al 1980 Astrophys. J. 238 394
Chapter 9
Black holes
9.1 Primordial black holes

Black holes are regions of space from which light cannot escape. It
might therefore appear that little could be learned about these objects from
measurements of the extragalactic background light (EBL). In fact, experimental
data on EBL intensity constrain black holes more strongly than any of the other
dark-matter candidates we have discussed so far. Before explaining how this
comes about, we distinguish between ‘ordinary’ black holes (which form via the
gravitational collapse of massive stars at the end of their lives) and primordial
black holes (PBHs) which could have arisen from the collapse of overdense
regions in the early Universe. The existence of the former is very nearly an
established astronomical fact, while the latter remain hypothetical. However,
it is the latter (PBHs) which are of most interest to us as potential dark-matter
candidates.
The reason for this is as follows. Ordinary black holes come from baryonic
progenitors (i.e. stars) and can hence be classified with the baryonic dark matter
of the Universe. (They are, of course, not ‘baryonic’ in all respects, since among
other things their baryon number is not defined.) Ordinary black holes are
therefore subject to the nucleosynthesis bound (4.4) on the density of baryonic
matter, which limits them to less than 5% of the critical density. PBHs are
not subject to this bound because they form during the radiation-dominated era,
before nucleosynthesis begins. A priori, nothing prevents them from making
up most of the density in the Universe. Moreover, they constitute cold dark
matter because their velocities are low. (That is, they collectively obey a
dustlike equation of state, even though they might individually be better described
as ‘radiation-like’ than baryonic.) PBHs were first proposed as dark-matter
candidates by Zeldovich and Novikov in 1966 [1] and Hawking in 1971 [2].
Black holes contribute to the EBL via a process discovered by Hawking
in 1974 and often called Hawking evaporation [3]. Photons cannot escape
from inside the black hole, but they are produced at or near the horizon by
170
Initial mass distribution 171
quantum fluctuations in the surrounding curved spacetime. These give rise to

a net flux of particles which propagates outward from a black hole (of mass M) at
a rate proportional to M −2 (with the black-hole mass itself dropping at the same
rate). For ordinary, stellar-mass black holes, this process occurs so slowly that
contributions to the EBL are insignificant and the designation ‘black’ remains
perfectly appropriate over the lifetime of the Universe. PBHs, however, can
have masses far smaller than those of a star, leading to correspondingly higher
luminosities. Those with M º 1015 g (equivalent to the mass of a small asteroid)
would, in fact, evaporate quickly enough to shed all their mass over less than
∼10 Gyr. They would already have expired in a blaze of high-energy photons and
other elementary particles as M → 0.
Our goal in this chapter will be to estimate the impact of this process
on the intensity of the EBL, and also to consider the corresponding problem
for the higher-dimensional analogues of black holes known as solitons. In
four dimensions this is a calculation that has been refined by many people and
has entered into the ‘standard lore’ of dark-matter astrophysics. For the most
part we will be content to review these well-established results, commenting
as appropriate on the ways in which they have been extended and modified
by later research. We will confirm that observations of the γ -ray background
exclude a significant role for four-dimensional black holes under the most natural
assumptions. Solitons are not as easy to constrain but we will find that the
bolometric intensity of the background allows us to put useful limits on these
objects as well.
9.2 Initial mass distribution
Our sources of background radiation in this chapter are the PBHs themselves.
We will take these to be distributed homogeneously throughout space. This is
not necessarily realistic since their low velocities mean that PBHs will tend to
collect inside the potential wells of galaxies and galaxy clusters. Clustering is
not of great concern to us here, however, for the same reason already discussed
in connection with axions and WIMPs (i.e. we are interested primarily in the
combined contributions of all PBHs to the diffuse background). A complication
is introduced, however, by the fact that PBHs cover such a wide range of masses
and luminosities that we can no longer treat all sources in the same way (as we
did for the galaxy halos in previous chapters). Instead we must define quantities
like number density and energy spectrum as functions of PBH mass as well as
time, and integrate our final results over both parameters.
The first step is to identify the distribution of PBH masses at the time when
they formed. There is little prospect of probing the time before nucleosynthesis
experimentally, so any theory of PBH formation is necessarily speculative to
some degree. However, the scenario requiring the least extrapolation from known
physics would be one in which PBHs arose via the gravitational collapse of
172 Black holes
small initial density fluctuations on a standard Robertson–Walker background.

It is furthermore reasonable to assume that the equation of state had the usual
form (2.27) and that the initial density fluctuations were distributed as
δ = (Mi /Mf )−n . (9.1)
Here Mi is the initial mass of the PBH, Mf is the mass lying inside the particle
horizon (or causally connected Universe) at PBH formation time, and is a
proportionality constant.
An investigation of PBH formation under these conditions was carried out in
1975 by Carr [4], who showed that the process is favoured over an extended range
of masses only if n = 23 . Proceeding on this assumption, he found that the initial
mass distribution of PBHs formed with masses between Mi and Mi + dMi per
unit comoving volume is
−β
Mi
n(Mi ) dMi = ρf Mf−2 ζ dMi (9.2)
Mf
where ρf is the mean density at formation time. The parameters β and ζ are
formally given by 2(2γ − 1)/γ and exp[−(γ − 1)2 /2 2 ] respectively, where
γ is the equation-of-state parameter in (2.27). However, in the interests of lifting
every possible restriction on conditions prevailing in the early Universe, we follow
Carr [5] in treating β and ζ as free parameters, not necessarily related to γ and
. Insofar as the early Universe was governed by the equation of state (2.27), β
will take values between 2 (dustlike or ‘soft’) and 3 (stiff or ‘hard’), with β = 52
corresponding to the most natural situation (i.e. γ = 43 as for a non-interacting
relativistic gas). But we will allow β to take values up to 4, corresponding to
‘superhard’ early conditions. The parameter ζ can be understood physically as
the fraction of the Universe which goes into PBHs of mass Mf at time tf . It is a
measure of the initial inhomogeneity of the Universe.
The fact that equation (9.2) has no exponential cut-off at high mass is
important because it allows us (at least in principle) to obtain a substantial
cosmological density of PBHs. Since 2 β 4, however, the power-law
distribution is dominated by PBHs of low mass. This is the primary reason why
PBHs turn out to be so tightly constrained by data on background radiation. It is
the low-mass PBHs whose contributions to the EBL via Hawking evaporation are
the strongest.
Much subsequent effort has gone into the identification of alternative
formation mechanisms which could give rise to a more favourable distribution of
PBH masses (i.e. one peaked at sufficiently high mass to provide the requisite
CDM density without the unwanted background radiation from the low-mass
tail). We pause here to survey some of these developments. Perhaps the least
speculative possibility is that PBHs arise from a post-inflationary spectrum of
density fluctuations which is not perfectly scale-invariant but has a characteristic
length of some kind [6]. The parameter ζ in (9.2) would then depend explicitly on
Evolution and number density 173
the inflationary potential (or analogous quantities). This kind of dependence has
been discussed, for example, in the context of two-stage inflation [7], extended
inflation [8], chaotic inflation [9], ‘plateau’ inflation [10], hybrid inflation [11]
and inflation via isocurvature fluctuations [12].
A narrow spectrum of masses might also be expected if PBHs formed during
a spontaneous phase transition rather than arising from primordial fluctuations.
The quark–hadron transition [13], grand unified symmetry-breaking transition
[14] and Weinberg–Salam phase transition [15] have all been considered in this
regard. The initial mass distribution in each case would be peaked near the
horizon mass Mf at transition time. The quark–hadron transition has attracted
particular attention because PBH formation at this time would be enhanced by
a temporary softening of the equation of state; and because Mf for this case
is coincidentally close to M , so that PBHs might be responsible for MACHO
observations of microlensing in the halo [16]. Cosmic string loops have also
been explored as possible seeds for PBHs with a peaked mass spectrum [17, 18].
And considerable interest has recently been generated by the discovery that PBHs
could provide a physical realization of the theoretical phenomenon known as
critical collapse [19]. If this is so, then initial PBH masses would no longer
necessarily be clustered near Mf .
While any of these proposals can, in principle, concentrate the PBH
population within a narrow mass range, all of them face the same problem of fine-
tuning if they are to produce the desired present-day density of PBHs. In the case
of inflationary mechanisms it is the form of the potential which must be adjusted.
In others it is the bubble nucleation rate, the string mass per unit length or the
fraction of the Universe going into PBHs at formation time. The degree of fine-
tuning required is typically of order one part in 109 . Thus, while modifications
of the initial mass distribution can weaken the ‘standard’ constraints on PBH
properties (which we derive later), they do not as yet have a compelling physical
basis. Similar comments apply to PBH-based explanations for specific classes of
observational phenomena. It has been suggested, for instance, that PBHs with the
right mass could be responsible for certain kinds of γ -ray bursts [20–22] or for
long-term quasar variability via microlensing [23, 24]. Other connections have
been drawn to diffuse γ -ray emission from the galactic halo [25, 26] as well as
the MACHO events mentioned earlier [27, 28]. These suggestions are intriguing,
but experimental confirmation of any one (or more) of them would raise almost
as many questions as it answers.
9.3 Evolution and number density
In order to obtain the comoving number density of PBHs from their initial mass
distribution, we use the fact that PBHs evaporate at a rate which is inversely
174 Black holes
proportional to the square of their masses:
dM α
= − 2. (9.3)
dt M
This applies to uncharged, non-rotating black holes, which is a reasonable
approximation in the case of PBHs since these objects discharge quickly relative
to their lifetimes [29] and also give up angular momentum by preferentially
emitting particles with spin [30]. The parameter α depends in general on the
PBH mass M and its behaviour was worked out in detail by Page in 1976
[31]. The PBHs which are of most importance for our purposes are those with
4.5 × 1014 g M 9.4 × 1016 g. Black holes in this range are light enough (and
therefore ‘hot’ enough) to emit massless particles (including photons) as well as
ultrarelativistic electrons and positrons. The corresponding value of α is
α = 6.9 × 1025 g3 s−1 . (9.4)
For M > 9.4 × 1016 g, the value of α drops to 3.8 × 1025 g3 s−1 because the
larger black hole is ‘cooler’ and no longer able to emit electrons and positrons.
EBL contributions from PBHs of this mass are of lesser importance because of
the shape of the mass distribution.
As the PBH mass drops below 4.5 × 1014 g, its energy kT climbs above the
rest energies of successively heavier particles, beginning with muons and pions.
As each mass threshold is passed, the PBH is able to emit more particles and the
value of α increases further. At temperatures above the quark–hadron transition
(kT ≈ 200 MeV), MacGibbon and Webber have shown that relativistic quark and
gluon jets are likely to be emitted rather than massive elementary particles [32].
These jets subsequently fragment into stable particles, and the photons produced
in this way are actually more important (at these energies) than the primary photon
flux. The precise behaviour of α in this regime depends on one’s choice of particle
physics. A plot of α(M) for the standard model is found in the review by Halzen
et al [33], who also note that α climbs to 7.8 × 1026 g3 s−1 at kT = 100 GeV, and
that its value would be at least three times higher in supersymmetric extensions
of the standard model where there are many more particle states to be emitted.
As we will shortly see, however, EBL contributions from PBHs at these
temperatures are suppressed by the fact that the latter have already evaporated
away all their mass and vanished. If we assume for the moment that PBH
evolution is adequately described by (9.3) with α = constant as given by (9.4),
then integration gives
M(t) = (Mi3 − 3αt)1/3 . (9.5)
The lifetime tpbh of a PBH is found by setting M(tpbh ) = 0, giving tpbh = Mi3 /3α.
Therefore the initial mass of a PBH which is just disappearing today (tpbh = t0 ) is
given by
M∗ = (3αt0 )1/3 . (9.6)
Cosmological density 175
Taking t0 = 16 Gyr and using (9.4) for α, we find that M∗ = 4.7 × 1014 g. A
numerical analysis allowing for changes in the value of α over the full range of
PBH masses with 0.06 m,0 1 and 0.4 h 0 1 leads to a somewhat larger
result [33]:
M∗ = (5.7 ± 1.4) × 1014 g. (9.7)
PBHs with M ≈ M∗ are exploding at redshift z ≈ 0 and consequently dominate

the spectrum of EBL contributions. The parameter M∗ is therefore of central
importance in what follows.
We now obtain the comoving number density of PBHs with masses between
M and M + dM at any time t. Since this is the same as the comoving number
density of PBHs with initial masses between Mi and Mi + dMi at formation time,
we can write
n(M, t) dM = n(Mi ) dMi . (9.8)
Inverting equation (9.5) to get Mi = (M 3 + 3αt)1/3 and differentiating, we find

from equations (9.2) and (9.8) that
2 3 −(β+2)/3
n(M, t) dM = M
M∗
M
M∗
+
t
t0
d
M
M∗
(9.9)
where we have used (9.6) to replace M∗3 with 3αt0 . Here the parameter is
formally given in terms of the various parameters at PBH formation time by
= (ζρf /Mf )(Mf /M∗ )β−1 and has the dimensions of a number density. As
we will see, it corresponds roughly to the comoving number density of PBHs of

mass M∗ . Following Page and Hawking [34], we allow to move up or down as
required by observational constraints. The theory to this point is thus specified by

two adjustable input parameters: the PBH normalization
of-state parameter β. In terms of the dimensionless variables
τ ≡ t/t0 (9.9) reads

and the equation-
≡ M/M∗ and
n( , τ ) d = ( + τ ) 2 3 −(β+2)/3
d. (9.10)
number density of PBHs with mass ratios between and

