Dark Matter: The Particles We Cannot See

The Evidence for Dark Matter

The first strong evidence for dark matter came from galaxy rotation curves. In the 1970s, astronomer Vera Rubin and collaborator Kent Ford measured how fast stars orbit the centers of spiral galaxies at different distances from the galactic center. Newton's gravity predicts that stars far from the center, where most of the visible mass is concentrated, should orbit slowly, just as distant planets orbit the sun more slowly than inner planets. Instead, Rubin found that orbital speeds remain roughly constant out to the visible edges of galaxies and beyond. This "flat rotation curve" implies that galaxies are embedded in enormous halos of invisible mass extending far beyond the visible disk. The total mass inferred from the rotation curves exceeds the visible mass by roughly a factor of ten.

Galaxy clusters provide independent evidence. Fritz Zwicky first noticed in 1933 that galaxies in the Coma Cluster were moving far too fast to be held together by the gravitational pull of the visible matter alone. He calculated that the cluster must contain roughly 400 times more mass than what was visible, coining the term "dunkle Materie" (dark matter). Modern measurements using gravitational lensing, the bending of light from background galaxies by the cluster's gravity, confirm that galaxy clusters contain roughly five times more mass than their visible stars and gas. X-ray observations of the hot gas between galaxies in clusters show the same mass discrepancy: the gas is too hot and too diffuse to be gravitationally bound without additional invisible mass.

The Bullet Cluster, a pair of galaxy clusters that collided roughly 150 million years ago, provides particularly compelling evidence. During the collision, the hot gas in each cluster (which constitutes most of the visible mass) interacted and slowed down, lagging in the middle, while the galaxies and the dark matter (which does not interact with itself or with gas except through gravity) passed through each other and separated. Gravitational lensing maps show that most of the mass is located with the galaxies, not with the gas, exactly what dark matter predicts. The Bullet Cluster is difficult to explain without dark matter, because the gravitational mass and the visible mass are in different places.

The cosmic microwave background (CMB), the afterglow radiation from the Big Bang, provides the most precise measurement of dark matter's abundance. The CMB contains tiny temperature fluctuations, about one part in 100,000, that encode the density variations in the universe when it was only 380,000 years old. These fluctuations form a characteristic pattern of peaks and troughs when analyzed by angular scale, and the heights and positions of these peaks are exquisitely sensitive to the universe's composition. The Planck satellite's measurements of the CMB determine that ordinary matter (baryons) constitutes about 5% of the universe's total energy content, dark matter constitutes about 27%, and dark energy (which drives the accelerating expansion) constitutes about 68%. No modification of these proportions can match the observed CMB pattern without dark matter.

Large-scale structure formation provides a fourth line of evidence. Computer simulations of cosmic evolution starting from the CMB-era density fluctuations can reproduce the observed distribution of galaxies in the universe, including galaxy clusters, filaments, voids, and the cosmic web structure seen in galaxy surveys, only if dark matter is included. Without dark matter, the gravitational pull of ordinary matter alone would be too weak to form the structures we observe in the time available since the Big Bang. Dark matter, which decoupled from radiation earlier than ordinary matter and could begin clumping gravitationally sooner, acted as scaffolding around which ordinary matter collected to form galaxies.

What Dark Matter Is Not

Dark matter is not ordinary matter that simply does not glow. Astronomers have considered whether dark matter could consist of dim or dead stars, rogue planets, black holes, or other compact objects made of normal baryonic matter (collectively called MACHOs, Massive Astrophysical Compact Halo Objects). Microlensing surveys, which monitor millions of stars for brief brightness increases caused by the gravitational lensing of a foreground compact object, have ruled out MACHOs as the primary dark matter component across most of the plausible mass range. More importantly, Big Bang nucleosynthesis, the process that created the light elements in the first minutes after the Big Bang, places a firm upper limit on the total amount of ordinary matter in the universe at about 5% of the total energy content, far too little to account for the 27% measured for dark matter.

Dark matter is not neutrinos, despite neutrinos being weakly interacting and massive. Neutrinos travel at nearly the speed of light (they are "hot dark matter"), which means they would have streamed out of density enhancements in the early universe rather than clumping gravitationally. Simulations of structure formation with hot dark matter produce a universe where large structures form first and fragment into smaller ones (a "top-down" scenario), the opposite of what is observed. The actual universe formed "bottom-up," with small structures (galaxies) forming first and merging into larger ones (clusters), consistent with "cold dark matter," particles that move slowly compared to light and can gravitationally clump on small scales.

