Across decades, the hunt for a dark matter particle has looked at many possible solutions—but so far, humanity hasn’t produced a clear answer. Is dark matter a neutrino? An axion? A figment of our imagination? Scientists don’t agree, though experiments from XENON to ADMX continue to strive towards giving us an answer.

“We have to be extremely open-minded about what it might be,” James Bullock, a professor of physics and astronomy at UC Irvine, told Ars. “Dark matter could be even more interesting than we were thinking it was going to be 20 or 30 years ago.”

The built-up confusion surrounding dark matter today can be extremely hard to parse. Recent headlines declared dark matter may not even exist, and even dedicated followers could be forgiven for asking how scientists came up with the idea in the first place. So to better understand dark matter’s place in the Universe, it may be helpful to take a look back at how our ideas about this mysterious material started and evolved over time—it's time to traverse a condensed history of dark matter.

The first mention

As laid out in a recent review by Gianfranco Bertone and Dan Hooper, the earliest references to dark matter only hint at the modern understanding. Toward the end of the 19th century, new images from the budding field of astronomical photography revealed dark regions in the sky. Stars did not appear to be evenly distributed, and scientists wondered if this was because dark regions lacked stars altogether, or if some absorbing matter was blocking their view of other stars.

Lord Kelvin, a Scots-Irish physicist, was one of the first scientists who attempted to estimate the number of dark bodies in the Milky Way galaxy. He used estimates drawn from the observed velocity dispersion of the stars—how fast these stars were orbiting around the core of the galaxy. Information about the speed of these stars allowed him to estimate the mass of the galaxy. There was a difference between that mass and the stars we can see. In one of his Baltimore Lectures on molecular dynamics and the wave theory of light, he concluded that “many of our stars, perhaps a great majority of them, may be dark bodies.”

Henri Poincaré, a French mathematician and physicist, responded to Lord Kelvin’s ideas in his 1906 work “The Milky Way and Theory of Gases”—explicitly using the term “dark matter,” or “matière obscure” in the writing's original French. Though impressed by Lord Kelvin’s ideas, he disagreed with the man’s conclusions. Poincaré wrote that “since his number is comparable to that which the telescope gives, then there is no dark matter, or at least not so much as there is of shining matter.”

Poincaré ended his paper on an uncertain note. Physicists, armed with a mistaken understanding of what powered stars, estimated that they can only exist for “fifty millions of years” before dying out. Given the relative brevity of a sun’s lifespans on the cosmological timescale, wouldn’t we expect to see a greater amount of dark matter in the form of dead stars? He left the question unanswered.

It’s easy to see that in this early period, dark matter literally meant dark matter: regions in the sky that lacked light and that scientists speculated may represent dark bodies. But hold onto your hats (and 19th century monocles)—the next few decades would bring about significant changes in our understanding of dark matter.

Data leads to new questions

The first major evidence that dark matter may actually be much more common than previously thought came from the work of Swiss-American astronomer Fritz Zwicky. After studying the Coma galaxy cluster, he determined that it did not contain enough visible matter to hold itself together. While the 800 galaxies he studied should have a velocity dispersion of 80 kilometers per second, he found that the real value was closer to 1,000 kilometers per second. This meant stars were traveling so fast that they should escape their mutual gravitational pull.

The fact that they hadn’t implied the galaxies had more mass than could be accounted for with visible matter alone. “If this would be confirmed,” Zwicky wrote, in his 1933 paper. “We would get the surprising result that dark matter is present in much greater amount than luminous matter.” The additional mass from this theorized dark matter would help explain how the galaxy cluster is able to hold together via gravitational attraction.

Later calculations showed that Zwicky’s estimate of the mass-to-light ratio was too large, by a factor of about 8, which meant that his estimates for the amount of dark matter were too high. Even still, his work paved the way for our understanding that most of the mass of a galaxy cluster is actually not in the form of atoms.

Like others that came before him, Zwicky still felt that dark matter was composed of material such as cold stars, other solid bodies, and gases. As of yet, the scientific community had no compelling evidence that this missing mass could be anything else.

Galactic rotation curves offer new clues

As cosmology matured as a science in the 1960s and 1970s, American astronomer Vera Rubin discovered something unusual about the rotation curves of galaxies. Rubin used an image tube spectrograph developed by her collaborator, astronomer Kent Ford, in order to observe spiral galaxies. Since spiral galaxies have most of their stars clustered near the core, scientists assumed that the majority of the mass and gravity would be concentrated near the center, too.