This+isdthe atcomoving
time t = t τ .
0
9.4 Cosmological density

To obtain the present mass density of PBHs with mass ratios between
+ d , we multiply equation (9.10) by the PBH mass M = M∗
τ = 1 so that
and
and put
ρpbh ( , 1) d = M ∗
1−β
(1 + −3 −(β+2)/3
) d . (9.11)
176 Black holes
The total PBH density is then found by integrating over

from zero to infinity.
The integral can be solved by changing variables to x ≡ −3 , whereupon
∞
ρpbh = 13 M∗ x a−1 (1 − x)−(a+b) dx.
0
(9.12)
Here a ≡ 13 (β − 2) and b ≡ 43 . The solution is
ρpbh = kβ M∗ kβ ≡
(a)(b)
3(a + b)
(9.13)
where (x) is the gamma function. Allowing β to take values from 2 through 5
2
(the most natural situation) and up to 4, we obtain:

 ∞ (β = 2)


 1.87 (β = 2.5)
kβ = 0.88 (β = 3) (9.14)



 0.56 (β = 3.5)
0.40 (β = 4).
latter
M . That is, can be thought of as the present number density of PBHs, if the
The combined mass density of all PBHs in the Universe is thus of order ρ ≈
∗
pbh
(9.13) can be recast as a relation between and the PBH density parameter
are imagined to consist predominantly of objects with M ≈ M . Equation
∗
k M
= ρ /ρ
pbh pbh :
crit,0
β ∗
= pbh . (9.15)
ρ crit,0
PBH number density and density are thus interchangeable as free

pbh
limit due to Page and Hawking of º 10 pc [34], then equations (2.36),

parameters. If we adopt the most natural value for β (=2.5) together with an upper
4 −3
−8 −2
this upper limit on holds (as we will confirm under the simplest assumptions),
(9.7), (9.14) and (9.15) together imply that is, at most, of order ∼10 h . If
pbh 0
then there is little hope for PBHs to make up the dark matter.
Equation (9.14) shows that one way to boost their importance would be to
assume a soft equation of state at formation time (i.e. a value of β close to 2 as
for dustlike matter, rather than 2.5 as for radiation). Physically this is related to
the fact that low-pressure matter offers little resistance to gravitational collapse.
Such a softening has, in fact, been shown to occur during the quark–hadron
transition [16], leading to increases in pbh for PBHs which form at that time
(subject to the fine-tuning problem noted in section 9.2). For PBHs which arise
from primordial density fluctuations, however, conditions of this kind are unlikely
to hold throughout the formation epoch. In the limit β → 2 equation (9.2) breaks
down in any case because it becomes possible for PBHs to form on scales smaller
than the horizon [4].
Spectral energy distribution 177
9.5 Spectral energy distribution

Hawking [35] proved that an uncharged, non-rotating black hole emits bosons
(such as photons) in any given quantum state with energies between E and E +dE
at the rate
s dE
d Ṅ = . (9.16)
2π [exp(E/kT ) − 1]
Here T is the effective black-hole temperature, and s is the absorption
coefficient or probability that the same particle would be absorbed by the black

hole if incident upon it in this state. The function d Ṅ is related to the spectral
energy distribution (SED) of the black hole by d Ṅ = F(λ, ) dλ/E, since
we have defined F(λ, ) dλ as the energy emitted between wavelengths λ and

λ + dλ. Here we anticipate the fact that F will depend explicitly on the PBH mass
as well as wavelength. The PBH SED thus satisfies
F(λ, ) dλ = 2π [exp(E/kT
E dE
s
) − 1]
. (9.17)
The absorption coefficient s is, in general, a complicated function of

E as well as the quantum numbers s (spin), (total angular momentum) and m

and
(axial angular momentum) of the emitted particles. Its form was first calculated
by Page [31]. At high energies, and in the vicinity of the peak of the emitted
spectrum, a good approximation is given by [36]
s ∝ M 2 E 2 . (9.18)
This approximation breaks down at low energies, where it gives rise to errors of
order 50% for (GME/ c3 ) ∼ 0.05 [37] or (with E = 2π c/λ and M ∼ M∗ ) for
λ ∼ 10−3 Å. This is adequate for our purposes, as we will find that the strongest
constraints on PBHs come from those with masses M ∼ M∗ at wavelengths
λ ∼ 10−4 Å.
Putting (9.18) into (9.17) and making the change of variable to wavelength
λ = hc/E, we obtain the SED
C λ dλ
F(λ, ) dλ = exp(hc/kT 2 −5
λ) − 1
(9.19)
where C is a proportionality constant. This has the same form as the blackbody
spectrum, equation (3.22). One should, however, keep in mind that we have
made three simplifying assumptions in arriving at this equation. First, we have
neglected the black-hole charge and spin (as justified in section 9.3). Second, we
have used an approximation for the absorption coefficient s . And third, we have
treated all the emitted photons as if they are in the same quantum state whereas, in
fact, the emission rate (9.16) applies separately to the = s (= 1), = s + 1 and
= s + 2 modes. There are thus actually three distinct quasi-blackbody photon
178 Black holes
spectra with different characteristic temperatures for any single PBH. However,
Page [31] has demonstrated that the = s mode is overwhelmingly dominant,
with the = s+1 and = s+2 modes contributing less than 1% and 0.01% of the
total photon flux respectively. Equation (9.19) is thus a reasonable approximation
to the SED of the PBH as a whole.
To fix the value of C we use the fact that the total flux of photons (in all
modes) radiated by a black hole of mass M is given [31]

Ṅ = d Ṅ =
∞

F(λ, ) dλ
= 5.97 × 1034 s−1
−1
M
. (9.20)

λ=0 hc/λ 1 g
Inserting (9.19) and recalling that M = M∗ , we find that
∞
λ−4 dλ
.
hc
C = (5.97 × 1034 g s−1 ) (9.21)
0 exp(hc/kT λ) − 1 M∗ 3
The definite integral on the left-hand side of this equation can be solved by
switching variables to ν = c/λ:
∞ −3
ν 2 dν/c3 hc
= (3)ζ(3) (9.22)
0 exp(hν/kT ) − 1 kT
where (n) and ζ (n) are the gamma function and Riemann zeta function
respectively. We then apply the fact that the temperature T of an uncharged,
non-rotating black hole is given by
c3
T = . (9.23)
8πκ G M
Putting (9.22) and (9.23) into (9.21) and rearranging terms leads to
(4π)6 hG 3 M∗2
C = (5.97 × 1034 g s−1 ) . (9.24)
c5 (3)ζ(3)

Using (3) = 2! = 2 and ζ (3) = ∞k=1 k
−3 = 1.201 along with (9.7) for M ,
∗
we find
C = (270 ± 120) erg Å s−1 .
4
(9.25)
We can also use the definitions (9.23) to define a useful new quantity:
2

hc 4π
λpbh ≡ = G M∗ = (6.6 ± 1.6) × 10−4 Å. (9.26)
kT c
The size of this characteristic wavelength tells us that we will be concerned
primarily with the high-energy γ -ray portion of the spectrum. In terms of C
and λpbh the SED (9.19) now reads
) = exp(C λ /λ/λ) − 1 .
F(λ,
2
pbh
5
(9.27)
depend on time through the PBH mass ratio .

While this contains no explicit time-dependence, the spectrum does of course
Luminosity 179
9.6 Luminosity

To compute the PBH luminosity we employ equation (3.1), integrating the
SED F(λ, ) over all wavelengths to obtain:
∞
L( ) = C 2
0
λ−5 dλ

exp( λpbh /λ) − 1
. (9.28)
This definite integral is also solved by means of a change of variable to frequency

ν, with the result that
L( ) = C (λ
2
pbh )
−4
(4)ζ(4). (9.29)
Using equations (9.7), (9.24) and (9.26) along with the values (4) = 3! = 6 and
ζ (4) = π 4 /96, we can put this into the form
L( ) = L
pbh
−2
(9.30)
where
(5.97 × 1034 g s−1 )π 2 hc3
L pbh = = (1.0 ± 0.4) × 1016 erg s−1 .
512ζ (3)G M∗2
≈ 1) is not very luminous. A PBH of 900 kg or so might theoretically be

Compared to the luminosity of an ordinary star, the typical PBH (of mass ratio
expected to reach the Sun’s luminosity; however, in practice it would already

have exploded, having long since reached an effective temperature high enough
of black holes in general can be emphasized by using the relation

to recast equation (9.30) in the form
≡ M/M∗
to emit a wide range of massive particles as well as photons. The low luminosity
−2
L M
= 1.7 × 10−55 . (9.31)
L M
This expression is not strictly valid for PBHs of masses near M , having been
derived for those with M ∼ M∗ ∼ 1015 g. For more massive PBHs, the
luminosity is if anything lower. (In fact, a black hole of Solar mass would be
colder than the CMB and would absorb radiation faster than it could emit it.) So,
Hawking evaporation or not, most black holes are indeed very black.
9.7 Bolometric intensity

To obtain the total bolometric intensity of PBHs out to a look-back time t0 − tf ,
we substitute the PBH number density (9.10) and luminosity (9.28) into the

integral (2.14) as usual. This time, however, number density n(t) is to be replaced
by n( , τ ) d , the number density of PBHs with mass ratios between and
180 Black holes
integrate
+ d at time t = t τ . Similarly, L() takes the place of L(t). We then
over PBH mass as well as time t in order to obtain the total intensity.
0
Employing τ as an integration variable, we find:

1 ∞
d
Q = ct0 L pbh R̃(τ ) dτ
τf (c (τ )
3 + τ )ε
. (9.32)

Here ε ≡ (β + 2)/3 and c (τ ) is a minimum cut-off mass, equal to the mass of
the lightest PBH which has not yet evaporated at time τ . This arises because the
initial PBH mass distribution (9.2) must, in general, have a non-zero minimum
Mmin in order to avoid divergences at low mass. As soon as the lightest PBHs
have formed and started to evaporate, however, the cut-off begins to drop from its
initial value of c (0) = Mmin /M∗ . If Mmin is of the order of the Planck mass

as suggested by Barrow et al [38], then one finds that c (τ ) drops to zero well
before the end of the radiation-dominated era. Since we are concerned with times

later than this, we can safely set c (τ ) = 0 in what follows.

Using (9.15) for , we therefore rewrite equation (9.32) as
1 ∞
Q = Q pbh pbh R̃(τ ) dτ
d
τf 0 ( 3 + τ )ε (9.33)
where
ct0 ρcrit,0 L pbh
Q pbh = . (9.34)
kβ M∗
In this form, equation (9.33) can be used to put a rough upper limit on pbh from
the bolometric intensity of the background light [39]. Let us assume that the
Universe is flat, as suggested by most observations (chapter 4). Then its age t0
can be obtained from equation (2.70) as
2t˜0
t0 = . (9.35)
3H0
Here t˜0 ≡ t˜m (0) where t˜m (z) is the dimensionless function

1 −1 1 − m,0
t˜m (z) ≡ sinh . (9.36)
1 − m,0 m,0 (1 + z)3
Equation (9.35) is necessary if we are to obtain the right dependence on Hubble’s

constant h 0 in our results, and equation (9.36) will prove useful in specifying not
only the value of t0 but τf as well. Putting (9.35) into (9.34) and using equations
(2.19), (2.36), (9.7) and (9.30), we find that
(5.97 × 1034 g s−1 )π 2 hc4 ρcrit,0 t˜0

Q pbh =
768ζ(3)G H0kβ M∗3
= (2.3 ± 1.4)h 0 t˜0 kβ−1 erg s−1 cm−2 . (9.37)
Bolometric intensity 181
We are now ready to evaluate equation (9.33). To begin with we note that the
integral over mass has an analytic solution:
∞ ( 13 )(ε − 13 )
+ τ)
d 1
ε
= kε τ 3 −ε kε ≡ . (9.38)
0 ( 3 3(ε)
If we further specialize to the EdS case (m,0 = 1) then t˜0 = 1 and

equation (2.61) implies:
R̃(τ ) = τ 2/3 . (9.39)
Putting equations (9.38) and (9.39) into (9.33), we find that
1
1−ε 1 − τf2−ε
Q = Q pbh pbh kε τ dτ = Q pbh pbh kε . (9.40)
τf 2−ε
The parameter τf is simply obtained for the EdS case by inverting (9.39) to give
τf = (1 + z f )−3/2 . The subscript ‘f’ (by which we usually mean ‘formation’) is
here a misnomer since we do not integrate back to PBH formation time, which
occurred in the early stages of the radiation-dominated era. Rather we integrate
back to the redshift at which processes like pair production become significant
enough to render the Universe approximately opaque to the (primarily γ -ray)
photons from PBH evaporation. Following Kribs et al [37] this is z f ≈ 700.
Using this value of z f and substituting equations (9.37) and (9.38) into (9.40),
we find that the bolometric intensity of background radiation due to evaporating
PBHs in an EdS Universe is

