Evolution of Matter in the Universe

Matter and energy content of the Universe control its geometry and expansion.
In the early Universe, the density was dominated by relativistic particles. At a
cosmic time of a second these were photons, neutrinos and electron pairs.


Their
influence on the expansion has become negligible in the present epoch, and now
baryons, non-baryonic “dark matter” and “dark energy” dominate the large-scale
dynamics and geometry of the Universe (Fig. 1). Baryons are the well-known
constituents of ordinary matter. For the existence of the other two components,
we have only indirect, but increasingly compelling evidence.
Although the influence of baryons on the overall dynamics and geometry of the
present Universe is relatively minor, their physical properties are unique.
Among the major forms of matter and energy that populate the present Universe,
only baryonic matter participates in all the physical forces known to us: the
strong forces (transmitted by gluons), the electromagnetic forces (transmitted by
photons), the weak forces (transmitted by the W± and Z0 bosons), and gravity
(transmitted by gravitons). These four physical interactions enable baryons to
self-organize, form a multitude of microscopic and macroscopic structures and,

indeed, create all the variety and beauty that we observe in the World.
The Expanding Universe
When in 1916, Albert Einstein formulated the theory of General Relativity,
space and time became objects of physical enquiry, along with matter and
radiation1
. For the first time this allowed development of scientific cosmologies
that made predictions that could be compared with observations. Thus General
Relativity provides the framework for cosmological models, but matter and
energy content determine which model is valid for our Universe. A variety of
observations made during the last fifty years show that we live in a Universe that
expanded out of an extremely dense and hot phase2
. We depict, in Figure 1, the
matter and energy components prevailing in the cosmos at three selected times.
Not only did the cosmic density decrease by many orders of magnitude duringthe time-span covered by Figure 1, but also the mixture of the matter and
energy components changed drastically.

The abundances in Figures 1a and 1b are quantitatively based on results of
elementary particle physics and established thermodynamic rules. In the present
Universe (Fig. 1c) the density of baryonic matter is well-established. Less well
established are as yet the densities of dark matter and dark energy, but progress
is expected from ongoing and future research.
The Epoch of the Quark-Gluon Plasma
The elementary particles of baryonic matter are quarks. There are six kinds of
quarks and six kinds of antiquarks. The nature of the strong interaction does not
allow quarks to occur in isolation3
, but they can exist as mesons (quark-antiquark
pairs), as baryons (three quarks), and as antibaryons (three antiquarks).
The quark-gluon plasma epoch is the earliest phase of the evolving Universe for
which we can investigate microscopic processes in the laboratory. At that time
the assemblage of quarks, antiquarks, gluons and other elementary particles
behaved somewhat like a liquid. This epoch ended when, at a cosmic time of
~100 microseconds, the expanding and cooling Universe was approaching a
density of 5×1016 kg/m3 (or 50 million tons per cubic centimetre) and a temper￾ature of 1012 K. The quark-gluon plasma became unstable and separated, form￾ing mesons, baryons and antibaryons. Mesons decayed while baryons and
antibaryons annihilated each other, all within microseconds. A very tiny fraction
of the baryons was spared, but this was enough for populating all the galaxies in
the Universe.