Modified gravity theories, such as Modified Newtonian Dynamics (MOND), attempt to explain dark matter phenomena by changing the law of gravity at very low accelerations rather than invoking invisible particles. MOND successfully fits many galaxy rotation curves with a single adjustable parameter, but it struggles with galaxy clusters (requiring some additional mass even after modification), cannot reproduce the CMB acoustic peaks without adding a dark matter-like component, and cannot explain the Bullet Cluster's separation of mass from gas. Most physicists consider dark matter particles to be a more complete and successful explanation, though MOND-inspired approaches continue to be studied.

Leading Dark Matter Candidates

WIMPs (Weakly Interacting Massive Particles) have been the leading dark matter candidate for decades. A WIMP would be a new particle with a mass in the range of roughly 10 GeV to 10 TeV that interacts with ordinary matter through the weak nuclear force (or a similarly weak new force). The appeal of WIMPs comes from the "WIMP miracle": a particle with weak-force-strength interactions and a mass near the electroweak scale would naturally freeze out of the hot early universe at the correct abundance to explain the observed dark matter density. This numerical coincidence, where the known physics of the weak force automatically produces the right amount of dark matter, motivated enormous experimental effort. Supersymmetry, the most popular theoretical extension of the Standard Model, naturally predicts a WIMP candidate: the lightest neutralino, a mixture of the superpartners of the neutral gauge bosons and Higgs bosons.

Axions are ultralight hypothetical particles originally proposed in 1977 to solve a separate problem in QCD: the strong CP problem (why the strong force does not violate CP symmetry despite having no reason not to). The Peccei-Quinn mechanism introduces a new symmetry that dynamically drives the strong CP-violating parameter to zero, and the axion is the quantum of the associated field. If axions exist with masses in the range of roughly 1 to 100 microelectronvolts (10^-6 to 10^-4 eV), they would be produced abundantly in the early universe and could constitute all of the dark matter. Despite their tiny mass, they would behave as cold dark matter because they are produced as a coherent field rather than as individual thermal particles. The ADMX experiment (Axion Dark Matter eXperiment) at the University of Washington searches for dark matter axions by looking for their conversion into photons in a strong magnetic field inside a microwave cavity. ADMX has reached the sensitivity needed to detect the theoretically predicted axion signal and is systematically scanning through the candidate mass range.

Sterile neutrinos are hypothetical neutrinos that do not interact through any Standard Model force, communicating with ordinary matter only through gravity and through mixing with the three known active neutrinos. With masses in the keV range (thousands of electronvolts), sterile neutrinos could serve as warm dark matter, intermediate between the hot dark matter of ordinary neutrinos and the cold dark matter of WIMPs. A sterile neutrino with a mass of about 7 keV could produce a detectable X-ray line at 3.5 keV when it decays. In 2014, two independent groups reported evidence for an unidentified 3.5 keV line in X-ray observations of galaxy clusters and the Andromeda galaxy, potentially consistent with decaying sterile neutrino dark matter. However, subsequent observations have produced conflicting results, and the interpretation remains controversial.

Primordial black holes formed in the very early universe, not from stellar collapse but from the gravitational collapse of extreme density fluctuations, have been proposed as dark matter. These could range in mass from far below the mass of a planet to thousands of solar masses. Observational constraints from microlensing, CMB spectral distortions, gravitational wave backgrounds, and dynamical effects have ruled out primordial black holes as the dominant dark matter component across most of the mass range, but windows remain open, particularly around 10^-12 to 10^-8 solar masses (roughly asteroid masses). The detection of gravitational waves from unexpectedly heavy black hole mergers by LIGO has renewed interest in the possibility that at least some black holes may be primordial in origin.

The Three Ways to Hunt Dark Matter

Direct detection experiments search for the rare instances when a dark matter particle scatters off an atomic nucleus in a carefully shielded underground detector. If the Milky Way's dark matter halo consists of WIMPs with masses around 100 GeV, roughly 100,000 of them should pass through every square centimeter of the Earth's surface every second. Occasionally, one should collide with a nucleus in a detector target, depositing a tiny amount of energy (a few keV) that can be measured as a nuclear recoil. The challenge is distinguishing this faint signal from the overwhelmingly larger background of cosmic rays, radioactive decay, and environmental radiation, which is why these experiments operate deep underground (to shield from cosmic rays) and use ultra-radiopure materials (to minimize radioactive backgrounds).

