What Is Dark Matter?

The first strong evidence came from the rotation of galaxies. Stars at the outer edge of a spiral galaxy orbit far faster than the visible matter can account for, and at those speeds they should have been flung away long ago.

Measurements by Vera Rubin and colleagues in the 1970s showed this pattern across many galaxies. Either there is a great deal of unseen mass holding them together, or our understanding of gravity at these scales is incomplete.

Galaxy rotation is not the only line of argument, which is what makes the case strong:

• •Gravitational lensing, where light bends around more mass than is visible
• •The cosmic microwave background, whose pattern requires non luminous matter
• •Galaxy cluster dynamics, moving too fast to stay bound without extra mass
• •Structure formation simulations, which only match observations with dark matter included

One observation is often described as the strongest single piece of evidence. The bullet cluster is a collision between two galaxy clusters, and it separates the components in a revealing way.

The hot gas, which is most of the ordinary matter, slowed down and piled up in the middle. But gravitational lensing shows the bulk of the mass carried straight through and sits away from the gas. That separation is difficult to explain by modifying gravity, since gravity should track wherever the visible matter is.

The leading candidates are particles that interact through gravity and possibly the weak nuclear force, but not electromagnetically, which is why they neither emit nor absorb light.

Weakly interacting massive particles were the favourite for decades, but increasingly sensitive detectors deep underground have repeatedly found nothing, narrowing the possibilities considerably. Attention has shifted towards alternatives including axions and various lighter particles, and searches are ongoing.

Dark matter is inferred entirely from its gravitational effects. No experiment has detected a dark matter particle, despite decades of increasingly sensitive searches.

There is a serious minority position, modified Newtonian dynamics and its relatives, arguing that gravity behaves differently at very low accelerations. These models fit galaxy rotation curves well but struggle with clusters and with the cosmic microwave background. Most cosmologists favour dark matter while acknowledging that not finding it is a genuine problem.

The two are routinely confused and are not related beyond both being poorly understood. Dark matter has gravity and holds structures together. Dark energy appears to push space apart and drives the accelerating expansion of the universe.

The proportions are roughly five percent ordinary matter, twenty seven percent dark matter and sixty eight percent dark energy. Put plainly, everything ever observed directly, including every star, planet and galaxy, accounts for about a twentieth of the contents of the universe, and the rest is named after our ignorance rather than described by our knowledge.

Physicists can say something about dark matter even without detecting it, because its speed in the early universe leaves fingerprints. If the particles moved close to the speed of light, they would have streamed out of small clumps of matter and smoothed them away. Galaxies would then have formed late, by giant structures breaking into smaller pieces. This is called hot dark matter, and neutrinos, which really do have a tiny mass, were once a natural candidate.

Computer simulations and surveys of how galaxies are actually arranged ruled that picture out. Real structure appears to have built up from small clumps merging into larger ones, which requires slow moving, or cold, dark matter. Neutrinos exist and contribute a little, but they are far too light and fast to be the main ingredient. This is a good example of indirect reasoning: the pattern of galaxies across the sky narrows down the properties of a substance nobody has ever seen.

The galaxy rotation measurements of the 1970s were not the first hint. In 1933 the Swiss astronomer Fritz Zwicky studied the Coma Cluster, a large group of galaxies, and estimated how fast its members were moving. They were moving far too fast for the visible mass to hold the cluster together, and he suggested that a large amount of unseen matter must be present. He used a German phrase meaning dark matter. For decades the idea attracted little attention, partly because the measurements were uncertain, until the galaxy rotation results made the problem impossible to ignore.

Experiments that search for dark matter particles use three broad strategies. Direct detection uses large tanks of very cold, pure material, such as liquid xenon, and waits for a particle to bump into an atomic nucleus and produce a tiny flash or electrical signal. These detectors sit deep underground, under rock that blocks most cosmic rays, because ordinary radiation would otherwise swamp the rare signal. A second approach tries to create dark matter in particle colliders and spot energy that goes missing. A third uses telescopes to look for radiation produced if dark matter particles meet and destroy one another.

An obvious first suggestion is that the missing mass is ordinary matter that is hard to see: dim stars, stray planets, cold gas or black holes. Astronomers nicknamed compact objects of this kind MACHOs, short for massive compact halo objects, and tested the idea in the 1990s. If a galaxy's halo were full of them, one would occasionally pass in front of a distant star, and its gravity would act as a lens, briefly brightening the star in a recognisable way. Surveys monitored millions of stars in the Large Magellanic Cloud for years. They found some events, but far too few for MACHOs to explain most of the halo.

A separate argument comes from the first few minutes after the Big Bang. The amounts of deuterium and helium formed then depend on how dense ordinary matter was, because protons and neutrons had to meet in order to fuse. Measuring those amounts in very old gas gives a density of ordinary matter that matches the roughly five percent figure found independently from the cosmic microwave background. There is simply not enough ordinary matter to supply the gravity astronomers observe. Whatever dark matter is, it is almost certainly not made of the protons and neutrons that build stars, planets and people.

A rough calculation shows how the rotation evidence works. For an object in a circular orbit, its speed and distance reveal how much mass lies inside the orbit: using Newton's law of gravity, that mass equals the speed squared times the radius, divided by the gravitational constant. The Sun travels around the centre of the Milky Way at roughly 230 kilometres per second, about 26,000 light years out. Plugging in those values gives around 100 billion times the mass of the Sun inside our orbit. That figure alone is not the puzzle, because the inner galaxy holds a great many stars.

The puzzle appears further out. If almost all the mass sat inside the Sun's orbit, speeds should fall with distance, just as Neptune moves round the Sun far more slowly than Earth: a star four times further out would travel at only half the speed. Instead, radio measurements of hydrogen gas well beyond the visible disc show speeds staying roughly constant. Because the enclosed mass equals speed squared times radius divided by the constant, a steady speed means the mass keeps growing in step with distance, even where there are hardly any stars. That growing, unseen mass is what astronomers call the dark matter halo.

Dark matter is unseen mass inferred from galaxy rotation, gravitational lensing and the structure of the universe, it makes up about a quarter of everything, and despite decades of searching nobody has detected a particle of it.