Galaxies, Rotation Curves & Dark Matter — The Hidden Mass of the Universe

Galaxy Structure: What We Can See

A typical spiral galaxy like the Milky Way is a gravitationally bound system containing 10&sup9;–10¹² stars, interstellar gas and dust, and (we now believe) a dominant dark-matter component. Morphologically the visible matter is organised into:

• Bulge: a dense central spheroid of old, red stars. The Milky Way’s bulge has mass ∼2×10¹⁰ M_☉ and radius ∼3 kpc. Most large spirals host a supermassive black hole at the centre (Milky Way: Sgr A*, 4.15×10⁶ M_☉).
• Stellar disk: a flat, rotating disk of stars following an exponential surface-density profile Σ(R) = Σ₀ e^−R/R_d with scale radius R_d ∼3 kpc for the Milky Way.
• Gas disk: a thicker layer of neutral hydrogen (HI) and molecular clouds, extending beyond the stellar disk. Radio 21-cm emission maps the HI distribution and, crucially, allows rotation-curve measurements far beyond the visible optical disk.
• Stellar halo: a sparse spherical distribution of old globular clusters and halo stars, extending to ∼100 kpc.

The total visible mass of a galaxy like the Milky Way is ∼5×10¹⁰ M_☉ in stars plus ∼10¹⁰ M_☉ in gas. But the dynamical mass inferred from kinematics is ∼10¹² M_☉ — roughly 20 times larger. The discrepancy is the dark matter problem.

Rotation Curves — The Smoking Gun

For a test particle in circular orbit at radius R from a mass distribution M(<R), Newtonian gravity requires:

In our own Solar System, planets follow this Keplerian v ∝ R^−1/2 relation perfectly: Mercury orbits at 47.9 km/s, Neptune at 5.4 km/s. If the mass in a spiral galaxy were concentrated at the centre like the Solar System, stars in the outer disk should show the same declining curve.

Vera Rubin and Kent Ford began systematic measurements of galaxy rotation in the late 1960s using the Doppler shift of Hα emission lines. By 1980, their paper on 21 spirals revealed an unmistakable pattern: the rotation curves are flat at large radii, with v(R) ≈ constant out to the farthest point measurable. Subsequent 21-cm HI surveys (extending to several times the optical radius) confirmed flat or rising curves in hundreds of galaxies.

A flat curve v(R) = v_flat implies M(<R) ∝ R — mass continues to grow linearly with radius even where there are essentially no visible stars. The invisible mass required typically exceeds the visible mass by a factor of 5–30.

The NFW Dark Matter Halo Profile

N-body simulations of structure formation in a cold dark matter (CDM) universe consistently produce dark matter halos with a characteristic density profile. Julio Navarro, Carlos Frenk, and Simon White (1995, 1996) found that halos spanning 10 orders of magnitude in mass all follow the same functional form — the NFW profile:

The NFW profile has two contrasting behaviours: at small radii (r « r_s) it scales as ρ ∝ r²/r−1 = r⁻¹ (cuspy inner core); at large radii (r » r_s) it scales as ρ ∝ r⁻³. Integrating gives the enclosed mass:

The NFW profile does not produce an exactly flat curve independently, but combined with the stellar and gas disks the total v(r) is remarkably flat over the observable range — matching observations for well-studied spirals.

The cusp-core controversy

The inner r⁻¹ NFW cusp is predicted by CDM simulations but observations of dwarf galaxies often suggest a flat central density core (ρ = const). This cusp-core problem remains an active research area. Proposed solutions include: baryonic feedback (supernova explosions blow out gas, reducing the central mass concentration and “heating” the dark matter), warm dark matter (WDM) with a free-streaming cutoff that smooths small-scale structure, or self-interacting dark matter (SIDM) which creates isothermal cores through collisions.

Freeman Disk and Bulge Components

The visible stellar disk contributes its own gravitational potential. For a thin exponential disk (Freeman 1970), the circular-speed contribution at radius R in the plane of the disk is computed from elliptic integrals, but an excellent analytic approximation (van der Kruit & Searle) uses modified Bessel functions:

The bulge approximated as a Hernquist sphere (density ρ ∝ 1/[r(r+a)³]) or a de Vaucouleurs R^1/4 spheroid adds a steep rising contribution to v(r) near the centre. A good decomposition of a spiral rotation curve therefore separates: bulge (inner rise), disk (intermediate hump), and DM halo (outer flat part). This mass-to-light ratio fitting is the standard method for inferring dark matter fractions in individual galaxies.