 0 (β = 2)


 2.3 ± 1.4 erg s−1 cm−2 (β = 2.5)
Q = h 0 pbh × 6.6 ± 4.2 erg s−1 cm−2 (β = 3) (9.41)

 −1 cm−2

 17 ± 10 erg s (β = 3.5)
45 ± 28 erg s−1 cm−2 (β = 4).
This vanishes for β = 2 because kβ → ∞ in this limit, as discussed in section 9.4.
The case β = 4 (i.e. ε = 2) is evaluated with the help of L’Hôpital’s rule, which
gives limε→2 (1 − τf2−ε )/(2 − ε) = − ln τf .
All the values of Q in (9.41) are far higher than the actual bolometric EBL
intensity in an EdS Universe, which is 25 Q ∗ = 1.0 × 10−4 erg s−1 cm−2 from
(2.49). Moreover this background is already well accounted for by the integrated
light from galaxies. A firm upper bound on pbh (for the most natural situation
with β = 2.5) is therefore
pbh < (4.4 ± 2.8) × 10−5 h −1

0 . (9.42)
For harder initial equations of state (β > 2.5) the PBH density would have to
be even lower. PBHs in the simplest formation scenario are thus eliminated as
important dark-matter candidates, even without reference to the γ -ray spectrum.
182 Black holes
β=4
10 β = 3.5
β=3
1 β = 2.5
β = 2.2
)
Q*
−2
0.1
cm
−1
Q ( h0 erg s
0.01
0.001
0.0001
1e−05
1e−05 0.0001 0.001 0.01 0.1 1

Ωpbh
Figure 9.1. The bolometric intensity due to evaporating primordial black holes as a
function of their collective density pbh and the equation-of-state parameter β. We
have assumed that m,0 = bar + pbh with bar = 0.016h −2 0 , h 0 = 0.75 and
,0 = 1 − m,0 . The horizontal dotted line indicates the approximate bolometric
intensity (Q ∗ ) of the observed EBL.
This conclusion is based on equation (9.30) for PBH luminosity, which is valid
over the most important part of the integral. Corrections for higher and lower
masses will not alter the bound (9.42) by more than an order of magnitude.
The results in (9.41) assume an EdS cosmology. For more general
flat models containing both matter and vacuum energy, integrated background
intensity would go up, because the Universe is older, and down because Q ∝
pbh . The latter effect is the stronger one, so that our constraints on pbh will
be weakened in a model such as CDM (with m,0 = 0.3, ,0 = 0.7). To
determine the importance of this effect, we can re-evaluate the integral (9.33)
using the general formula (2.68) for R̃(τ ) in place of (9.39). We will make
the minimal assumption that PBHs constitute the only CDM, so that m,0 =
pbh + bar with bar given by (4.4) as usual. Equation (2.70) shows that the
parameter τf is given for arbitrary values of m,0 by τf = t˜m (z f )/t˜0 where the
function t˜m (z) is defined as before by (9.36).
Carrying out the integration in (9.33), we obtain the plots of bolometric
intensity Q as a function of pbh shown in figure 9.1. As before Q is proportional
to h 0 because it goes as both ρpbh = pbh ρcrit,0 ∝ h 20 and t0 ∝ h −1
0 . Since Q → 0
for β → 2 we have chosen a minimum value of β = 2.2 as being representative
of ‘soft’ conditions.
Spectral intensity 183
Figure 9.1 confirms that, regardless of cosmological model, PBH

contributions to the background light are too high unless pbh 1. The values
in equation (9.41) are recovered at the right-hand edge of the figure where pbh
approaches one, as expected. For all other models, if we impose a conservative
upper bound Q < Q ∗ (as indicated by the faint dotted line) then it follows that
pbh < (6.9 ± 4.2) × 10−5h −1 0 for β = 2.5. This is about 60% higher than the
limit (9.42) for the EdS case.
9.8 Spectral intensity

Stronger limits on PBH density can be obtained from the γ -ray background,
where these objects contribute most strongly to the EBL and where we have
good data (as summarized in section 8.7). Spectral intensity is found as usual
by substituting the comoving PBH number density (9.10) and SED (9.27) into
equation (3.5). As in the bolometric case, we now integrate over PBH mass
well as time τ = t/t0 , so that
as
∞
Iλ (λ0 ) =
ct0 1 2
4π τf
R̃ (τ ) dτ

c (τ )

n( , τ )F( R̃λ0 , ) d . (9.43)
Following the discussion in section 9.7 we set
c (τ ) = 0. In light of our
bolometric results it is unlikely that PBHs make up a significant part of the
dark matter, so we no longer tie the value of m,0 to pbh . Models with
m,0 > bar pbh must therefore contain a second species of cold dark matter
(other than PBHs) to provide the required matter density. Putting (9.10) and (9.27)
into (9.43) and using (9.7), (9.15) and (9.24), we find that
1 ∞ ( + τ ) d .
4 3 −ε
exp[λ / R̃(τ )λ ] − 1
Iλ (λ0 ) = Ipbh pbh R̃ −3 (τ ) dτ (9.44)
τf 0 pbh 0
Here the dimensional prefactor is a function of both β and λ0 and reads
(5.97 × 1034 g s−1 )(4π)5 G 3 M∗ ρcrit,0 t˜0

Ipbh =
3ζ (3)c5 kβ H0 λ40
−4
λ0
= [(2.1 ± 0.5) × 10−7 CUs]h 0 kβ−1 t˜0 . (9.45)
Å
We have divided through by the photon energy hc/λ0 to put the results in units of
CUs as usual. The range of uncertainty in Iλ (λ0 ) is noticeably smaller than that in
Q, equation (9.37). This is because Iλ (λ0 ) depends only linearly on M∗ whereas
Q is proportional to M∗−3 . (This, in turn, results from the fact that Iλ ∝ C ∝ M∗2
whereas Q ∝ L pbh ∝ M∗−2 . One more factor of M∗−1 comes from ∝ ρpbh /M∗
in both cases.) Like Q, Iλ depends linearly on h 0 since integrated intensity in
either case is proportional to both ρpbh ∝ ρcrit,0 ∝ h 20 and t0 ∝ h −1
0 .
184 Black holes
(a) (b)
10 10
SAS−2 (TF82) SAS−2 (TF82)
COMPTEL (K96) COMPTEL (K96)
(upper limits) (upper limits)
EGRET (S98) EGRET (S98)
β=4 β=4
β = 3.5 β = 3.5
1 1
β=3 β=3
β = 2.5 β = 2.5
β = 2.2 β = 2.2
Iλ ( CUs )
Iλ ( CUs )
0.1 0.1
0.01 0.01
1e−06 1e−05 0.0001 0.001 0.01 1e−06 1e−05 0.0001 0.001 0.01
λ(Å) λ(Å)
Figure 9.2. The spectral intensity of the diffuse γ -ray background from evaporating
primordial black holes in flat models, as compared with experimental limits from
the SAS-2, COMPTEL and EGRET instruments. The left-hand panel (a) assumes
m,0 = 0.06, while the right-hand panel (b) is plotted for m,0 = 1 (the EdS case).
All curves assume pbh = 10−8 and h 0 = 0.75.
Numerical integration of equation (9.44) leads to the plots in figure 9.2,

where we have set pbh = 10−8 . Following Page and Hawking [34] we have
chosen values of m,0 = 0.06 in the left-hand panel (a) and m,0 = 1 in the right-
hand panel (b). (Results are not strictly comparable in the former case, however,
since we assume that ,0 = 1 − m,0 rather than ,0 = 0.) Our results
are in good agreement with the earlier ones except at the longest wavelengths
(lowest energies), where PBH evaporation is no longer well described by a simple
blackbody SED, and where the spectrum begins to be affected by pair production
on nuclei. As expected the spectra peak near 10−4 Å in the γ -ray region. Also
plotted in figure 9.2 are the data from SAS-2 [40], COMPTEL [41] and EGRET
[42] (bold lines and points).
By adjusting the value of pbh up or down from its value of 10−8 in

figure 9.2, we can match the theoretical PBH spectra to those measured (e.g.
by EGRET), thereby obtaining the maximum value of pbh consistent with
Spectral intensity 185
observation. For β = 2.5 this results in

(4.2 ± 1.1) × 10−8 h −1 (m,0 = 0.06)
pbh < 0 (9.46)
(6.2 ± 1.6) × 10−8 h −1
0 (m,0 = 1).
These limits are three orders of magnitude stronger than the one from bolometric
intensity, again confirming that PBHs in the simplest formation scenario cannot
be significant contributors to the dark matter. Equation (9.46) can be put into the

form of an upper limit on the PBH number-density normalization with the help
of (9.15), giving

(2.2 ± 0.8) × 104h 0 pc−3 (m,0 = 0.06)
<
(3.2 ± 1.1) × 104h 0 pc−3 (m,0 = 1).
(9.47)
These numbers are in good agreement with the original Page–Hawking bound of
< 1 × 104 pc−3 , which was obtained for h 0 = 0.6 [34].
Subsequent workers have refined the γ -ray background constraints on pbh
and in a number of ways. An important development was the realization
by MacGibbon and Webber [32] that PBHs whose effective temperatures have
climbed above the rest energy of hadrons are likely to give off more photons by
indirect processes than by direct emission. This is because hadrons are composite
particles, made up of quarks and gluons. It is these elementary constituents
(rather than their composite bound states) which should be emitted from a very
hot PBH in the form of relativistic quark and gluon jets. Numerical simulations
and accelerator experiments indicate that these jets subsequently fragment into
secondary particles whose decays (especially those of the pions) produce a far
greater flux of photons than that emitted directly from the PBH. The net effect is
to increase the PBH luminosity, particularly in low-energy γ -rays, strengthening
the constraint on pbh by about an order of magnitude [36]. The most recent
upper limit obtained in this way using EGRET data (assuming m,0 = 1) is
pbh < (5.1 ± 1.3) × 10−9 h −20 [43].
Complementary bounds on PBH contributions to the dark matter have come
from direct searches for those evaporating within a few kpc of the Earth. Such
limits are subject to more uncertainty than ones based on the EBL because they
depend on assumptions about the degree to which PBHs are clustered. If there
is no clustering then (9.42) can be converted into a stringent upper bound on the

local PBH evaporation rate, ˙ < 10−7 pc−3 yr−1 . This, however, relaxes to
˙ º 10 pc−3 yr−1 if PBHs are strongly clustered [33], in which case limits
from direct searches could potentially become competitive with those based on
the EBL. Data taken at energies near 50 TeV with the CYGNUS air-shower array

have led to a bound of ˙ < 8.5 × 105 pc−3 yr−1 [44]; and a comparable limit of
˙ < (3.0 ± 1.0) × 106 pc−3 yr−1 has been obtained at 400 GeV using an imaging
atmospheric Čerenkov technique developed by the Whipple collaboration [45].
The strongest constraint yet reported has come from balloon measurements of the
186 Black holes
cosmic-ray antiproton flux below 0.5 GeV, from which it has been inferred that
˙ < 0.017 pc−3 yr−1 [46].
Other ideas have been advanced which could weaken the bounds on PBHs as
dark-matter candidates. It might be, for instance, that these objects leave behind
stable relics rather than evaporating completely [47]. This, however, raises a new
problem (similar to the ‘gravitino problem’ discussed in section 8.6) because such
relics would have been overproduced by quantum and thermal fluctuations in the
early Universe. Inflation can be invoked to reduce their density but must be finely
tuned if the same relics are to make up an interesting fraction of the dark matter
today [48]. A more promising possibility has been opened up by the suggestion of
Heckler [49, 50] that particles emitted from the surface interact strongly enough
above a critical temperature to form a black-hole photosphere. This would make
the PBH appear cooler as seen from a distance than its actual surface temperature,
just as the Solar photosphere makes the Sun appear cooler than its core. (In the
case of the black hole, however, one has not only an electromagnetic photosphere
but a QCD ‘gluosphere’.) The reality of this effect is still under debate [43]
but preliminary calculations indicate that it could reduce the intensity of PBH
contributions to the γ -ray background by 60% at 100 MeV and by as much as
two orders of magnitude at 1 GeV [51].
Finally, as discussed already in section 9.2, the limits obtained here can be
weakened or evaded if PBH formation occurs in such a way as to produce fewer
low-mass objects. The challenge faced in such proposals is to explain how a
distribution of this kind comes about in a natural way. A common procedure is to
turn the question around and use observational data on the present intensity of the
γ -ray background as a probe of the original PBH formation mechanism. Such an
approach has been applied, for example, to put constraints on the spectral index of
density fluctuations in the context of PBHs which form via critical collapse [37],
or inflation with a ‘blue’ or tilted spectrum [52]. Even if they do not exist, in
other words, primordial black holes provide a valuable window on conditions in
the early Universe, where information is otherwise scarce.
9.9 Higher-dimensional ‘black holes’
In view of the fact that conventional black holes are disfavoured as dark-matter
candidates, it is worthwhile to consider alternatives. One of the simplest of
these is the extension of the black-hole concept from the four-dimensional
(4D) spacetime of general relativity to higher dimensions. Higher-dimensional
relativity, also known as Kaluza–Klein gravity, has a long history and underlies
modern attempts to unify gravity with the standard model of particle physics
(see [53] for a review). The extra dimensions have traditionally been assumed
to be compact, in order to explain their non-appearance in low-energy physics.
The past few years, however, have witnessed a surge of interest in non-
compactified theories of higher-dimensional gravity [54–56]. In such theories
Higher-dimensional ‘black holes’ 187
the dimensionality of spacetime can manifest itself at experimentally accessible

