Dark matter and dark energy

Observations continue to indicate that the Universe is dominated by invisible components — dark matter and dark energy. Shedding light on this cosmic darkness is a priority for astronomers and physicists.

What is the composition of the Universe?

Deep view — a slice of the Hubble Space Telescope's view of the visible Universe. Credit: NASA & A. RIESS (STSCI)

In terms of their contribution to the mean energy density, the contents of the Universe are approximately 75% dark energy, 20% dark matter and 5% normal (atomic) matter, with smaller contributions from photons and neutrinos. These measurements rely on the validity of the hot Big Bang model, general relativity and the cosmological principle (that the Universe is uniform on the largest scales). The breadth and depth of experiments and observations that support these underlying tenets give us confidence that this model of the cosmos has a solid foundation.

What is the evidence for dark matter?

We can infer the presence of dark matter through indirect methods, despite not being able to see it (Fig. 1, overleaf). Newton's laws state that the mass of a body can be determined by the motion of its satellites. Thus, it has been calculated that the mass of galaxy clusters is far larger than that of their constituent galaxies, and that the mass of galaxies is far larger than the combined mass of their constituent stars and interstellar gas. And there is plenty more corroborating evidence. Yet there is very good reason to expect that this extra 'stuff' is not normal matter. Such an abundance of normal matter would be difficult to conceal from the prying eyes of astronomers, and would furthermore leave a distinct signature in the cosmic microwave background (CMB) radiation (relic radiation from the Big Bang), and in the properties of galaxies and clusters, that is simply not seen.

Figure 1: Dark matter and how it might be detected.

a, b, The rotational velocity of its stars and gas indicates that the Milky Way is embedded in a dark-matter halo extending out to a radius of about 200 kiloparsecs (kpc). High-energy γ-rays may be produced by the annihilation of dark-matter particles in neighbouring dwarf spheroidal galaxies and near the Galactic Centre, where the dark-matter density is expected to be highest. The dark-matter density may also be enhanced in the tidal stream of matter that trails from the Sagittarius dwarf galaxy and entangles the Milky Way. c, Earth's orbit through the Galactic dark-matter halo may produce a modulation of the dark-matter flux identified in experiments that aim to detect dark matter directly: a smaller (by about 10%) flux is expected when Earth moves in the same direction as the dark-matter 'wind' from the Galactic halo (in winter) than when it moves against it (in summer).

Why can't we conclude that Newton's laws break down at the distance scales of galaxies or clusters?

This might have been a reasonable hypothesis a few decades ago. However, any alternative gravity theory that accounts for the observed galaxy and cluster dynamics must also explain the vast body of data on gravitational lensing (the deflection of light from distant sources), the CMB and large-scale structures. At the same time, it must also satisfy a suite of precise constraints on gravity obtained within the Solar System.

How much dark matter is there nearby?

The orbital velocities of stars in the Milky Way suggest a mean mass density of dark matter in our neighbourhood of about a third of a proton mass per cubic centimetre. For perspective, this is 106 times greater than the mean density of the cosmos, but 24 orders of magnitude smaller than the mean density of water. Because whatever objects make up dark matter move in the same Galactic gravitational potential well as stars, we know that they must be moving with velocities of about 200 kilometres per second. Earth's orbit around the Sun implies that the amount of dark matter incident on the Earth varies by about 10% from summer to winter (Fig. 1). Furthermore, the distribution of galactic dark matter may not be smooth; galaxy formation is an ongoing process, and computational studies suggest that there may be a significant amount of dark-matter substructure in the form of clumps and tidal streams.

What is the best bet for the nature of dark matter?

From the vast array of proposals, the most promising ideas involve novel elementary particles. Among the candidates that have withstood long-standing theoretical scrutiny are weakly interacting massive particles (WIMPs) and axions. WIMPs, like neutrinos, interact only weakly with ordinary matter. They arise naturally in extensions to the standard model of particle physics (for example, in supersymmetry or in models with large extra dimensions). Detection of WIMPs is one of the primary goals of the Large Hadron Collider (LHC) at CERN near Geneva, Switzerland. The other candidate, the axion, is an elementary particle hypothesized to explain some of the symmetries of the strong interactions that bind quarks in protons and neutrons. There are other possibilities, so it is necessary to keep an open mind. However, constraints on the strength of the interaction of dark-matter particles with ordinary matter, their stability against decay and their 'coldness' — dark-matter particles today must move slowly compared with the speed of light — allow the range of possibilities to be pared down.