Symmetry Breaking in the Very Early Universe
The survival of some baryonic matter at the end of the quark-gluon plasma
epoch remains an unresolved problem of cosmology. An excess of baryons over
antibaryons could result from a difference in the behaviour of matter and anti￾matter. Andrei Sakharov, winner of the Nobel Prize for Peace in 1975, has
listed observations that could account for the excess of baryons. The violation
of the time-reversal invariance, found in the decay of neutral K-mesons is an
example. A necessary condition is that protons should decay, however slowly,
into mesons. No such proton instability has been found so far, but experiments
revealed that the lifetime of the proton far exceeds the age of the Universe. This
ascertains that even on a cosmic time scale the number of baryons will not
decrease by spontaneous decay.
Baryon-antibaryon annihilation is a strong interaction process. Therefore, in a
homogenous Universe only a totally insignificant amount of antibaryons should
have left the Big Bang. Indeed, even though the identification of primordial anti￾matter is complicated by interactions of cosmic rays with matter that produce
proton-antiproton pairs, experiments as well as observations have not given any
indication of the presence of primordial antibaryons. In fact, the fraction of
antiprotons found in cosmic rays is fully compatible with such a secondary ori￾gin.
Primordial Nucleosynthesis (0.1 s to 3 min in cosmic time)
During the epoch of primordial nucleosynthesis, lasting from ~100 milliseconds
to ~3 minutes in cosmic time, the physics is well-known, so that we can make
quantitative predictions for microscopic and macroscopic processes.
R.V. Wagoner, W.A. Fowler and F. Hoyle4 formulated the theory of Standard Big
Bang Nucleosynthesis (SBBN) in 1967. Based on Einstein’s General Relativity,
SBBN assumes a homogeneous and isotropic Universe during the epoch of
nucleosynthesis, and neglects degeneracies of leptons. When, with the LEP col￾lider at CERN, it was shown that there exist three neutrino flavours (Nν = 3), this
was included in the SBBN theory. Recent results of neutrino oscillation experi￾ments assure us that the rest masses of all these neutrinos are very low, low
enough to be negligible during the nucleosynthesis epoch. Thus, the baryonic
density remains the only important free parameter in the SBBN theory.
The sequence of events during the epoch of primordial nucleosynthesis is as fol￾lows: At a cosmic age of 10 milliseconds, the temperature had decreased to
1011 K. Mesons and heavier leptons had virtually all decayed, and only protons and
neutrons, the lightest variety of baryons, remained. As a result, energy
density and expansion rate were completely dominated by relativistic particles,
i.e. photons, neutrinos and electrons5
, with protons and neutrons being only
minor constituents (Fig. 1a). Since neutrons are heavier than protons, the
neutron/proton ratio decreased with decreasing temperature through the weak
interaction until, at a cosmic time of ~1 second and a temperature of ~1010 K, the
weak interaction became ineffective, and the neutron/proton ratio was frozen-in
at a value of one fifth. Afterwards, beta decay of the neutrons slowly decreased
this ratio further until all neutrons were bound in stable nuclei.
Nucleosynthesis, i.e. the fusion of protons and neutrons into deuterium and
heavier nuclei, effectively began when the temperature had decreased to
109 K at a cosmic time of ~100 seconds, and it was completed 200 seconds later.
Since all the nuclei of atomic mass A = 5 and A = 8 are extremely short-lived,
the production could not go beyond the isotopes of the lightest three
elements (Fig. 2). Of these, only deuterium (D or 2
H), the heavy isotope of
56
Figure 2. Origin of nuclei6
. Because all
nuclei having an atomic mass 5 or 8 are
extremely short lived, the nucleosynthe￾sis in the early Universe is limited to the
lightest three elements. All nuclei with
atomic mass A above 11 are produced in
stars. For species with mixed origin,
such as 7
Li, the relative proportions
change with time and location hydrogen, was created exclusively (>99%) during the first few minutes in the
life of the Universe.
The Universal Density of Baryonic Matter
The predicted Big Bang production of the isotopes of hydrogen and helium is
shown in Figure 3. 1
H and 4
He represent more than 99.9% of the total mass.
D and 3
He are rare and, as Figure 3 shows, their yields depend inversely on bary￾onic density. This is analogous to chemical reactions, where the yields of inter￾mediate products decrease with increasing supply of reacting partners. Since the