energies. We focus on the prototypical five-dimensional (5D) case, although the
extension to higher dimensions is straightforward in principle.
Let us begin by recalling that black holes in standard 4D relativity are
described geometrically by the Schwarzschild metric. In isotropic coordinates
this reads
2
1 − G Ms /2c2r G Ms 4
ds =
2
c 2 2
dt − 1 + (dr 2 + r 2 d2) (9.48)
1 + G Ms /2c2r 2c2r
where d2 ≡ dθ 2 + sin2 θ dφ 2 . This is a description of the static, spherically-

symmetric spacetime around a pointlike object (such as a collapsed star or
primordial density fluctuation) with Schwarzschild mass Ms . As we have seen,
it is unlikely that such objects can make up the dark matter.
If the Universe has more than four dimensions, then the same object must
be modelled with a higher-dimensional analogue of the Schwarzschild metric.
Various possibilities have been explored over the years, with most attention
focusing on a 5D solution discussed in detail by Gross and Perry [57], Sorkin [58]
and Davidson and Owen [59] in the early 1980s. This is now generally known as
the soliton metric and reads:
2ξ κ 2
ar − 1 a 2r 2 − 1 ar + 1 2ξ(κ−1)
ds =2
c dt −
2 2
(dr 2 + r 2 d2 )
ar + 1 a 2r 2 ar − 1

ar + 1 2ξ
− dy 2 . (9.49)
ar − 1
Here y is the new coordinate and there are three metric parameters (a, ξ, κ)
rather than just one (Ms ) as in equation (9.48). Only two of these are
independent, however, because a consistency condition (which follows from the
field equations) requires that ξ 2 (κ 2 − κ + 1) = 1. In the limit where ξ → 0,
κ → ∞ and ξ κ → 1, equation (9.49) reduces to (9.48) on 4D hypersurfaces
y = constant. In this limit we can also identify the parameter a as a = 2c2 /G Ms
where Ms is the Schwarzschild mass.
We wish to understand the physical properties of this solution in four
dimensions. To accomplish this we do two things. First, we assume that Einstein’s
field equations in their usual form hold in the full five-dimensional spacetime.
Second, we assume that the Universe in five dimensions is empty, with no 5D
matter fields or cosmological constant. The field equations then simplify to
Ê AB = 0. (9.50)
Ê
Here AB is the 5D Ricci tensor, defined in exactly the same way as the 4D one
except that spacetime indices A, B run over 0–4 instead of 0–3. Putting a 5D
metric such as (9.49) into the vacuum 5D field equations (9.50), we find that we
188 Black holes
recover the 4D field equations (2.1) with a non-zero energy–momentum tensor

Ìµν . Matter and energy, in other words, are induced in 4D by pure geometry in
5D. It is by studying the properties of this induced-matter energy–momentum
tensor (Ìµν ) that we learn what the soliton looks like in four dimensions.
The details of the mechanism just outlined [60] and its application to solitons
in particular [61, 62] have been well studied and we do not review this material
here. It is important to note, however, that the Kaluza–Klein soliton differs from
an ordinary black hole in several key respects. It contains a singularity at its
centre; but this centre is located at r = 1/a rather than r = 0. (The point r = 0
is, in fact, not even part of the manifold, which ends at r = 1/a.) And its event
horizon, insofar as it has one, also shrinks to a point at r = 1/a. For these reasons
the soliton is better classified as a naked singularity than a black hole.
Solitons in the induced-matter picture are further distinguished from
conventional black holes by the fact that they have an extended matter distribution
rather than having all their mass compressed into the singularity. It is this feature
which proves to be of most use to us in putting constraints on solitons as dark-
matter candidates [63]. The time–time component of the induced-matter energy–
momentum tensor gives us the density of the solitonic fluid as a function of radial
distance:

c2 ξ 2 κa 6r 4 ar − 1 2ξ(κ−1)
ρs (r ) = . (9.51)
2π G(ar − 1)4 (ar + 1)4 ar + 1
From the other elements of Ìµν one finds that pressure can be written ps = 13 ρs c2 ,
so that the soliton has a radiation-like equation of state. In this respect it more
closely resembles a primordial black hole (which forms during the radiation-
dominated era) than one which arises as the endpoint of stellar collapse. The
elements of Ìµν can also be used to calculate the gravitational mass of the fluid
inside r :
2c2 ξ κ ar − 1 ξ
Mg (r ) = . (9.52)
Ga ar + 1
At large distances r 1/a from the centre the soliton’s density (9.51) and
gravitational mass (9.52) go over to
c2 ξ 2 κ 2c2ξ κ
ρs (r ) → Mg (r ) → Mg (∞) = . (9.53)
2π Ga 2r 4 Ga
The second of these expressions shows that the asymptotic value of Mg is, in
general, not the same as Ms [Mg (∞) = ξ κ Ms for r 1/a], but reduces to it in
the limit ξ κ → 1. Viewed in four dimensions, the soliton resembles a hole in the
geometry surrounded by a spherically-symmetric ball of ultrarelativistic matter
whose density falls off at large distances as 1/r 4 . If the Universe does have more
than four dimensions, then objects like this should be common, being generic to
5D Kaluza–Klein gravity in exactly the same way black holes are to 4D general
relativity.
Let us, therefore, assess their impact on the background radiation, using the
same methods as usual, and assuming that the fluid making up the soliton is in fact
composed of photons (although one might also consider ultrarelativistic particles
such as neutrinos in principle). We do not have spectral information on these so
we proceed bolometrically. Putting the second of equations (9.53) into the first
gives
G Mg2
ρs (r ) ≈ . (9.54)
8πc2 κr 4
Numbers can be attached to the quantities κ, r and Mg as follows. The first
(κ) is technically a free parameter. However, a natural choice from the physical
point of view is κ ∼ 1. For this case the consistency relation implies ξ ∼ 1
also, guaranteeing that the asymptotic gravitational mass of the soliton is close
to its Schwarzschild one. To obtain a value for r , let us assume that solitons are
distributed homogeneously through space with average separation d and mean
density ρ̄s = s ρcrit,0 = Ms /d 3 . Since ρs drops as r −4 whereas the number
of solitons at a distance r climbs only as r 3 , the local density of solitons is
largely determined by the nearest one. We can therefore replace r by d =
(Ms /s ρcrit,0 )1/3 . The last unknown in (9.54) is the soliton mass Mg (= Ms if
κ = 1). The fact that ρs ∝ r −4 is reminiscent of the density profile of the galactic
dark-matter halo, equation (6.14). Theoretical work on the classical tests of 5D
general relativity [64] and limits on violations of the equivalence principle [65]
also suggests that solitons are likely to be associated with dark matter on galactic
or larger scales. Let us therefore express Ms in units of the mass of the Galaxy,
which from (6.15) is Mgal ≈ 2 × 1012 M . Equation (9.54) then gives the local
energy density of solitonic fluid as

Ms 2/3
ρs c2 ≈ (3 × 10−17 erg cm−3 )h 0 s
8/3 4/3
. (9.55)
Mgal
To get a characteristic value, we take Ms = Mgal and adopt our usual values
h 0 = 0.75 and s = cdm = 0.3. Let us, moreover, compare our result to the
average energy density of the CMB, which dominates the spectrum of background
radiation (figure 1.2). The latter is found from (5.37) as ρcmb c2 = γ ρcrit,0 c2 =
4 × 10−13 erg cm−3 . We therefore obtain
ρs
≈ 7 × 10−6. (9.56)
ρcmb
This is of the same order of magnitude as the limit set on anomalous contributions
to the CMB by COBE and other experiments. We infer, therefore, that the
dark matter may consist of solitons but that they are probably not more massive
than galaxies. Similar arguments can be made on the basis of tidal effects and
gravitational lensing [63]. To go further and put more detailed constraints on
these candidates from background radiation or other considerations will require a
more detailed investigation of their microphysical properties.
190 Black holes
Let us summarize our results for this chapter. We have noted that standard
(stellar) black holes cannot provide the dark matter insofar as their contributions to
the density of the Universe are effectively baryonic. Primordial black holes evade
this constraint but we have reconfirmed the classic results of Page, Hawking and
others: the collective density of such objects must be negligible, for otherwise
their presence would have been obvious in the γ -ray background. In fact, we
have shown that their bolometric intensity alone is sufficient to rule them out as
important dark-matter candidates. These constraints may be relaxed if primordial
black holes form in such a way as to favour objects of higher mass, but it is
not clear that such a distribution can be shown to arise in a natural way. As
an alternative, we have considered black-hole like objects in higher-dimensional
gravity. If the world does have more than four dimensions, as suggested by
modern unifield-field theories, then these should exist and could be the dark
matter. This is consistent with the bolometric intensity of the EBL, but there are a
number of theoretical issues to be worked out before a more definitive assessment
of their potential can be made.
References
[1] Y B Zeldovich and I D Novikov 1966 Sov. Astron. 10 602
[2] Hawking S W 1971 Mon. Not. R. Astron. Soc. 152 75
[3] Hawking S W 1974 Nature 248 30
[4] Carr B J 1975 Astrophys. J. 201 1
[5] Carr B J 1976 Astrophys. J. 206 8
[6] Khlopov M Y, Malomed B A and Zeldovich Y B 1985 Mon. Not. R. Aston. Soc. 215
575
[7] Nasel’skii P D and Polnarëv A G 1985 Sov. Astron. 29 487
[8] Hsu S D H 1990 Phys. Lett. B 251 343
[9] Carr B J and Lidsey J E 1993 Phys. Rev. D 48 543
[10] Ivanov P, Naselsky P and Novikov I 1994 Phys. Rev. D 50 7173
[11] Garcı́a-Bellido J, Linde A and Wands D 1996 Phys. Rev. D 54 6040
[12] Yokoyama J 1997 Astron. Astrophys. 318 673
[13] Crawford M and Schramm D N 1982 Nature 298 538
[14] Kodama H, Sasaki M and Sato K 1982 Prog. Theor. Phys. 68 1979
[15] Hawking S W, Moss I G and Stewart J M 1982 Phys. Rev. D 26 2681
[16] Jedamzik K 1997 Phys. Rev. D 55 R5871
[17] Hawking S W 1989 Phys. Lett. B 231 237
[18] Polnarev A and Zembowicz R 1991 Phys. Rev. D 43 1106
[19] Niemeyer J C and Jedamzik K 1998 Phys. Rev. Lett. 80 5481
[20] Cline D B and Hong W 1992 Astrophys. J. 401 L57
[21] Cline D B, Sanders D A and Hong W 1997 Astrophys. J. 486 169
[22] Green A M 2002 Phys. Rev. 65 027301
[23] Hawkins M R S 1993 Nature 366 242
[24] Hawkins M R S 1996 Mon. Not. R. Astron. Soc. 278 787
[26] Cline D B 1998 Astrophys. J. 501 L1
[27] Green A M and Liddle A R 1999 Phys. Rev. 60 063509