What experiments or observations can help?

Clearly, the most compelling resolution to the dark-matter problem would be the direct detection of dark-matter particles. Currently, there are some 20 experimental projects seeking to detect WIMPs by observing the 10–100 kiloelectronvolts of energy that would be deposited in a detector when a WIMP from the Galactic halo scatters from an atomic nucleus in the detector and makes it recoil. The target nuclei in some of these experiments are located in metallic crystals; the nuclear recoil is then detected through the recoiling energy collected in the detector. The challenge in these and other dark-matter detection experiments is to distinguish the signature of dark matter from the crowd of terrestrial-radiation backgrounds. But the current generation of experiments is becoming sufficiently sensitive that it will soon be possible to vet some of the leading particle-physics models for dark matter. The discovery of unknown particles at the LHC would greatly narrow the range of dark-matter candidates and boost our confidence that we are on the right track. But it would not eliminate the need for an in situ astrophysical detection.

Haven't there already been claims of dark-matter detection?

Yes. The DAMA experiment, operating deep underground at the Gran Sasso National Laboratory in Italy, has reported detection of the tell-tale annual modulation in dark-matter flux consistent with Earth's orbit through the Galactic dark-matter halo. This signal has not been corroborated by other experiments. Because other experiments use different target nuclei, the various results can only be compared in the context of specific theories of dark matter. The mass of the simplest, 'supersymmetric' WIMPs and their couplings to normal matter, proposed to explain the DAMA result, have been excluded by the other experiments.

How else can we see dark matter?

Although individual WIMPs are in theory stable, pairs of WIMPs can 'annihilate', producing high-energy photons and cosmic rays in the form of positrons (antielectrons), antiprotons and neutrinos. Detection of such particles might provide indirect evidence for dark matter. The most likely nearby sources of these annihilation products would be the Galactic Centre, where the dark-matter density is high, or the cores of some of the dark-matter-dominated dwarf galaxies surrounding the Milky Way (Fig. 1). One telling clue would be monoenergetic γ-rays. There is a host of ground-based, balloon- and satellite-borne experiments looking for these clues.

What about the cosmic-ray experiments ...?

In 2008, PAMELA (Payload for Antimatter Matter Exploration and Light-nuclei Astrophysics), a satellite-borne cosmic-ray experiment, and the balloon-borne ATIC (Advanced Thin Ionization Calorimeter) experiment, reported an excess flux of high-energy cosmic-ray positrons. These observations might be a consequence of WIMP annihilation, but the observed flux is higher, by several orders of magnitude, than the simplest WIMP models predict. One interpretation is that WIMP dark matter is more complicated than previously thought. However, more prosaic astrophysical explanations (such as particle acceleration by nearby pulsars) must be excluded before the anomaly can be attributed to dark matter.

... and future possibilities for studying dark matter?

Experiments to detect dark matter directly aim to exploit the dark-matter WIMP 'wind' (Fig. 1) and isolate the characteristic annual modulation in the WIMP flux from other background signals of terrestrial origin. Meanwhile, Gaia, a satellite mission set to launch in the near future, aims to chart the position and motion of about 109 nearby stars; this map will be used to trace out the gravitational field of the Milky Way, and thereby infer the dark-matter distribution in its dark-matter halo. A variety of experiments, including that using the recently launched Fermi Gamma-ray Space Telescope, will look for γ-rays from WIMP annihilation. And high-energy neutrino telescopes, such as IceCube at the South Pole, will look for neutrinos produced by the annihilation of WIMPs that have accumulated in the Sun and Earth.

What about dark energy?

The observation that the expansion of the Universe is speeding up (Fig. 2), instead of slowing down owing to the mutual gravitational attraction of matter, indicates that there is much more to the Universe than we understand at present. The leading interpretation is that the Universe is filled by something — dubbed dark energy — that 'antigravitates'. Whereas the possibility for gravitational repulsion does not exist in Newtonian gravity, it does exist in general relativity. The equivalence between matter and energy means that gaseous pressures caused by thermal molecular motions can be a source of gravitational fields. The gravitational field of a fluid with sufficiently negative pressure is repulsive. Although it may be difficult to imagine how molecular motions can give rise to a negative pressure, it has been realized that some of the quantum fields that arise in elementary-particle theory allow for fluids with negative pressure. Dark energy is thus simply the negative-pressure fluid that is postulated to account for cosmic acceleration.