The current generation of direct detection experiments uses multi-tonne liquid xenon detectors. XENONnT at Gran Sasso in Italy uses 5.9 tonnes of liquid xenon. LUX-ZEPLIN (LZ) at the Sanford Underground Research Facility uses 7 tonnes. PandaX-4T at the China Jinping Underground Laboratory uses 4 tonnes. These detectors work as dual-phase time projection chambers: a particle interaction in the liquid xenon produces both scintillation light and ionization electrons. The light is detected by arrays of photomultiplier tubes at the top and bottom of the detector, while the electrons drift upward through the liquid and are extracted into a thin gas layer above, where they produce a second, proportional light signal. The ratio of the two signals distinguishes nuclear recoils (potential dark matter) from electron recoils (background). These experiments have reached astonishing sensitivities, probing WIMP-nucleon cross-sections below 10^-47 cm2, but have not yet detected a dark matter signal. The next generation, including DARWIN/XLZD (a proposed 40-tonne xenon detector), aims to push sensitivity to the "neutrino fog," the level where coherent neutrino-nucleus scattering from solar and atmospheric neutrinos becomes an irreducible background.

Indirect detection searches for the products of dark matter annihilation or decay. If dark matter particles are their own antiparticles (as many WIMP candidates are), two dark matter particles meeting in a dense region of the galaxy could annihilate into Standard Model particles: gamma rays, neutrinos, positrons, or antiprotons. Space-based experiments like the Fermi Gamma-ray Space Telescope scan the sky for excess gamma rays from the galactic center, dwarf galaxies, and galaxy clusters. The Alpha Magnetic Spectrometer (AMS-02) on the International Space Station measures cosmic ray positrons and antiprotons with high precision, looking for excesses above astrophysical backgrounds. Some intriguing anomalies have been observed, including an excess of gamma rays from the galactic center region detected by Fermi, but none has been conclusively attributed to dark matter rather than astrophysical sources.

Collider production aims to create dark matter particles in high-energy collisions at accelerators. If dark matter interacts with ordinary matter through any force, however weak, then sufficiently energetic collisions should occasionally produce dark matter particles. These particles would escape the detector without being measured, manifesting as "missing energy" or "missing transverse momentum" in the collision debris. The LHC searches for missing energy events in many different production channels, and the results constrain the masses and interaction strengths of various dark matter candidates. No dark matter signal has been found at the LHC, but the constraints complement and sometimes exceed those from direct and indirect detection experiments, particularly for low-mass dark matter candidates.

The Dark Matter Problem Today

The absence of a confirmed dark matter detection, despite decades of increasingly sensitive searches, has broadened the theoretical landscape. The WIMP paradigm, while not excluded, is under pressure: the simplest WIMP models predict cross-sections that have already been probed and rejected by direct detection experiments. The supersymmetric neutralino, once the most natural WIMP candidate, is constrained by both the failure to find supersymmetric particles at the LHC and the null results of direct detection experiments. This has led to increased interest in alternatives: axions, sterile neutrinos, hidden sector particles that interact with ordinary matter only through gravity and possibly through a new "dark photon" mediator, and other exotic possibilities.

The field is also grappling with tensions between dark matter predictions and observations on small scales. Cold dark matter simulations predict that galaxies should have cuspy density profiles (density rising steeply toward the center) and hundreds of satellite galaxies around each Milky Way-sized galaxy. Observations suggest that many galaxies have cored density profiles (constant density in the center) and fewer satellites than predicted. These discrepancies may be resolved by better modeling of ordinary matter effects (supernova feedback, gas dynamics, stellar winds) on dark matter distributions, or they may indicate that dark matter has self-interactions or other properties not captured by the simplest cold dark matter models.

Whatever dark matter turns out to be, it represents a particle (or particles) that makes up the majority of the matter in the universe and is not described by the Standard Model. Its identification would be one of the most important discoveries in the history of science, simultaneously solving an astrophysical mystery and revealing new fundamental physics. The range of experimental approaches being pursued, from underground xenon detectors to axion haloscopes to space-based gamma-ray telescopes to the LHC, ensures that the search covers a broad range of possibilities, maximizing the chances that the nature of dark matter will be revealed.