[28] Green A M 2000 Astrophys. J. 537 708
[29] Carter B 1974 Phys. Rev. Lett. 33 558
[30] Page D N 1976 Phys. Rev. D 14 3260
[31] Page D N 1976 Phys. Rev. D 13 198
[32] MacGibbon J H and Webber B R 1990 Phys. Rev. D 41 3052
[33] Halzen F, Zas E, MacGibbon J H and Weekes T C 1991 Nature 353 807
[34] Page D N and Hawking S W 1976 Astrophys. J. 206 1
[35] Hawking S W 1975 Commun. Math. Phys. 43 199
[36] MacGibbon J H and Carr B J 1991 Astrophys. J. 371 447
[37] Kribs G D, Leibovich A K and Rothstein I Z 1999 Phys. Rev. D 60 103510
[38] Barrow J D, Copeland E J and Liddle A R 1991 Mon. Not. R. Astron. Soc. 253 675
[39] Overduin J M and Wesson P S 1992 Vistas Astron. 35 439
[40] Thompson D J and Fichtel C E 1982 Astron. Astrophys. 109 352
[41] Kappadath S C et al 1996 Astron. Astrophys. Suppl. Ser. 120 619
[42] Sreekumar P et al 1998 Astrophys. J. 494 523
[43] Carr B J and MacGibbon J H 1998 Phys. Rep. 307 141
[44] Alexandreas D E et al 1993 Phys. Rev. Lett. 71 2524
[45] Connaughton V et al 1998 Astropart. Phys. 8 178
[46] Maki K, Mitsui T and Orito S 1996 Phys. Rev. Lett. 76 3473
[47] MacGibbon J H 1987 Nature 329 308
[48] Barrow J D, Copeland E J and Liddle A R 1992 Phys. Rev. D 46 645
[49] Heckler A F 1997 Phys. Rev. D 55 480
[50] Heckler A F 1997 Phys. Rev. Lett. 78 3430
[51] Cline J M, Mostoslavsky M and Servant G 1999 Phys. Rev. D 59 063009
[52] Kim H I, Lee C H and MacGibbon J H 1999 Phys. Rev. D 59 063004
[53] Overduin J M and Wesson P S 1997 Phys. Rep. 283 303
[54] Wesson P S et al 1996 Int. J. Mod. Phys. A11 3247
[55] Arkani-Hamed N, Dimopoulos S and Dvali G 1998 Phys. Lett. B 429 263
[56] Randall L and Sundrum R 1999 Phys. Rev. Lett. 83 4690
[57] Gross D J and Perry M J 1983 Nucl. Phys. B 226 29
[58] Sorkin R D 1983 Phys. Rev. Lett. 51 87
[59] Davidson A and Owen D A 1985 Phys. Lett. B 155 247
[60] Wesson P S and Ponce de Leon J 1992 J. Math. Phys. 33 3883
[61] Liu H and Wesson P S 1992 J. Math. Phys. 33 3888
[62] Wesson P S and Ponce de Leon J 1994 Class. Quantum Grav. 11 1341
[63] Wesson P S 1994 Astrophys. J. 420 L49
[64] Liu H and Overduin J M 2000 Astrophys. J. 538 386
[65] Overduin J M 2000 Phys. Rev. D 62 102001
Chapter 10
Conclusions
Why is the sky dark at night, rather than being filled with the light of distant
galaxies? We have answered this question both qualitatively and quantitatively.
The brightness of the night sky is set by the finite age of the Universe, which
limits the number of photons that galaxies have been able to contribute to the
extragalactic background light. Expansion further darkens an already-black sky
by stretching and dimming the light from distant galaxies. This is, however, a
secondary effect. If we could freeze the expansion without altering the lifetime
of the galaxies, then the night sky would brighten by no more than a factor of
two to three at optical wavelengths, depending on the evolutionary history of the
galaxies and the makeup of the cosmological fluid.
What makes up the dark matter? This question still awaits a definitive
answer. However, we have seen that the machinery which settles the first
question goes some way toward resolving this one as well. Most dark-matter
candidates are not, in fact, perfectly black. Like the galaxies, they contribute
to the extragalactic background light with characteristic signatures at specific
wavelengths. Experimental data on the spectral intensity of this light therefore tell
us what the dark matter can (or cannot) be. It cannot be vacuum energy decaying
into photons, because this would lead to levels of microwave background radiation
in excess of those observed. It cannot consist of axions or neutrinos with
rest energies in the eV range, because these would produce too much infrared,
optical or ultraviolet background light, depending on their lifetimes and coupling
parameters. It could consist of supersymmetric weakly interacting massive
particles (WIMPs) such as neutralinos or gravitinos, but data on the x-ray and
γ -ray backgrounds imply that these must be very nearly stable. The same data
exclude a significant role for primordial black holes, whose Hawking evaporation
produces too much light at γ -ray wavelengths. Higher-dimensional analogues of
black holes known as solitons are more difficult to constrain but an analysis based
on the integrated intensity of the background radiation at all wavelengths suggests
that they could be dark-matter objects if their masses are not larger than those of
galaxies.
192
Conclusions 193
Table 10.1. Constraints on dark-matter candidates from background radiation.

Candidate Relevant waveband Prognosis Caveat(s)
Vacuum energy Microwave Ruled out Other decay modes
Multi-eV axions IR, optical Unlikely Rest mass/coupling
Light neutrinos UV Ruled out Rest mass/lifetime
WIMPs X-ray to γ -ray Plausible Lifetime/parameters
Black holes γ -ray Ruled out Formation mechanisms
Solitons (bolometric) Possible Theoretical parameters
The relevant waveband and overall prognosis for these dark-matter

candidates are summarized in table 10.1 along with possible caveats for each
case. Thus, theories in which the vacuum does not couple explicitly to matter
or radiation cannot be ruled out in this way. ‘Invisible’ axions and neutrinos
with very low rest masses would evade our bounds, although neutrinos in this
case could not provide a significant fraction of the dark matter. Very weakly-
coupled axions or long-lived neutrinos are also possible in principle, but face
other problems. Stable or extremely long-lived supersymmetric WIMPs remain
viable, subject to a number of theoretical parameters (such as the supersymmetry-
breaking energy scale) whose values have yet to be worked out. Primordial black
holes could conceivably play a larger role on the cosmic stage if they form in
such a way as to give them a more sharply-peaked mass distribution. Solitons are
the proverbial dark-horse candidates: they cannot be excluded, but a decisive test
must await more detailed investigation of their theoretical properties.
Table 10.1, of course, does not exhaust the list of possibilities. Many other
kinds of dark matter have been proposed, and most can be similarly constrained
by means of their contributions to the background light. We mention a few of
these without going into details. Very massive objects (VMOs) are dark-matter
clumps with masses in the range 100–105 M . They might, for instance, be
black holes which formed from the collapse of even more massive progenitors.
These high-redshift progenitors would also have been extremely luminous and it
is their light (redshifted into the infrared band) which puts the strongest limits
on VMOs as dark-matter candidates [1]. Warm-dark-matter (WDM) candidates
furnish another example. These are particles with rest energies in the keV range
whose existence would help resolve certain problems with conventional theories
of structure formation. While they are not part of most extensions of the standard
model, two possible WDM candidates are light gravitinos and a fourth generation
of ‘sterile’ neutrinos which interact more weakly than those of the standard model.
These could leave a mark in the x-ray background [2]. A number of other dark-
matter particles which have been proposed are very massive and would affect
primarily the γ -ray background. Among these are WIMPzillas, non-thermal
194 Conclusions
relics with rest energies in the range 1012–1016 GeV [3]. Gluinos [4] and axinos
[5], the supersymmetric counterparts of gluons and axions, are other examples.
High-energy γ -rays also provide us with a crucial experimental window on
some of the dark-matter candidates predicted by higher-dimensional unified-field
theories such as string and M theory [6] and brane-world scenarios [7, 8].
The dark night sky is thus our most versatile dark-matter detector. Its
potential in this regard is summed up nicely by a quote from the eighteenth-
century English amateur astronomer and cosmologist Thomas Wright. In his
1750 book titled An original theory or new hypothesis of the Universe, he wrote
as follows:
We may justly suppose that so many radiant bodies were not created
barely to to enlighten an infinite void, but to make their much more
numerous attendants visible. [9]
Of the attendants we have discussed, which is most likely to make up the dark
matter? It would be foolhardy to answer this too confidently, but we will close
this book with some opinions of our own. The results we have obtained here
support the view that the dark matter cannot involve ‘ordinary’ particles, by which
we mean those of the standard model of particle physics (including those which
may have fallen into black holes). We think it unlikely that baryons, neutrinos
or black holes will be found to form the bulk of the dark matter. Vacuum energy,
similarly, is a possibility only insofar as it does not decay into baryons or photons.
This leaves us with the other ‘exotic’ candidates. There is no doubt that the
frontrunners here are axions and supersymmetric WIMPs. These particles are
theoretically well motivated and physically well defined enough to be testable.
Both, however, arise in the context of particle-physics theories which do not yet
make room for gravity. It may be that the nature of the dark matter will continue
to resist understanding until a more fully unified theory of all the interactions is
at hand. In our view such a theory is likely to involve more than four dimensions.
Thus, while we are justified at present in directing most of our theoretical and
experimental efforts at WIMPs and axions, experience also suggests to us that
the dark matter may involve higher-dimensional objects such as solitons and
branes.
A final word is in order about the phrase ‘dark matter’, which while natural is
perhaps semantically unfortunate. It has been known since Einstein that mass and
energy are equivalent: in recent years particle physicists have found it increasingly
difficult to define the mass of a particle (as opposed, for example, to the energy
of a resonance); it is now acknowledged by workers in general relativity that the
energy of the gravitational field in that theory cannot be localized; and in higher-
dimensional extensions of general relativity, energy in spacetime is derived from
the geometry of higher dimensions in a way that intimately mixes the properties
of ‘matter’ and ‘vacuum’. The darkness of the night sky is consistent with other
cosmological data in telling us that there is something out there which gravitates
or curves spacetime. We are of the opinion that this substance, whatever we call it,
Conclusions 195
will eventually be identified. However, while some of it may be in a form which

is familiar, some may also be in a form which is truly unexpected.
References
[1] Carr B J and Sakellariadou M 1999 Astrophys. J. 516 195
[2] Abazajian K, Fuller G M and Tucker W H 2001 Astrophys. J. 562 593
[3] Ziaeepour H 2001 Astropart. Phys. 16 101
[4] Berezinsky V, Kachelrieß M and Ostapchenko S 2002 Phys. Rev. D 65 083004
[5] Kim H B and Kim J E 2002 Phys. Lett. B 527 18
[6] Benakli K, Ellis J and Nanopoulos D V 1999 Phys. Rev. D 59 047301
[7] Hall L J and Smith D 1999 Phys. Rev. D 60 085008
[8] Hannestad S and Raffelt G G 2001 Phys. Rev. Lett. 87 051301
[9] Jaki S L 2001 The Paradox of Olbers’ Paradox 2nd edn (Pinckney, MI: Real View
Books) p 112
Appendix A
Bolometric intensity integrals
A.1 Radiation-dominated models

This appendix is a summary of analytic solutions to the bolometric intensity
integrals (2.42) and (2.45) in chapter 2. We begin with the situation in which
the cosmic fluid is primarily composed of radiation-like matter (such as photons
or light neutrinos), so that r,0 = 0 but m,0 = ,0 = 0. Equations (2.42) and
(2.45) give
1+z f
Q dx
= (A.1)
Q∗ 1 x3 (1 − r,0 ) + r,0 x 2
1+z f
Q stat dx
= . (A.2)
Q∗ 1 x2 (1 − r,0 ) + r,0 x 2
We assume that the density of relativistic matter is positive. Equation (A.1) then
has two solutions, depending on whether the Universe is open or closed. For open
models (r,0 < 1),

Q 1 1 + r,0 z f (2 + z f ) r,0
= − +
Q∗ 2(1 − r,0 ) 2(1 − r,0 )(1 + z f )2 4(1 − r,0 )3/2

1 + r,0 z f (2 + z f ) + 1 − r,0
× ln
1 + r,0 z f (2 + z f ) − 1 − r,0

1 − 1 − r,0
× . (A.3)
1 + 1 − r,0
This simplifies considerably in the limit z f → ∞, for which

Q 1 r,0 1 − 1 − r,0
→ + ln . (A.4)
Q∗ 2(1 − r,0 ) 4(1 − r,0 )3/2 1 + 1 − r,0
196
Matter-dominated models 197
For closed models (r,0 > 1), the solution is different:

Q −1 1 + r,0 z f (2 + z f ) r,0
= + +
Q∗ 2(r,0 − 1) 2(r,0 − 1)(1 + z f ) 2 2(r,0 − 1)3/2

r,0 − 1 1
× cos−1
r,0 (1 + z f )

−1 r,0 − 1
− cos . (A.5)
r,0
This also simplifies in the limit z f → ∞, as follows:

Q −1 r,0 π −1 r,0 − 1
→ + − cos . (A.6)
Q∗ 2(r,0 − 1) 2(r,0 − 1)3/2 2 r,0
For the static case, (A.2) has a single solution for both open and closed models,
as follows:
Q stat 1 1 + r,0 z f (2 + z f )
= − . (A.7)
Q∗ 1 − r,0 (1 − r,0 )(1 + z f )
In the limit z f → ∞, this reduces to

Q stat 1 − r,0
→ . (A.8)
Q∗ 1 − r,0
These solutions are plotted in figure 2.2 of the main text.
A.2 Matter-dominated models

We proceed to models in which the cosmic fluid is dominated by dustlike rather
than relativistic matter, so that m,0 = 0 but r,0 = ,0 = 0. If vacuum
energy can be neglected (a supposition made by most cosmologists until quite
recently), then this is a good description of conditions in the present Universe.
The case m,0 ≈ 0.3 (i.e. the OCDM model) has received special attention in the
literature. (The other important special case is of course the Einstein–de Sitter
model with m,0 = 1, treated in section 2.7.) Equations (2.42) and (2.45) read
1+zf
Q dx
= (A.9)
Q∗ 1 x (1 − m,0 ) + m,0 x
3
1+zf
Q stat dx
= . (A.10)
Q∗ 1 x (1 − m,0 ) + m,0 x
2
We assume that the matter density is positive, as before. Equation (A.9) then has
two solutions, depending on whether the Universe is open or closed. For open
198 Bolometric intensity integrals
models (m,0 < 1),

Q 1 3m,0
= 1−
Q∗ 2(1 − m,0 ) 2(1 − m,0 )

1 + m,0 z f 1 3m,0
− −
2(1 − m,0 )(1 + z f ) 1 + z f 2(1 − m,0 )
2

3m,0 1 + m,0 z f − 1 − m,0
+ ln
8(1 − m,0 )5/2 1 + m,0 z f + 1 − m,0

1 + 1 − m,0
× . (A.11)
1 − 1 − m,0
In the limit z f → ∞ this simplifies to

Q 1 3m,0
→ 1−
Q∗ 2(1 − m,0 ) 2(1 − m,0 )

