A Brief History of the Universe

Last: 14. Black Holes and Quasars Next: 16. REVIEW

Looking out into space is also looking back in time. We see evidence the expansion of the universe is accelerating, and clear proof that the early universe was hot, dense, and almost perfectly smooth. Even further back we come to the origin of helium, the formation of matter itself, and finally to hints of an initial phase of violent expansion which forever hides the events giving rise to our universe.


Redshift and Expansion

Standard candles more powerful than Cephied variables enable us to measure distances and study the expansion of the universe on very large scales.

We find that Hubble's law works very well out to distances of many hundred Mpc.

  Hubble diagram

Large-Scale Structure

Using Hubble's law to compute distances from redshifts, we have charted positions of galaxies on enormous scales. These charts show the `frothy' pattern of galaxies.   2dF galaxy redshift survey
The 2dF Galaxy Redshift Survey [ANU]

Structure Formation

This frothy structure can be produced by computer simulations which start with an almost perfectly smooth distribution of dark matter and allow gravity to pull things together.

For the simulations to work, however, the dark matter must start out `cold', meaning zero random motion at very early times. This is an important clue about the nature of dark matter.

  Simulation of dark matter clustering
The Millennium Simulation [MPA]

It's also necessary to start with tiny variations in the matter density; a perfectly uniform universe would remain uniform since each particle would feel the same pull in all directions.

Interpretation of Redshift

Hubble's law fails for extremely distant galaxies. Galaxies with redshifts z > 6 are known; the simple equations we've been using imply that any object with z > 1 has a recession speed greater than the speed of light, and a distance greater than the distance light can travel in the time the universe has existed!

Instead, we interpret large redshifts as measuring how much the universe has expanded since the light was emitted. For example, if we observe a galaxy with a redshift z = 2, we know that the universe has expanded by a factor of 1 + z = 3 since the light from that galaxy was emitted.    

This example illustrates another problem with the simple equations: by `distance' do we mean the distance when the light was emitted, or the distance now? In contrast, there's no ambiguity with interpreting 1 + z as the expansion factor.

Supernovae as Standard Candles

To directly measure distances to far-away galaxies we need more powerful standard candles. Type Ia Supernovae — thermonuclear explosions triggered when white dwarf stars collpase — can be used as `standard bombs'; they are
  • very luminous, and therefore visible at extremely large distances
  • easy to recognize — a bright point of light suddenly appearing in a distant galaxy is almost certainly a supernova
  • well-standardized, because they occur when the mass of a white dwarf reaches 1.4 MSun
  SN Ia in a distant galaxy

With such powerful candles it became possible to measure changes in the expansion rate of the universe. We expected to find that the expansion was slowing down due to gravity. Instead...

Expansion Rate

...the expansion is speeding up!

To prevent gravity from slowing the expansion, some form of `antigravity' is needed. The most promising possibility is a repulsive force associated with empty space itself (Einstein's `cosmological constant'). This force becomes more and more effective as the universe expands, because the expansion creates more and more empty space.

A possible recipe for the universe contains about 5% ordinary matter, 25% dark matter, and 70% `dark energy' — in effect, energy associated with this form of antigravity.

  Accelerating expansion
Dark energy [physicsweb]

The Microwave Background

The oldest light we can detect comes from a time long before the first stars and galaxies formed. This is an almost perfectly uniform sea of microwave photons coming from all directions. The spectrum of this radiation is a perfect match to a `black-body' with a temperature of 2.725 K — the temperature of the universe today.   COBE/FIRAS data and black-body spectrum
Ned Wright's Cosmology Tutorial [UCLA]

Nothing in the present universe produces radiation so perfectly matched to a black-body spectrum. This radiation is a relic of a time when the entire universe was a black body, with a temperature of about 3000 K.


The temperature of the universe was higher at early times, exceeding ∼3000 K at a redshift z > 1000. Above this temperature, hydrogen atoms are ionized into protons and electrons, and the electrons scatter light waves (before). This scattering makes the universe a very good black body.

As the universe expanded it cooled, and at z ≈ 1000 the electrons and protons (re)combined to form hydrogen atoms (after). This cleared the `fog' of free electrons and allowed the light to travel freely.

Recombination occurred when the universe was about t ≈ 400,000 yr old.

Before and after H atoms form

Microwave Fluctuations

Microwave fluctuations
WMAP Data Product Images [NASA]

The cosmic microwave radiation is almost equally bright in all directions, with slight fluctuations — just a few parts in 100,000. This shows the early universe was not perfectly uniform; the variations are just large enough to form large-scale structure.

Concordance in Cosmology

Fluctuations in the microwave radiation provide key information about cosmology. For example, `lumps' in the microwave sky tend to have angular sizes of about 1° (first peak at right), implying that the universe has no overall curvature. The curve represents a `best bet' model of the universe:
  • t0 = 13.7 Gyr old
  • H0 = 71 km ⁄ sec ⁄ Mpc
  • 4% ordinary matter, 23% dark matter, 73% `dark energy'
  Microwave fluctuations
WMAP Data Product Images [NASA]

These parameters are very close to the results obtained from other measurements!

Primordial Nucleosynthesis

If we follow the expansion back to extremely early times, the temperature of the universe is expected to exceeded the highest temperatures found in the cores of stars. Is there any evidence for such extreme temperatures?

At temperatures of many billions of degrees, atomic nuclei fragment into individual neutrons (n) and protons (p). Reactions involving these particles as well as electrons (e) and neutrinos (ν) maintain equal numbers of ps and ns:

n + ν ↔ p + e

As the temperature fell, ps began to outnumber ns for two reasons: first, ns are ∼0.2% heavier than ps (and thus harder to make), and second, free ns spontaneously decay into ps with a half-life of ∼10 min (that is, half the remaining ns decay every 10 min).