Figure 2: Cosmic acceleration and dark energy.

Type Ia supernovae, which result from the explosion of white-dwarf stars, are thought to be standard candles (objects of known brightness). This property allows astronomers to determine how far away such supernovae are, based on their apparent brightness as observed on Earth — the dimmer the object seems to be, the higher the value of its magnitude and the farther away it is. The observation that these supernovae are dimmer than expected, at a given recessional velocity, has led to the conclusion that the Universe's expansion has been accelerating over approximately the past 5 billion years, before which the expansion was decelerating. The cause of this cosmic acceleration is widely attributed to dark energy.

What is the best bet for the nature of dark energy?

The simplest candidate for dark energy is Einstein's cosmological constant, which denotes a perfectly uniform fluid with negative pressure that is associated with the lowest energy (vacuum) state of the Universe. However, the observationally required value of the cosmological constant is 10120 times smaller than the theoretical expectation. Alternatively, dark energy might be due to a fluid of unknown particles, similar to the axion but much smaller in mass — quantum theory predicts that such particles could supply the requisite negative pressure to accelerate the cosmic expansion.

How reliable are the known laws of gravitation on cosmological scales?

General relativity works. It has been extremely well tested in the Solar System, and it is used to make sense of a vast catalogue of astrophysical and cosmological observations. These successes do not preclude the possibility of variations in the laws of gravitation on cosmological length scales. A Pandora's box of gravitational theories has been proposed to explain the accelerated cosmic expansion. But it is proving surprisingly difficult to tinker with gravity without running up against the precision constraints in the Solar System, and so far there are no compelling alternatives.

Could dark matter and dark energy be related?

It seems reasonable to consider the possibility of a 'dark sector', beyond the standard model of particle physics, containing a dark-matter particle and a dark-energy field. Both seem to require unknown sources of gravitational fields, one attractive and the other repulsive, but there have been no convincing proposals that unify the two phenomena.

Could cosmic acceleration be caused by any other phenomena?

One might consider new forms of gravitation (whereby normal matter produces the same antigravitational effect as dark energy), new electromagnetic effects (whereby distant supernovae are artificially dimmed; Fig. 2), or some other flaw in our fundamental assumptions (such as the statistical homogeneity and isotropy of the Universe on the largest length scales). The current state of observations does not favour one of these alternatives, but we must keep an open mind.

What recent observations have helped to refine the dark-energy problem?

The observations of 'baryon acoustic oscillations' have been used to corroborate and refine the evidence for cosmic acceleration. These cosmic ripples made by primordial sound waves are imprinted on the CMB and on the distribution of galaxies. By measuring how the wavelength of the ripples varies with the distance from Earth, one can chart the history of the cosmic expansion.

What experiments can help to determine the nature of dark energy?

There is a decided absence of compelling theoretical explanations for the physics underlying cosmic acceleration, so the approach to date has been to gather more of the same type of data in the hope that some clue will pop out. Apart from using supernovae and baryon acoustic oscillations, other methods that measure the rate at which normal and dark matter cluster under the influence of gravitation in an accelerating Universe are also progressing. One promising technique uses gravitational lensing in the 'weak' regime to set constraints on dark energy; in this regime, instead of the strong bending of light that results in highly distorted images in the form of elongated arcs, the images of distant sources are only weakly stretched and magnified by foreground matter (the 'lenses'). Another technique uses X-ray emissions of hot gas in galaxy clusters to determine the depth of their gravitational potential wells. But despite the promise of these methods, it may be difficult to determine the underlying physics of cosmic acceleration. On the other hand, this seems to be the only way to tackle such a challenging and fundamental problem.


Caldwell, R. & Kamionkowski, M. The physics of cosmic acceleration. Annu. Rev. Nucl. Part. Sci. (in the press); preprint available at (2009).

Frieman, J. A., Turner, M. S. & Huterer, D. Dark energy and the accelerating Universe. Annu. Rev. Astron. Astrophys. 46, 385–432 (2008).

Hooper, D. & Baltz, E. A. Strategies for determining the nature of dark matter. Annu. Rev. Nucl. Part. Sci. 58, 293–314 (2008).

Hogan, J. & Brumfiel, G. Unseen Universe. Nature 448, 240–248 (2007).

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Caldwell, R., Kamionkowski, M. Dark matter and dark energy. Nature 458, 587–589 (2009).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.