early 1970s, deuterium abundance measurements in the solar wind, meteorites,
Jupiter and the Galactic interstellar gas were used to derive the primordial abun￾dance of deuterium. Values of D/H in the range 3-5 × 10-5 were obtained from
which a universal baryonic density of
3-6 × 10-28 kg/m3 was calculated7
, cor￾responding to about 0.2 atoms per
cubic metre. A general consensus
existed on these values7 until, in 1994,
deuterium was measured by absorp￾tion of radiation from distant quasars
in intervening clouds of gas. Since the
investigated clouds are extremely old
and virtually free of heavier elements
(i.e. they have nearly “zero metallici￾ty”), their deuterium abundance
should be close to primordial. The
problem was that widely varying D/H
ratios were reported, ranging from
3×10-5 to 2×10-4. These results
reopened a broad discussion not only
on the usefulness of galactic data for
deriving primordial abundances, but
also on the reliability of the SBBN the￾ory for calculating the universal bary￾onic density.
This was the situation when in May 1997 ISSI convened the workshop on
“Primordial Nuclei and their Galactic Evolution”8
. It became clear at this
workshop that the low D/H ratios in distant clouds reported by D. Tytler and associates9
, and not the high values found by other authors, could be reconciled
with the 3
He and deuterium abundances in the Solar System and the present-day
Galaxy.
The 3
He and deuterium abundances11,12 in the Protosolar Cloud and the Local
Interstellar Cloud (Figs. 4 and 5) demonstrated that the principal effect of stellar
processing is the conversion of deuterium into 3
He with the sum, D+3
He remain￾ing nearly constant13 (Fig. 6). This was supported by new theoretical work14
showing that 3
He from incomplete hydrogen burning does not have a large effect
on the chemical evolution in the Galaxy.
The best current estimates of the primordial D/H and (D+3
He)/H ratios are com￾pared in Figure 3 with the theoretically predicted values. It is evident that both ratios give a universal baryon/photon ratio of (5.8±0.6)×10-10 and a present-day universal density of baryonic matter of σB = (4.1±0.4)×10-31 g/cm3 or about0.2 atoms per cubic metre. The baryon/photon ratio is one of the fundamentalnumbers of cosmology. So far, it is known only empirically. Any theory aboutthe earliest phases of the Big Bang will have to predict a value that is compati￾ble with the number derived from deuterium and 3
He.
Since the sum of D and 3He is nearly independent of galactic evolution, the
primordial baryonic density can be derived from this sum with little, if any,
extrapolation. As Figure 6 shows, (D+3
He)/H in the two galactic samples and in
the distant low-metallicity clouds are nearly the same. Thus, at the time of
primordial nucleosynthesis, the baryonic densities in the far-away regions of
these clouds and in our part of the Universe were the same, which is evidence
for a homogenous Universe at the time of primordial nucleosynthesis.
Deuterium as a Tracer of Natural Processes
From that time on, deuterium in the Universe has been continuously decreasing,
as stars are destroying, but not producing it. This “one-way” behaviour makes
deuterium a unique tracer for physical and chemical processes in nature16,17.
Deuterium in our bodies is authentic Big Bang stuff. Its relatively high abun￾dance of 2-3 grams in each of us is due to chemical enrichment in the cold
molecular cloud from which the Solar System formed16,18.
How Much Helium from the Big Bang, How Much from Stars?
The primordial abundance of 4
He is best obtained by extrapolating the helium
abundance measured in H II regions to “zero metallicity”, i.e. to vanishing O/H
or N/H ratios. For many years, several authors using this method consistently
derived a primordial helium mass fraction near 23%. Then, at the 1997 ISSI
workshop, Trinh Xuan Thuan and Yuri Izotov reported on their observations of
H II regions in blue dwarf galaxies (Fig. 7), and showed that the primordial mass
fraction had to be increased to 24.5%19, a value that has since been adopted. This
may seem like a small change but, as the following discussion will show, it is
significant for analyzing the physical processes in the early Universe.
The Big Bang production of 4
He depends only weakly on the baryonic density
(Fig. 3). Thus, by using the baryon/photon ratio as determined above, 4
He can be
used for testing the validity of the SBBN theory, or, to express it more general￾ly, the validity of the laws of physics under the extreme conditions that were
prevalent in the early Universe.