Light Element Formation

When the temperature of the universe had fallen to 109 K (∼100 sec after the universe began), ps outnumbered ns by a ratio of about 7:1. At this temperature, ps and ns could start combining to build up atomic nuclei. The first reaction produced `heavy hydrogen' (2H, a.k.a. deuterium or D) via

p + nD + γ

As in stars, D gets involved in other reactions as soon as it forms. Thus almost all of the D quickly combined to produce helium (4He).

  Nucleosynthetic reactions

Within a few minutes, all the available ns — about 12.5% by mass — had combined with an equal mass of ps to form 4He. Other reactions produced tiny amounts of other light elements, mostly leftover D and lithium (7Li).

Light Element Abundances

A robust prediction of the hot universe idea is that 4He makes up ∼25% (by mass) of the ordinary matter in the universe. This is exactly what observations show. Almost all of the 4He in the universe was made in the first few minutes; only a small amout has been made in stars since then.

The amount of D (and 7Li) produced depend on the total amount of ordinary matter in the universe: more matter yields less D.

  BBN predicitons
Galaxies and Cosmology, Ch. 6, Fig. 13

The observed abundances of D and 7Li both indicate that ordinary matter makes up only a small fraction of the universe — about 3 or 4%. This is in good agreement with the matter fraction in the `best bet' model already described.

Cosmic Abundances

Some elements are more common than others; the pattern of abundances reflects stellar processes:
  • Li, Be, B are rare; they are not produced in stars.
  • C and O are common products of Type II SN, as are many other light elements with even numbers of protons.
  • Iron-group elements (Fe, Co, Ni) are common products of Type Ia SN.
  • Elements beyond the iron group require energy to create; they are rare.
  • Pb is more common than other heavy elements; it's produced when heavier ones (eg, Th, U) decay.
  Abundances of elements
Abundance of Elements [GreenSpirit]

Origin of Matter

Following the expansion back to even earlier times, we expect to encounter even higher temperatures. The physics of temperatures as high as 1016 K can be studied in particle accelerators, and we can use this knowledge to guess what happened in the very early universe.

When the universe was 10-8 sec old it was so hot (1016 K) and dense that even protons and neutrons would have fragmented into smaller particles called quarks (q). At these temperatures, the radiation (γ) was so energetic it made quark-antiquark pairs:

γ  +  γ  ↔  q  + 


This reaction does not favor either matter or antimatter; both appear on the same footing. Thus at this stage the universe contained equal amounts of matter and antimatter — quite unlike the situation today!


At some stage as the universe expanded and cooled down, a slight imbalance between matter and antimatter developed. The mechanism which created this imbalance is not well understood, but some particle-physics theories do allow reactions which slightly favor matter over antimatter.

If these theories are correct then quarks would outnumber antiquarks by a tiny factor — roughly one extra quark for every 109 quark-antiquark pairs. As the temperature fell the radiation became less energetic, and no longer be able to create quarks and antiquarks. Once this happened most of the quarks and antiquarks annihilated each other:

q  + 

 →  γ  +  γ

The slight excess of quarks surviving this mutual massacre could go on to form the matter we see around us.

Quark Confinement

Six types of quarks are known to exist, but ordinary matter contains only two: the `up' quark (u) and the `down' quark (d). These particles have fractional charges of +(2/3) and -(1/3), respectively.

When the universe was 10-6 sec old it had a temperature of 1013 K. At this stage, individual protons (p) and neutrons (n) could begin to form, via the reactions

u + u + dp

d + d + un

  Internal structure of nucleons

Since this time, quarks have been confined inside ps, ns, and other sub-atomic particles.


The original `Big-Bang' model for the formation of the universe has a few unsolved questions:

Some people simply assume that the universe began uniform, flat, and free of monopoles. However, there is a speculative explanation which answers all three of these questions: an early stage of extremely rapid expansion known as inflation.

The Inflationary Period

When the universe was only about 10-35 sec old, Einstein's cosmological `constant' could have become very large, creating a powerful antigravitational force. This would have caused the universe to expand exponentially, increasing its size by a factor of 1050 or more.   Inflation vs standard theory

Inflationary Answers

If inflation actually occurred, it would provide answers to the horizon, flatness, and monopole problems:

In addition, inflation could account for the tiny variations in density needed to produce the microwave background fluctuations and the large-scale structure of galaxies and clustwers we observe today.

The Multiverse

If inflation happened once, it could and probably did happen more than once, creating multiple universes with different physical conditions and even different laws of physics. We may be living in one of the universes where life can exist, while other universes may be less hospitable.   Chaotic inflation of multiple universes

Because light moves too slowly to carry information from one universe to another, we have no way of proving that our universe `descended' from another one. In solving the horizon, flatness, and monopole problems, inflation permanently erased any trace of earlier conditions. But if we could show that our universe has spawned others, it might be plausible that ours is not the first or only universe to exist.

Cosmic History

Returning to less speculative ideas, we can now present a history of the universe from early times to the present.

Chaotic inflation of multiple universes

We can contrast this history with the one presented the first week of class. Then, our perspective was centered on events leading to our planet and human life. Now, we see events as the universe itself might recount them.   Log scale showing ages of things
Last: 14. Black Holes and Quasars Next: 16. REVIEW

Joshua E. Barnes (barnes@ifa.hawaii.edu)
Last modified: November 30, 2006
Valid XHTML 1.0 Transitional
z = 0.25 0.5 1.0 2.0 4.0 8.0
Radial distance (Mpc) 986 1856 3271 5172 7204 9018
Light travel time (Gyr) 2.88 4.95 7.62 10.18 11.93 12.83