By looking at the rotation curves of stars in these galaxies—graphs of the velocity of the stars versus their distance from the center of the galaxy—astronomers could determine the distribution of mass. Typically, stars farther from the center should be moving more slowly than stars near the center—just as Neptune moves more slowly around the sun than Mercury. In most rotation curves, then, the line of the graph starts out high and then falls as it moves right, away from the galaxy’s core. This is consistent with most of the mass existing in the center of the system.

But Rubin found something different in the spiral galaxies she studied: instead of sloping down, their rotation curves seemed to level off. This implied that the stars in the far outer regions of these spiral galaxies were moving just as quickly as stars near the center. The observed, visible mass of a galaxy did not have enough gravity to hold these fast-moving stars together. As a result, Rubin concluded that these galaxies contain ten times as much dark matter as visible matter, which helped bind the galaxies together. Over the course of the 1970s, other scientists, despite some initial skepticism, confirmed these findings. A large “halo” of dark matter surrounds each galaxy.

With this, researchers seemed to finally have incontrovertible evidence that the Universe was not just composed of visible matter. The discovery prompted a gradual change in how “dark matter” was conceived. No longer just cold stars and solid bodies, dark matter came to be viewed as the material that made up the majority of the Universe. And with that change came a new player on the scene: particle physics.

Particle physics enters

MAssive Compact Halo Objects (such as brown dwarf stars and black holes) were once a contender to explain dark matter. But they gradually fell out of favor after a few crucial observations: data from the EROS project in the late 1990s suggested that MACHOs weren’t numerous enough to account for all the mass of needed. In addition, other observations showed that the gravitational effects of dark matter occurred in places were MACHOs—which reside at the edges of galaxies—did not exist.

With regular objects off the table, an unusual pairing grew: particle physicists and astrophysicists. Prior to dark matter, the two groups had had little to do with one another. Indeed, when Fermilab created its theoretical astrophysics group in 1983, many scientists wondered what particle physics research problems could be aided by astrophysics. But by the latter half of the 1980s, the idea that dark matter was made up of a not-yet-discovered subatomic particle was gaining steam, leading to more and more collaborations between particle physicists and astrophysicists.

These collaborations produced more than a few ideas: perhaps dark matter was composed of the electrically neutral, weakly interacting neutrino, or perhaps of the hypothesized, extremely light axion. Perhaps it was made up of some light supersymmetric particle that scientists had not yet discovered. In short, scientists began to search for WIMPs—Weakly Interacting Massive Particles—instead of MACHOs.

Whether WIMPs or MACHOs, though, some scientists remained skeptical of the quest for dark matter, suggesting perhaps it was our theory of gravity that was wrong. An altered theory of gravity could be able to explain Rubin’s unusual data without the need for new particles or matter. One of the leading alternatives to dark matter has been MOND, or MOdified Newtonian Dynamics. However, scientists have yet to discover a version of MOND that can elegantly and simply explain the observed gravitational effects.

The COBE satellite was just a few instruments and a large cooling system.

Credit: Lawrence Berkeley Lab

Precision cosmology and the 1992 COBE Experiment

By the end of the 1980s, more and more scientists accepted the idea that most of the mass in the Universe consisted of cold (slow-moving) dark matter—and experiments in the next few decades only gave more support to this view.

A big one came when John Mather and George Smoot kicked off a new era of cosmology in the 1990s with their work on the COsmic Background Explorer, which looked at the cosmic microwave background (CMB). The CMB is essentially radiation leftover after the Big Bang, providing evidence about the state of the early Universe. Previous data had shown that the CMB was even, reflecting the large-scale evenness of the Universe. But on slightly smaller scales, the Universe was lumpy, with both large voids and large clumps of galaxies. Inflation was one possible explanation: if the early Universe had small quantum fluctuations, they would have greatly enlarged when the Universe expanded exponentially in the moments after the Big Bang, leading to the lumpiness observed today.

To find evidence of this, Mather, Smoot, and their team sought to find tiny fluctuations in the CMB. In 1992, the COBE team announced that it found minuscule temperature fluctuations—only one part in 100,000—that were the remains of those fluctuations from the early Universe, pre-inflation. A new era of precision cosmology was born.

Tracy Slatyer, an Assistant Professor at MIT in the physics department, has extensively studied astrophysical data from the CMB. She told Ars that the rise of precision cosmology in the 1990s helped cement ideas that dark matter, rather than MOND, was responsible for some of the unusual data scientists had discovered. That’s because, for theories of inflation to fit new data about the CMB, the Universe had to have much more matter than visible matter alone could provide. Without it, the massive structures seen in today’s expanding Universe would not have had time to develop.

But a new study would soon provide even more enticing evidence for dark matter—and how it interacts with visible matter.

The Bullet Cluster

In 2006, scientists at Harvard University saw something spectacular through the Chandra Telescope: two galaxy clusters colliding. The collision left clumps of visible and invisible matter to be analyzed.