Stellar Production of Carbon and Heavier Elements
The gap in the sequence of stable nuclei at atomic masses 5 and 8 is
overcome by the 3-alpha nuclear reaction, producing 12C, the major isotope of
carbon20. Since this reaction involves three partners, a high density is required
for it to become effective. This condition is only fulfilled when stars have
evolved into red giants, with high enough central densities and with tempera￾tures of ~100 million degrees. Once 12C is present in a star, the synthesis con￾tinues to heavier elements as the star contracts further and increases its core tem￾perature.

The fusion of lighter nuclei into heavier ones continues up to the group
of elements around iron, which possesses the minimum free energy. Elements
beyond the iron group are produced by slow neutron capture in red giants, and
by the “r-process”, an extremely rapid capture of neutrons during super novae
explosions. The relative proportion of thorium, uranium and plutonium isotopes
in the Solar System proves that “r-process” synthesis was not restricted to some
violent early epoch, but has been going on throughout galactic history21.
Dark Matter....
In 1937 Fritz Zwicky discovered that the visible mass of the galaxies in the Coma
Cluster (Fig. 8: left) and other such clusters was far from sufficient to keep them
gravitationally bound, and he concluded that these clusters were held together by
a surplus of “dark matter” that astronomers could not readily account for.
During the last decades of the 20th century, it became increasingly clear that the
Universe harbours more gravitational attraction than could possibly be account￾ed for by the 0.2 atoms/m3 of matter that was derived from D and 3
He abundances At the ISSI workshop on “Matter in the Universe” held in March 2001, astronomi￾cal observations at various wavelengths including gamma-rays, and X-rays were
presented, along with results from gravitational lensing25. They all showed that non￾baryonic matter contributes most of the gravitational forces on the scale of clusters of
galaxies and even galaxies22, 26. Indeed, these clusters would fly apart, and galaxies -
including our own - would tend to disintegrate without the presence of an unknown
form of matter (Figs. 8: right and 9). The fluctuations in the Cosmic Microwave
Background (CMB) radiation demonstrated on a cosmic scale that, in addition to
baryonic matter, a non-baryonic form of matter must have come out of the Big Bang27.
....and Dark Energy
In recent years, observations of type IA supernovae explosions indicated that the
expansion of the Universe has been speeding-up during the past several billion
years.
 This effect is attributed to a “dark energy” with an equation of state that
combines positive energy with negative pressure28-30. In a medium with negative
pressure (p), the energy density (ε) decreases less rapidly, because the medium receives, but does not expend work. When p/ε is below –1/3, the negative pres￾sure overcomes the gravitational attraction, accelerating the expansion31.
In the years following the 2001 ISSI workshop, new and refined measurements
have firmed up the above conclusions30, 32, without fundamentally changing those
presented at the workshop25,26,33,34. Particularly, the CMB observations obtained
with the Wilkinson Microwave Anisotropy Probe (WMAP, Fig. 10) confirmed
the earlier results and improved quantitative predictions35.
The relative proportions of the major forms of energy in the Universe are shown
in Figure 1c. The share of baryonic matter is modest, but its density as derived
from deuterium and 3
He is very robust. If, for example, all the dark matter were
baryonic, the abundance of deuterium would be 80 times lower, and neither the
D absorption lines in Figure 4, nor the 3
He peak in Figure 5 would be noticeable.
Whereas dark and baryonic matter decelerate the expansion of the Universe, dark
energy tends to accelerate it 31. Since the densities of dark matter and baryonic
matter decrease more quickly than the density of dark energy, deceleration of the
expanding Universe turned into acceleration several billion years ago, and so it
will continue to expand into the distant future. That is our present understanding.

Building Cosmic Structure
Dark matter, not being affected by electromagnetic interactions, decoupled from
the photon gas very early and initiated the growth of cosmic structure long
before baryons could have done this. When at a cosmic time greater than
100,000 years baryons decoupled from photons, the baryons were drawn into
already existing blobs of dark matter and began to form the structures we observe. In places of strong enough con￾centration, baryonic matter, contracting
under its own weight, formed stars that
then produced carbon and heavier ele￾ments, essential ingredients of complex
molecules and crystals. These highly
organized systems of baryonic matter are
the crucial building blocks of comets,
solid planets and life.
The Nature of Dark Matter
The particles of dark matter have not yet been identified. Weakly Interacting
Massive Particles (WIMPs), but also virtually non-interacting light particles
(axions), are being considered. Experiments to detect WIMPs produced by
accelerators or natural WIMPs are underway (Fig. 11). Such measurements
could provide information on the mass and interaction properties of the dark￾matter particles and on their temperature in the solar neighbourhood. These
properties of the dark matter allow predictions regarding the evolution of medi￾um-scale structures, such as the number of dwarf galaxies in relation to fully
grown galaxies, the amount of baryonic matter falling quasi-continuously into
our own Galaxy, or the concentration of matter towards the centre of galaxies
and clusters. Such predictions are important for comparison with observations.
The direct detection of dark-matter particles is, in principle, made easier by its
concentration in the galactic potential well. For the neighbourhood of the Sun,
an energy density of 0.3 GeV/cm3 of dark matter is derived from the width and
depth of this well33. This corresponds to 580 grams of dark matter for the whole
volume of Earth. Comparison with the mass of the Earth of 6×1024 kilograms
shows just how much more locally concentrated and structured baryonic matter
is in the Universe, a direct consequence of the strong and electromagnetic forces.