32m,0 1 + 1 − m,0
+ ln . (A.12)
8(1 − m,0 )5/2 1 − 1 − m,0
For closed models (m,0 > 1), the solution is somewhat different:

Q −1 3m,0 1 + m,0 z f
= 1+ +
Q∗ 2(m,0 − 1) 2(m,0 − 1) 2(m,0 − 1)(1 + z f )

1 3m,0 32m,0
× + +
1 + zf 2(m,0 − 1) 4(m,0 − 1)5/2

−1 1 + m,0 z f −1 1
× tan − tan . (A.13)
m,0 − 1 m,0 − 1
This simplifies, as follows, for z f → ∞:

Q −1 3m,0
→ 1+
Q∗ 2(m,0 − 1) 2(m,0 − 1)
2

3m,0 π −1 1
+ − tan . (A.14)
4(m,0 − 1)5/2 2 m,0 − 1
The solution of (A.10) for a static Universe also depends on the spatial curvature.
For open models,

Q stat 1 1 + m,0 z f m,0
= − +
Q∗ 1 − m,0 (1 − m,0 )(1 + z f ) 2(1 − m,0 )3/2

1 + m,0 z f + 1 − m,0 1 − 1 − m,0
× ln . (A.15)
1 + m,0 z f − 1 − m,0 1 + 1 − m,0
Vacuum-dominated models 199
In the limit z f → ∞ this reduces to

Q stat 1 m,0 1 − 1 − m,0
→ + ln . (A.16)
Q∗ 1 − m,0 2(1 − m,0 )3/2 1 + 1 − m,0
For closed models, the solution is

Q stat −1 1 + m,0 z f m,0
= + +
Q∗ m,0 − 1 (m,0 − 1)(1 + z f ) (m,0 − 1)3/2

−1 1 + m,0 z f −1 1
× tan − tan (A.17)
m,0 − 1 m,0 − 1
with the following limit as z f → ∞:

Q stat −1 m,0 π 1
→ + − tan−1 . (A.18)
Q∗ m,0 − 1 (m,0 − 1)3/2 2 m,0 − 1
These results are displayed in figure 2.3 of the main text.
A.3 Vacuum-dominated models

We move on finally to the models filled with vacuum energy, so that r,0 =
m,0 = 0 but ,0 = 0. If evidence continues to build for a vacuum-dominated
Universe, then these may provide a better description of large-scale dynamics
than the matter-dominated models of the previous section, despite the fact that
they contain no matter at all. They will, moreover, approximate the real Universe
more and more closely as matter density falls with time, while that of the vacuum
energy stays constant.
We consider values in the range −0.5 ,0 1.5 in this section, although
a negative cosmological term is somewhat disfavoured both theoretically (through
the interpretation of ,0 as a large-scale energy density) and observationally
(since negative values of ,0 imply very short lifetimes for the Universe). We
exclude the special cases ,0 = 0 (Milne model) and ,0 = 1 (de Sitter model)
which are treated in section 2.7. Equations (2.42) and (2.45) take the forms
1+zf
Q dx
= (A.19)
Q∗ 1 x ,0 + (1 − ,0 )x 2
2
1+zf
Q stat dx
= . (A.20)
Q∗ 1 x ,0 + (1 − ,0 )x 2
Equation (A.19) for the expanding model has the same form as (A.2) for the static
model with relativistic matter, and the solution (good for both open and closed
models) is
Q 1 1 + (1 − ,0 )(2 + z f )z f
= − . (A.21)
Q∗ ,0 ,0 (1 + z f )
200 Bolometric intensity integrals
As z f → ∞, this reduces (for models with ,0 1) to

Q 1 − 1 − ,0
→ (,0 1). (A.22)
Q∗ ,0
For models with ,0 > 1, limiting values of EBL intensity occur as z f →
z max (,0 ) < ∞, as discussed in section 2.7 in the main text. In the expanding
case equation (A.21) gives simply
Q 1
→ (,0 > 1) (A.23)
Q∗ ,0
where we have used (2.58) for z max . For the static case, equation (A.20) has three
possible solutions, depending whether ,0 < 0, 0 < ,0 < 1, or ,0 > 1.
For models of the first kind,

Q stat 1 −1 −,0 1 −1 −,0
= cos − cos
Q∗ −,0 1 − ,0 1 + z f 1 − ,0
(A.24)
which has the following limit as z f → ∞:

Q stat 1 π −1 −,0
→ − cos . (A.25)
Q∗ −,0 2 1 − ,0
For models in which 0 < ,0 < 1, the solution of (A.20) reads

Q stat 1 1 + (1 − ,0 )(2 + z f )z f − ,0
= ln
Q∗ 2 ,0 1 + (1 − ,0 )(2 + z f )z f + ,0

1 + ,0
× (A.26)
1 − ,0
which reduces to the following as z f → ∞:

Q stat 1 1 + ,0
→ ln . (A.27)
Q∗ 2 ,0 1 − ,0
When ,0 > 1, finally, we obtain

Q stat 1 ,0 − 1 + (1 − ,0 )(2 + z f )z f
= ln
Q∗ 2 ,0 ,0 + 1 + (1 − ,0 )(2 + z f )z f

,0 + 1
× . (A.28)
,0 − 1
Vacuum-dominated models 201
In the limit z f → z max , where the latter is defined as in (2.58), this expression
reduces to
Q stat 1 ,0 + 1
→ ln (A.29)
Q∗ 2 ,0 ,0 − 1
a result close to that in (A.27). Figure 2.4 in the main text summarizes these
results.
Appendix B
Dynamics with a decaying vacuum
This appendix outlines the solution of equations (5.15), (5.16), (5.18) and (5.19)
in the flat decaying-vacuum cosmology which is the subject of chapter 5. These
equations specify the scale factor R and densities ρm , ρr and ρv of matter,
radiation and decaying vacuum energy respectively. We apply the boundary
conditions R(0) = 0, ρm (t0 ) = ρm,0 and ρr (t0 ) = ρr,0 .
The key is the differential equation (5.19), which gives R(t). This may be
solved analytically in three different regimes:
Regime 1: ρr + ρv ρm + ρc
Regime 2: ρr + ρv ρm , ρc = 0 (B.1)
Regime 3: ρr + ρv ρm + ρc , ρc = 0
Here we distinguish between the time-varying part of the vacuum energy density
(ρv ) and its constant part (ρc ). Regime 1 describes the radiation-dominated era.
Regime 2 is a good approximation to the matter-dominated era in the Einstein–
de Sitter (EdS) model, with ρm,0 = ρcrit,0 . Regime 3 should be used instead when
ρm,0 < ρcrit,0 , as in the CDM model (with ρm,0 = 0.3ρcrit,0 ) or the BDM
model (with ρm,0 = 0.03ρcrit,0 ). For brevity we refer to this as the vacuum-
dominated regime.
B.1 Radiation-dominated regime

Equation (5.19) takes the following form in regime 1 (with x = x r ):
2
Ṙ 8π Gρv
= . (B.2)
R 3x r
Inserting (5.18) for ρv (R) then produces a differential equation for R(t), which
can be solved to yield:
1/4(1−xr)
32π Gαv (1 − x r )2 t 2
R1 (t) = . (B.3)
3x r
202
Matter-dominated regime 203
Substitution back into (5.18) then gives the vacuum energy density:
αx r
ρv,1 (t) = t −2 (B.4)
(1 − x r )2
where α ≡ 3/(32π G) = 4.47 × 105 g cm−2 s2 . The corresponding radiation

density is, from (5.16):

1 − xr
ρr,1 (t) = ρv,1 (t). (B.5)
xr
To check these results one can imagine that regime 1 extends to the present day.
Dividing through by the critical density ρcrit = 3H 2/8π G, we obtain
ρv,1 + ρr,1 ρv,1 /x r 1
v+r,1 (t) = = = t −2 . (B.6)
ρcrit ρcrit [2(1 − x r )H ]2
Differentiating (B.3) with respect to time, we find that H0 = 1/[2(1 − x r )t0 ].
Thus v+r,1 = 1 at the present time, which is consistent with our assumption of
flatness.
B.2 Matter-dominated regime

For regime 2, equation (5.19) takes the form (with x = x m ):
2
Ṙ 8π Gρm
= . (B.7)
R 3
From the conservation equation (5.15), matter density is given by
ρm (R) = αm R −3 (B.8)
where αm = constant. Inserting this into (B.7) produces a differential equation

for R(t) whose solution is:
R2 (t) = (6π Gαm )1/3 t 2/3 . (B.9)
Substitution back into (5.18) then gives the vacuum energy in terms of the
constants αv and αm as
ρv,2 (t) = αv (6π Gαm t 2 )−4(1−xm )/3. (B.10)
One does not need to solve for αv and αm in this expression. The dependence on t
is enough, as we show by using (5.16) to write down the corresponding radiation
density:
1 − xm
ρr,2 (t) = ρv,2 (t) = αr t −8(1−xm )/3 (B.11)
xm
204 Dynamics with a decaying vacuum
where αr is a new constant. This is fixed in terms of the present radiation density
8(1−x m )/3
ρr,0 by the boundary condition ρr,2 (t0 ) = ρr,0 , so that αr = ρr,0 t0 . The
vacuum density can then be written in terms of αr as well. Using (5.16) again one
has
xm αr x m
ρv,2 (t) = ρr,2 (t) = t −8(1−xm )/3 . (B.12)
1 − xm 1 − xm
The quantities ρr and ρv will not necessarily be continuous across the phase
transition t = teq , but R 4(1−x)ρv will be, since it is conserved by (5.17).
Matter density during regime 2 is found by substituting (B.9) directly into
(B.8). This gives
1 −2
ρm,2 = t . (B.13)
6π G
To check this one can divide through by the critical density to obtain m,2 (t).
Recalling that the lifetime of a flat, matter-dominated Universe is t0 = 2/3H0,
one finds simply
−2
t
m,2 (t) = . (B.14)
t0
Thus m,2 = 1 in the limit t → t0 as required.
We do not need an expression for matter density in the radiation-dominated
era. For the sake of the completeness of the plots in figure 5.1, however, we
can extend ρm (t) into regime 1 by imposing continuity across the phase transition
t = teq . (This is an extra condition on the theory, which conserves particle number
R 3 ρm rather than density ρm .) Using equations (B.3), (B.8) and (B.13), we obtain
−3/2(1−xr)
1 t
ρm,1 (t) = 2
. (B.15)
6π Gteq teq
This expression joins smoothly onto (B.13) at t = teq as desired.
B.3 Vacuum-dominated regime

The derivation in this section follows the same logic as in the previous section but
involves a little more work. Equation (5.19) takes the following form in regime 3
(with x = x m ):
2
Ṙ 8π G
= (αm R −3 + ρc ) (B.16)
R 3
where we have used equation (B.8). This is the same as the differential
equation (2.64) and can be solved by the methods of section 2.8. Assuming
ρc > 0, the result is
2/3
αm
R3 (t) = sinh( 6π Gρc t) . (B.17)
ρc
Vacuum-dominated regime 205
We then proceed as for regime 2, substituting (B.17) into (5.18) to obtain an

expression for ρv,3 (t) in terms of αm and αv . This gives the time-dependence of
the radiation density ρr,3 (as well as ρv,3 ). Application of the boundary condition
ρr,3 (t0 ) = ρr,0 then gives
√ −8(1−xm )/3
sinh( 6π Gρc t)
ρr,3 (t) = ρr,0 √ . (B.18)
sinh( 6π Gρc t0 )
The corresponding vacuum density is, from (5.16),

xm
ρv,3 (t) = ρr,3 (t). (B.19)
1 − xm
The matter density is similarly found by putting (B.17) into (B.8) and applying
the boundary condition ρm,3 (t0 ) = ρm,0 to obtain
√ −2
sinh( 6π Gρc t)
ρm,3 (t) = ρm,0 √ . (B.20)
sinh( 6π Gρc t0 )
For plotting purposes, this can be extended into regime 1 by imposing continuity
on ρm at teq , as in the preceding section. The result is
√ −2 −3/2(1−xr)
sinh( 6π Gρc teq ) t
ρm,1 (t) = ρm,0 √ . (B.21)
sinh( 6π Gρc t0 ) teq
This joins smoothly onto (B.20) at t = teq as desired.