Here’s how it worked (this animation might help): after the collision, galaxies were left on the far edges, represented as a blue haze. Slowed by the collision and left closer to the center were gas clouds emitting X-rays, where most of the visible matter ended up. Where was dark matter after the collision? Scientists hypothesized that it would not interact with the gas clouds, and it should instead pass through the collision, winding up near the blue haze of galaxy clusters on the far edges.

That’s precisely what researchers saw. Using gravitational lensing, scientists were able to determine that most of the matter was near the galaxy clusters and not in the red haze of gas clouds where most visible matter was. The galaxy collision therefore separated dark and visible matter—and by doing so, allowed scientists to find evidence of the former.

For many scientists, the results cleared up any lingering doubts about the existence of dark matter. The hunt for dark matter was on—and today, scientists are conducting numerous innovative experiments to find some clues as to the makeup of this mysterious material.

Lux, a xenon-based dark matter detector.

Credit: Lawrence Berkeley Lab

Modern searches: Detection by scattering

One type of experiment aims to detect dark matter particles bumping into familiar ones. The two best-known initiatives of this type are called LUX and XENON. In both experiments, researchers fill large detectors with liquid xenon. They then search for small signals result when rarely interacting WIMPs collide with the nuclei of xenon atoms, giving energy to the atoms and producing photons that the equipment can detect.

So far, scientists have not been able to find the photons they are looking for, though these experiments have helped put some limits on the potential properties of dark matter.

Elena Aprile, a professor of physics at Columbia University and the woman who leads the XENON dark matter experiment, told Ars that it is “getting harder and harder experimentally” to find dark matter. Even though the XENON experiment has seen a few iterations—and the team has another, XENONnT, planned in 2019—there’s still no sought-after WIMP particle.

“Clearly most of us have this feeling that the story is coming to a conclusion, and at the same time we are pushing to the limits of what we know how to do,” Aprile told Ars. “People have attachment to the WIMP. 'Do I really like this guy? Is it time to move away from it?' It is very, very frustrating.”

Still, Aprile is determined to succeed. She wants her team to be the first to obtain direct evidence of the possible dark matter particle. “I don’t give up easily,” she said.

Annihilation and decay

Other searches have focused on the possibility that dark matter particles collide and annihilate in much the same way as visible matter. If this is true, then we might be able to see evidence in denser regions of dark matter, where collisions might produce an excess of energetic particles like positrons, the antimatter partners of electrons.

Scientists did find this excess in data from the Russian-European PAMELA satellite in 2008, and again in data from the Alpha Magnetic Spectrometer on the International Space Station in 2013. However, they weren’t able to pinpoint whether the excess was indeed the result of dark matter particle collisions—or less exotic sources such as pulsars (a type of neutron star). They also found that the energy signature of the particles observed in the 2013 data did not fit with their expectations of how dark matter would behave. Though this doesn’t rule out the possibility that these excesses are indirect evidence of dark matter, it does make it less likely.

The Axion experiments

Another type of experiment is searching for axions—theoretical particles that were first proposed to solve a different problem in physics, one involving the strong nuclear force. Axions are electrically neutral, interact weakly with light and other types of matter, and just happen to have properties that make them a strong dark matter candidate.

More good news is that axions are only consistent with dark matter if their mass is within a very narrow range, which makes the idea easy to test. A couple of different experiments are underway that explore the possibility of axions as dark matter.

First, there’s the CERN Axion Solar Telescope. Axions can be converted into light (photons), and light can be converted into axions. So, as light scatters off particles in the Sun, it might be transformed into axions, which then escape the Sun. The CERN Axion Solar Telescope aims to find such axions, using a dipole magnet from the Large Hadron Collider. Currently the telescope lacks the sensitivity needed to rule out axions as dark matter, although it could help constrain their properties.

There’s also the Axion Dark Matter eXperiment (ADMX) at the University of Washington. If they exist, axions in the dark matter halo of the Milky Way should be passing through Earth at all times. ADMX aims to “catch” such particles by stimulating them to decay into photons that can then be detected by a device called a Radio Frequency (RF) cavity. One major difficulty researchers face, though, is reducing the background noise enough to find the weak photon signal.

Looking beyond the WIMP

With so many frustrations in the search for dark matter, some are looking to alternative theories once again. James Bullock told Ars that the MOND model is a strong contender. “It’s striking how well it works in explaining the rotation speeds of certain types of galaxies,” he said. “That’s something that I personally pay a lot of attention to. It’s something you can’t sweep under the rug anymore.” Still, he noted that MOND theories need to get better at explaining large-scale observations such as the cosmic microwave background in order for them to be accepted.