The structural similarity between the results of this section and those of the
preceding one allows us to combine them into a single set of equations if we
define a function Ëm (t) such that

t (ρ = 0)
Ëm (t) ≡ sinh !√6π Gρ t " (ρc = 0). (B.22)
c c
The results of these two sections, finally, can

√ be expressed in terms of m,0 by
recalling that 8π Gρcrit,0 = 3H02 so that 6π Gρc = 32 1 − m,0 H0 for flat
models. Equations (5.20) through (5.26) and figure 5.1 in the main text summarize
these results.
Appendix C
Absorption by galactic hydrogen
In this appendix, we take up the problem (section 7.3) of calculating the

‘efficiency factor’ , or fraction of neutrino decay photons of energy E γ =
14.4 ± 0.5 eV which are absorbed by neutral hydrogen inside the galaxy before
they can escape and contribute to its luminosity. The number density of decay

photons is proportional to ν , equation (7.11).
Consider a decay photon emitted along the direction (θ, φ) at some point
P = (rp , z p ) in the halo, where r , z are cylindrical coordinates. We wish to find
out how much absorbing material lies between P and the edge of the halo, which
is described by the ellipsoidal surface:
2 2
x y z 2
+ + = ξ 2. (C.1)
r r h
Here x, y, z are rectangular coordinates, r = 8 kpc is the distance of the
Sun from the galactic centre, h = 3 kpc is the scale height of the halo and
ξ = rh /r is the halo scale factor. Outside the ellipsoidal surface we assume that
the neutrino number density drops off exponentially and can be neglected. The
value of ξ is fixed by specifying the total mass of the halo. We require this to be
Mh = (2±1)×1012 M , which from section 7.3 corresponds to rh = 70±25 kpc.
Thus ξ = 8.8 ± 3.1.
If d is the distance to the edge of the halo, then x = rp + d sin θ cos φ,
y = d sin θ sin φ and z = z p + d cos θ . Substitution into equation (C.1) yields
two roots:
2 2
B B C
d± = − ± − (C.2)
A A A
where A = h 2 sin2 θ + r 2 cos2 θ , B = h 2 r sin θ cos φ + r 2 z cos θ and C =
p p
h rp + r z p − r h ξ . The quantities d ± are the column lengths in the directions
2 2 2 2 2 2
Q + = (θ, φ) and Q − = (θ + π, φ + π).

In general, the fraction of (monochromatic) photons transmitted across a
distance d through an absorbing medium of ionization cross section σ and
206
Absorption by galactic hydrogen 207
number density n is given by f = e−τ where τ , the optical depth, is defined

d
by τ = 0 σ n dr . For neutral hydrogen the photoionization cross section is
σ = σH (13.6 eV/E γ )3 where σH = 6.3 × 10−18 cm2 [1]. The number density of
neutral hydrogen atoms in the Milky Way is roughly described by the following
function [2]: 2
1 z
n H (r, z) = n d f d (r ) exp − (C.3)
2 z d (r )
where:

r/r1 if r < r1
f d (r ) = exp[(r2 − r )/r3 ] if r > r2
1 otherwise
0.12 kpc + 0.023(r − r0 ) if r > r0
z d (r ) =
0.12 kpc otherwise.
Here n d = 0.35 cm−3 , r1 = 5 kpc, r2 = 13 kpc, r3 = 1.9 kpc and r0 = 9.5 kpc.
The constants σ = 5.3 × 10−18 cm2 (for 14.4 eV photons) and n d = 0.35 cm−3
may be usefully combined into a single absorption coefficient kd ≡ σ n d =
5730 kpc−1 . The large size of this number indicates that regions close to the
galactic plane (where n H ≈ n d ) will be totally opaque to decay photons.
The optical depths of columns P Q + and P Q − are:
d± 2
1 z
τp± (θ, φ) = kd f d (r ) exp − dr (C.4)
0 2 z d (r )
where x = rp + r sin θ cos φ, y = r sin θ sin φ, z = z p + r cos θ and

r = (x 2 + y 2 )1/2 . The probability that a photon emitted at P will reach the
edge of the halo (averaged over all decay directions) is then:
π π/2
e (rp , z p ) = 1
+ −
[e−τ p,+ (θ,φ) + e−τ p,− (θ,φ) ] sin θ dθ dφ. (C.5)
2π φ=0 θ=0
Thanks to cylindrical symmetry, we need integrate only over one quadrant of the
sky. It remains to average e over all points P in the halo. The average must be

weighted by the number density of decay photons ν to reflect the fact that more
photons are emitted near the centre ofthe halo than at its edges. Using (7.11) this
may be written as ν (rp , zp ) = [1 + (rp /r )2 (z/ h)2 ]−2 . So the final efficiency
factor is rh zmax (rp )
= 0 0
e (rp , z p ) ν (rp , z p )rp drp dz p
rh zmax (rp ) (C.6)
0 0 ν (rp , z p )rp drp dz p

where z max (rp = h ξ 2 − (rp /r )2 . Numerical integration of (C.6) produces the
values quoted in equation (7.17) of the main text. Figure C.1 illustrates the shape
208 Absorption by galactic hydrogen
10
-5
-10
-20 0 20
Figure C.1. Iso-absorption contours (light short-dashed lines) indicating the percentage
of decay photons produced inside the halo which escape absorption by neutral hydrogen
(solid lines) and reach the edge of the halo (heavy long-dashed lines). This plot assumes a
small dark-matter halo of radial size rh = 28 kpc.
of several ‘iso-absorption’ contours for a dark-matter halo with rh = 28 kpc [3],

considerably smaller than the ones we consider in chapter 7. In this case, the scale
sizes of the disc and halo are close enough that nearly 50% of the decay photons
are absorbed, cutting the luminosity of the neutrino halo in half.
References
[1] Bowers R and Deeming T 1984 Astrophysics II: Interstellar Matter and Galaxies
(Boston, MA: Jones and Bartlett) pp 383–91
[2] Scheffler H and Elsässer H 1988 Physics of the Galaxy and Interstellar Medium
(Berlin: Springer) pp 350–64
Index
Absorption density parameter (a ), 116, 124

by Earth’s atmosphere, 10 DFSZ, 116
by intergalactic dust, 63, 128, in clusters of galaxies, 124
135 in galaxy halos, 117
by neutral hydrogen, 135 invisible, 115
in the disc, 129, 130, 208 KSVZ (hadronic), 116
intergalactic, 63, 128, 136 multi-eV, 116
Active galactic nuclei (AGN), 164 production mechanisms
Age axionic string decay, 114
of metal-poor halo stars, 70 thermal, 115
of the galaxies, 7, 28 vacuum misalignment, 114
of the Universe (t0 ), 39, 70, 94, rest mass (m a ), 114, 116, 124
99, 180
Age crisis, 70
AMANDA, 148 B-band, 21, 50, 136
Anthropic principle, 80 Baade–Wesselink method, 68
Antipodal redshift, 80 Baryogenesis, 161
Apollo, 164 Baryon-to-photon ratio, 84, 98
Arrow of time, 91 Baryonic dark matter (BDM), 67
Asteroids, 2 density parameter (bar ), 67
Axinos, 194 Baryonic matter
Axion decay, 113 dark fraction ( f bdm ), 72
and cavity detectors, 115 density parameter (bar ), 72, 130
bolometric intensity due to, 119 BDM, see Baryonic dark matter
Feynman diagram, 115 Beige dwarfs, 73
halo luminosity due to, 118 Berkeley UVX, 142
lifetime (τa ), 115 Bianchi identities, 91
spectral intensity due to, 121 Big-bang theory, 6
Axions, 13, 76, 113, 146, 192 Big bounce
and red giants, 114, 116 in vacuum-dominated models, 32
and supernovae, 114, 116 with variable -term, 94
comoving number density (n a ), Black dwarfs, 72
115 Black holes, 13, 72, 170
coupling parameter (ζ ), 115, 122 higher-dimensional, see Solitons
209
210 Index
primordial, see Primordial black Cosmic microwave background

holes (CMB), 7, 13, 41
temperature, 178 angular power spectrum, 83, 94
Blazars, 164 temperature (Tcmb ), 103, 157
Boltzmann equation, 115, 146 Cosmic string loops, 78, 173
BOOMERANG, 83 Cosmological constant (), 68, 78
Brane-world scenario, 194 variable, see Variable
Brans–Dicke theory, 92 cosmological term
Brown dwarfs, 72 Cosmological-constant problem,
79, 85, 90, 113
CDM, see Cold dark matter Cosmological models
CDMS, 148 closed (spherical), 25, 83
Cepheid variables, 68 de Sitter, 30
Čerenkov detectors, 116, 149, 185 Eddington–Lemaı̂tre, 80
CGRO, 164 Einstein–de Sitter (EdS), 27, 30,
Charge-parity (CP), 113 47, 70, 73, 81, 98, 122, 133,
Circular velocity (vc ), 50, 118 152, 181
CMB, see Cosmic microwave flat (Euclidean), 24
background BDM, 48, 70, 72, 84, 98, 116
CNOC collaboration, 74 CDM, 48, 70, 81, 85, 98, 116,
COBE, 11, 49, 83, 95, 108, 122 152
Coincidence problem, 86 Milne, 31
Cold dark matter (CDM), 67, 70, OCDM, 48, 70
73, 85, 113, 146, 170 open (hyperbolic), 25, 85
density parameter (cdm ), 67, 76 radiation, 29
Comoving luminosity density (), SCDM, see EdS above
45 Cosmological scale factor (R), 17
at z = 0 (0 ), 21 as function of t, 37, 38, 71, 99,
relative to z = 0 (˜ ), 46 181
Comoving number density (n), 19 at present (R0 ), 17
at z = 0 (n 0 ), 117 relative to present ( R̃), 19
relative to z = 0 (ñ), 59 Critical density (ρcrit ), 24
COMPTEL, 153, 156, 160, 163, present value of (ρcrit,0 ), 24
164, 184 Curvature constant (k), 17, 26
Conformal rescaling, 92 Ê
Curvature scalar ( ), 17, 92
Conservation CUs, see Continuum units
of energy, 90, 91, 93 CYGNUS air-shower array, 185
of entropy, 103
of galaxies, 23 DAMA, 148
of particle number, 97 Dark energy, see Vacuum energy
Constrained minimal Dark matter, 13, 66, 192
supersymmetric model baryonic, see Baryonic dark
(CMSSM), 148 matter
Continuum units (CUs), 46, 108 cold, see Cold dark matter
Index 211
‘four elements’ of, 69 for perfect fluids, 22, 96

hot, see Hot dark matter for scalar fields, 93
warm, see Warm dark matter Equation of state, 22, 172
DASI, 83 dustlike, 22, 96, 172, 176
de Sitter model, see Cosmological radiation, 22, 96, 172
models scalar field, 23
DE-1, 142 stiff, 23, 172
Decaying-neutrino hypothesis, 77, superhard, 172
127, 142 vacuum, 22, 102
Dicke coincidences, 27 Equivalence principle, 189
DISM, see Interstellar medium, EURD, 142
diffuse Extragalactic background light
Dust (EBL), 11
and dimming of SNIa, 82 bolometric intensity of (Q), 19
and galaxy SEDs, 56 in de Sitter model, 30
and lensing statistics, 81 in EdS model, 30
extinction (ξ ), 136, 137 in Milne model, 31
Dust grains in models with matter and
fluffy silicate, 138 vacuum energy, 33
Mie scattering, 137 in radiation model, 30
PAH nanostructures, 138 typical value (Q ∗ ), 22
silicate/graphite, 137 spectral intensity of (Iλ ), 43
DUVE, 142 with blackbody SED, 53
with delta-function SED, 48
EBL, see Extragalactic background with Gaussian SED, 50
light with two-component SED, 59
Eclipsing binaries, 69
EDELWEISS, 148 Free-streaming neutrinos
EGRET, 153, 156, 160, 163, 164, bolometric intensity due to, 134
184 comoving luminosity density due
Einstein–de Sitter (EdS) model, see to, 134
Cosmological models Friedmann–Lemaı̂tre equation, 26
Einstein limit (,E ), 80
Einstein tensor, 92 Galaxies
Einstein’s ‘biggest blunder’, 79 faint blue excess, 58
Einstein’s field equations faint number counts, 81, 94
in 4 dimensions, 16 low surface brightness, 61
in 5 dimensions, 187 luminosity (L 0 ), 21
Electromagnetic cascades, 158, 160 luminosity density of, see
Electron–positron annihilation, 103 Comoving luminosity
Electroweak theory, 79 density
Energy–momentum tensor (Ìµν ), luminosity ratio (0 ), 57
17 merger parameter (η), 57
for induced matter, 188 morphological types, 55
212 Index
normal (quiescent), 55 bolometric intensity due to, 162

luminosity (L n ), 57 decay, see WIMP decay
number density (n n ), 57 decay lifetime (τg̃ ), 162, 166
number density of, see rest mass (m g̃ ), 161
Comoving number density spectral intensity due to, 162
peculiar velocities, 75
spectral energy distributions, 56 Hawking evaporation, 13, 170
starburst (active), 44, 55 HDM, see Hot dark matter
luminosity (L s ), 57 HEAO-1, 164
number density (n s ), 57 Higgsinos (h̃), 147, 159
starburst fraction ( f ), 57 Homestake, 78
at z = 0 ( f 0 ), 57 Hopkins Ultraviolet Telescope, 128
Galaxy clusters Hopkins UVX, 49, 121
baryon fraction, 74 Hot dark matter (HDM), 67, 77,
rich, 74 128
µ
Galaxy current (Jg ), 18 Hubble Deep Field (HDF), 46, 72
Galaxy distribution Hubble Key Project (HKP), 68
bias in, 81 Hubble Space Telescope (HST), 11,
power spectrum of, 75 49, 122
Galaxy evolution Hubble’s energy and number
luminosity density, 45 effects, 17
number density, 59, 75 Hubble’s law, 6
starburst fraction, 59 Hubble’s parameter (H ), 20
Gallex, 78 as function of t, 38
γ -ray band, 11, 147, 160, 163, 183 as function of z, 26, 119, 133
high-energy, 164 at present (H0), 20
low-energy, 164 normalized (h 0 ), 20, 70, 130
very-high-energy (VHE), 165 relative to present ( H̃ ), 21, 26
γ -ray bursts (GRBs), 173 HZT, 81
General relativity, 16
classical tests of, 189 ICS, see Inverse Compton
in higher dimensions, see scattering
Kaluza–Klein gravity Inflation, 23, 27, 73, 161, 186
GI theory, see Gravitational and reheat temperature, 161
instability theory Infrared (IR) band, 11, 12, 120
Gluinos, 194 far infrared (FIR), 109
Grand unified theories (GUTs), 79, near infrared (NIR), 48
115, 173 Integral flux, conversion to, 165
Gravitational instability (GI) theory, Intensity
75–77 of a single galaxy, 17
Gravitational lensing, 70, 80 of all galaxies, see Extragalactic
statistics of, 81, 94 background light
Gravitino problem, 161, 186 of galaxies in a shell
Gravitinos, 147, 193 bolometric, 19
Index 213
spectral, 42 Mass-to-light ratio, 74

‘reading-room’, 9, 39 Mass-to-light ratio (M/L), 72
Intergalactic medium (IGM), 135 Massive compact halo objects
Interstellar medium, diffuse (MACHOs), 72, 74
(DISM), 137 Matter density (ρm ), 23, 99
Inverse Compton scattering (ICS), Matter density parameter (m,0 ),
157, 160, 161 25, 67, 85, 130
Isothermal density profile, 151 evolution of (m ), 85
Matter–antimatter annihilation, 95
J-band, 49 Matter–radiation equality, epoch of
Jaffe profile, 118 (teq ), 97, 100
K-band, 49, 81 MAXIMA, 83
Kaluza–Klein gravity, 186 Metric tensor (gµν ), 17
Kitt Peak telescope, 12 MeV bump, demise of, 164
Kosmos, 164 Microlensing, 72
Milky Way
L-band, 49 dynamical mass, 74, 118, 130,
Large Magellanic Cloud (LMC), 151, 155
68, 72, 149 luminosity (B-band), 74
Large-scale structure formation, 73, Milne model, see Cosmological
81, 94, 127 models
Las Campanas telescope, 49, 122 Minimal supersymmetric model
Lee–Weinberg bound, 150 (MSSM), 147
LEP collider, 77
Light-travel time, 7 Naked singularities, 188
Lightest supersymmetric particle
Neutralinos, 147
(LSP), 146, 161
annihilation cross section, 150,
Long-baseline radio interferometry,
151
69
bolometric intensity due to, 152,
Look-back time, 19
155
LSND, 78
decay, see WIMP decay
Luminous matter
decay lifetime (τχ̃ ), 154, 156,
average density, 6
166
density parameter (lum ), 68, 71
in halos, see WIMP halos
rate of energy output, 5
Lyman-α absorbers, 72, 75, 84, 136 rest mass (m χ̃ ), 147
spectral intensity due to, 153,
M-theory, 194 155, 159
Mach’s principle, 92 Neutrino decay, 127, 150
MACHO survey, 72, 73, 173 Feynman diagrams, 129
MACHOs, see Massive compact lifetime (τν ), 127
halo objects spectral intensity due to, 141
Magnitude-redshift relation, 81 Neutrino halos, 128
MAP, 84 bolometric intensity due to, 133
214 Index
efficiency factor (), 129, 132, Pair production, 156, 158, 181
206 PBHs, see Primordial black holes
ellipsoidal density profile, 131, Peccei-Quinn (PQ) symmetry, 113
206 Phase transition, 97, 173
luminosity (L h ), 129, 133 Photinos, 147, 150, 151
mass (Mh ), 130 Pioneer 10, 46, 49, 121
spectral intensity due to, 140 Planck, 84
Neutrino oscillation, 78 Planck function, 51
Neutrinoless double-beta decay, 78 Planck’s law, 8
Neutrinos, 13, 67, 192 Primordial black holes (PBHs),
blackbody temperature (Tν ), 103 170, 192
bound, see Neutrino halos and γ -ray bursts, 173
density parameter (ν ), 67, 77, and γ -ray emission from halo,
78, 103, 130 173
free-streaming, see and MACHO events, 173
Free-streaming neutrinos and quasar variability, 173
in clusters of galaxies, 128 bolometric intensity due to,
number density (n ν ), 76 180–182
rest mass (m ν ), 77, 127 comoving number density, 175
sterile, 193 cut-off mass, 180
Neutron stars, 72 density (ρpbh ), 176
Non-minimal coupling, 92 density parameter (pbh ), 176,
Nucleosynthesis 181, 185
primordial, 71, 94, 98, 161 direct detection, 185
stellar, 5 equation-of-state parameter (β),
175
formation of
OAO-2, 49, 121, 142 during phase transition, 173
OGLE microlensing survey, 69 via collapse of small density
Olbers’ paradox, 1 fluctuations, 172
and absorption, 3, 64 via cosmic string loops, 173
and astronomy textbooks, 4 via critical collapse, 173, 186
and expansion, 4, 8, 35, 64, 192 via inflationary mechanisms,
and finite age, 4, 7, 35, 64, 192 172, 186
and galaxy distribution, 3 initial mass distribution, 172
and spatial curvature, 4 lifetime (tpbh ), 174
and the laws of physics, 4 local evaporation rate, 185
bolometric resolution of, 35 luminosity, 179
in de Sitter model, 31
normalization ( ), 175, 185
spectral resolution of, 63 photosphere, 186
Optical band, 10, 120 spectral intensity due to, 183, 184
Optical depth, 135, 207 stable relics, 186
in dust, 136 typical mass (M∗ ), 175
in neutral hydrogen, 136 Prognoz, 142
Index 215
Pure luminosity evolution (PLE), local energy density (ρs ), 189

52 Solrad-11, 142
Spectral energy distribution, 42
Quantum chromodynamics (QCD), blackbody, 52, 106, 177
79, 113 delta-function, 44
Quark and gluon jets, 185 Gaussian, 50, 121, 129, 150
Quark–hadron transition, 173, 176 inverse Compton scattering, 158
Quasars, 3 with cascades, 158
lensing frequency, 81 normalization, 42
spectra, see Lyman-α absorbers primordial black hole, 177
turnoff, 136 Static analogue
variability, 72, 173 bolometric intensity, 28
Quintessence, 23, 67, 92 spectral intensity, 43
Steady-state theory, 3
R-parity, 147, 154, 161, 167 Stefan’s law, 52, 103, 107
Radiation density (ρr ), 23, 99 String theory, 194
Radiation density parameter (r,0 ), Strong-CP problem, 113, 167
25, 67, 104 Subaru Deep Field, 61
Radio galaxy lobes, 75 Sunyaev–Zeldovich (SZ) effect, 68
Red clump stars, 69 Super-Kamiokande, 78
Red dwarfs, 72 Supernovae, Type Ia (SNIa), 81, 94,
Redshift (z), 6, 20 114, 164
Ricci tensor Supersymmetry, 127, 146, 174
Ê
in 4 dimensions ( µν ), 17 SUSY, see Supersymmetry
Ê
in 5 dimensions ( AB ), 187
Robertson–Walker metric, 17 Total density parameter (tot,0 ), 25,
26
SAGE, 78 Two Degree Field (2dF), 21
SAS-2, 153, 156, 160, 163, 164,
184 Ultraviolet (UV) band, 11, 150
Scalar field, 23, 90, 91 extreme ultraviolet (EUV), 128
Scalar–tensor theories, 92 far ultraviolet (FUV), 128
Schwarzschild metric, 187 near ultraviolet (NUV), 48, 128
SCP, 81 Unified field theory, 16
SED, see Spectral energy
distribution Vacuum decay, 13, 90
Selectrons, 147 and cosmic microwave
Sloan Digital Sky Survey (SDSS), background, 95
21, 61 bolometric intensity of, 106
Sneutrinos, 147 branching ratio (β), 103, 104
Soliton metric, 187 comoving luminosity density of
Solitons, 171, 192 (v,0 ), 105
density profile, 188, 189 coupling parameter (x), 97
gravitational mass, 188 into baryons, 95
216 Index
into radiation, 95, 97 indirect detection of, 148

luminosity (L v ), 104 pair annihilation of, 146
relativistic, 102 Feynman diagrams, 149
thermodynamical, 102 Whipple collaboration, 185
phenomenological model, 95 White dwarfs, 72
source regions, 101 Wien’s law, 44
spectral intensity of, 108 WIMP decay
Vacuum density (ρv ), 23, 99 branching ratio, 155, 158, 162
in terms of (ρ ), 24, 79, 96 Feynman diagrams, 154, 157
Vacuum density parameter (,0 ), one-loop, 147, 154, 165
25, 78, 82, 83, 85 tree-level, 147, 157, 161
evolution of ( ), 85 WIMP halos, 150
Vacuum energy, 67, 78, 192 cosmological density of, 151
Variable cosmological term, 80, 90 density profile, see Isothermal
Velocity dispersion, 121, 129, 150 density profile
Very massive objects (VMOs), 193 luminosity of, 150, 152, 154,
VIRGO consortium, 81 155, 158, 159, 162
Voyager 2, 142 WIMPs, see Weakly-interacting
massive particles
Warm dark matter (WDM), 193 WIMPzillas, 193
Weakly-interacting massive
particles (WIMPs), 13, 76, X-ray band, 11, 147, 160, 163, 193
146, 192 hard, 164
cosmological enhancement factor soft, 164
( f c ), 152
decay, see WIMP decay Zero-point field (ZPF), 67
direct detection of, 148 Zinos, 147, 151
in halos, see WIMP halos Zodiacal light, 46

Dark Sky Dark Matter

Uploaded by

Copyright:

Available Formats

Dark Sky Dark Matter

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dark Sky Dark Matter

Uploaded by

Copyright:

Available Formats

Dark Sky, Dark Matter

Series in Astronomy and Astrophysics

The Series in Astronomy and Astrophysics includes books on all aspects of

Other books in the series

Dust in the Galactic Environment, 2nd Edition

An Introduction to the Science of Cosmology

The Origin and Evolution of the Solar System

The Physics of the Interstellar Medium

Dust and Chemistry in Astronomy

The Physics of Interstellar Dust

Very High Energy Gamma Ray Astronomy

Dark Sky, Dark Matter

Institute of Physics Publishing

All rights reserved. No part of this publication may be reproduced, stored

British Library Cataloguing-in-Publication Data

ISBN 0 7503 0684 X

Library of Congress Cataloging-in-Publication Data are available

Series Editors: M Birkinshaw, University of Bristol, UK

Commissioning Editor: John Navas

Published by Institute of Physics Publishing, wholly owned by The Institute of

Typeset in LATEX 2ε by Text 2 Text, Torquay, Devon

4.6 Vacuum energy 78

J M Overduin and P S Wesson

The dark night sky

1.1 Olbers’ paradox

Figure 1.1. H W M Olbers (1758–1840).

1.2 A short history of Olbers’ paradox

times following the proposal of the steady-state cosmology by H Bondi, T Gold

1.3 The paradox now: stars, galaxies and Universe

1.4 The resolution: age versus expansion

1.5 The data: optical and otherwise

commonly referred to in the literature as the extragalactic background light

of photons we receive at the Earth is a combination due to various mechanisms,

The modern resolution and energy

2.1 Big-bang cosmology

2.2 The bolometric background

Êµν − 21 Êgµν − gµν = − 8πc4G Ìµν . (2.1)

The left-hand side of these equations describes the geometrical structure of

We now consider a multitude of galaxies. Suppose that these are distributed

where U µ ≡ (1, 0, 0, 0) is the galaxy four-velocity. The vanishing space

We will henceforth reserve the symbol n to denote comoving number density

dQ = dQ g n g dV = cn(t) R̃(t)L(t) dt. (2.12)

Here we have deﬁned the relative scale factor by

R̃(t) ≡ R(t)/R0 . (2.13)

2.3 From time to redshift

Here we have introduced a new quantity, Hubble’s parameter:

Here we have deﬁned the relative expansion rate by

H̃ (z) ≡ H (z)/H0. (2.22)

This can be measured experimentally by counting galaxies down to some

0 = (2.0 ± 0.2) × 108 h 0 L  Mpc−3

Q ∗ ≡ c0 /H0 = (2.5 ± 0.2) × 10−4 erg s−1 cm−2 . (2.25)

2.4 Matter and energy

Ìµν = (ρ + p/c2 )Uµ Uν + pgµν . (2.26)

electromagnetic radiation now is far below that of dustlike (pressure-free) matter.

ρ(t) = ρ0 [R(t)/R0 ]−3γ . (2.31)

In particular, for the single-component ﬂuids in (2.28):

ρr = ρr,0 (R/R0 )−4 (radiation)

2.5 The expansion rate

k = (R0 H0 /c)2 ( tot,0 − 1). (2.39)

H̃ (z) = [ r,0 (1 + z)4 + m,0 (1 + z)3 + ,0

0 = (2.0 ± 0.2) × 108 h 0 L Mpc−3

k = (R0 H0 /c)2 (tot,0 − 1). (2.39)

H̃ (z) = [r,0 (1 + z)4 + m,0 (1 + z)3 + ,0