Evidence for Dark Matter Beyond Rotation Curves

Gravitational lensing

General relativity predicts that mass bends light. Distant galaxy clusters act as gravitational lenses, distorting and magnifying background galaxies. The lensing strength probes the total mass, regardless of whether that mass emits light. Comparing lensing mass to X-ray gas mass (the dominant visible baryonic component in clusters) consistently shows a factor ∼5–7 more mass in dark matter. The bending angle for a light ray passing at impact parameter b near a mass M is:

The Bullet Cluster

The most direct evidence for dark matter as a separate component comes from the Bullet Cluster (1E 0657-558), two galaxy clusters that collided ∼100 Myr ago. Chandra X-ray observations show hot gas (the main baryonic component) slowed and shocked at the collision point. But weak gravitational lensing maps — which trace total mass — show the mass displaced ahead of the gas, coinciding with the galaxies and stars that sailed through without interaction. The dark matter, like the stars, passed straight through because dark matter interacts only gravitationally (and weakly, if at all, through other forces). This spatial separation of dark and baryonic mass is essentially impossible to explain with modified gravity.

Cosmic microwave background

The CMB power spectrum (acoustic peaks in temperature fluctuations at z ≈ 1100) depends sensitively on the density of baryons Ω_b vs. total matter Ω_m. Planck 2018 results give:

A universe without dark matter cannot reproduce the observed second and third acoustic peaks. The CMB data independently requires a non-baryonic matter component roughly 5 times more abundant than ordinary atoms.

Big Bang nucleosynthesis

BBN constrains the baryon density from the observed primordial abundances of D, ³He, ⁴He, and ⁷Li. It yields Ω_bh² ≈ 0.022 — in excellent agreement with CMB measurements but far below the total matter density required by structure formation and rotation curves. This cross-check from an independent epoch (t ≈ 3 min) firmly establishes that the dark matter cannot be baryonic.

What Is Dark Matter?

Despite overwhelming evidence for dark matter’s gravitational effects, its particle nature remains unknown. The main candidate classes are:

• WIMPs (Weakly Interacting Massive Particles, m ∼ 10–1000 GeV/c²): first-order favourite for decades because supersymmetric extensions of the Standard Model naturally predict a stable lightest supersymmetric particle (LSP) with approximately the right relic density (the “WIMP miracle”). Direct detection experiments (LUX, PandaX, XENONnT) have ruled out large regions of parameter space without a positive signal.
• Axions (m ∼ 1–1000 μeV/c²): ultra-light scalar particles originally proposed to solve the strong-CP problem in QCD. Axion dark matter forms Bose-Einstein condensates (fuzzy dark matter) that may resolve the cusp-core problem. Experiments: ADMX, CASPEr.
• Sterile neutrinos: hypothetical heavy neutrinos (keV–GeV) that mix weakly with active neutrinos. Would decay to X-rays: a tentative 3.5 keV X-ray line from galaxy clusters attracted much attention (2014) but remains unconfirmed.
• PBHs (Primordial Black Holes): black holes formed from density fluctuations in the very early universe. Microlensing surveys (MACHO, EROS, HSC) constrain PBHs to a small asteroid-mass window. The LIGO/Virgo gravitational-wave detections renewed interest in PBH dark matter, but current data do not support significant contributions.

Modified Gravity: MOND

Mordehai Milgrom proposed in 1983 that dark matter is unnecessary if Newton’s second law is modified at very low accelerations. His Modified Newtonian Dynamics (MOND):

MOND naturally explains the baryonic Tully-Fisher relation (a tight empirical correlation v_flat⁴ ∝ M_bary) that holds across five decades in galaxy mass. This is a striking success: in the dark matter paradigm the correlation must be explained by the fine-tuned relationship between baryonic mass and halo mass. MOND, however, struggles badly with galaxy clusters (where it still requires dark matter — or hot sterile neutrinos) and with the Bullet Cluster. A fully relativistic extension (TeVeS, Bekenstein 2004, and its successors) is needed for lensing, but these theories face severe theoretical challenges and tension with gravitational-wave observations.

Cosmological Structure: ΛCDM

The ΛCDM (Λ Cold Dark Matter) model combines dark matter with the cosmological constant Λ (dark energy) and describes the formation of large-scale structure from small initial quantum fluctuations amplified by inflation. Key parameters (Planck 2018):

In ΛCDM, dark matter is cold (non-relativistic at decoupling) and collisionless. It provides the gravitational scaffolding into which baryons fall to form galaxies. The cosmic web — filaments, voids, and cluster nodes — is the fossil record of dark matter structure formation. Simulations (Millennium, IllustrisTNG, EAGLE) reproduce the observed galaxy distribution, cluster mass function, and two-point correlation function with remarkable fidelity.

The two tensions in ΛCDM

Despite its success, ΛCDM faces two significant tensions in current data. The H₀ tension: local distance-ladder measurements (Cepheids + Type Ia SNe) give H₀ ≈ 73 km/s/Mpc, inconsistent with the CMB-inferred value of 67 km/s/Mpc at 4–5σ. The σ₈ tension: weak lensing surveys (KiDS, DES) prefer slightly less clustering amplitude than the CMB predicts. Whether these tensions indicate new physics beyond ΛCDM or systematic errors remains actively debated.

Binary Stars and Galactic Dynamics

Inside galaxies, stellar dynamics also reveals the mass distribution through methods other than rotation curves. The virial theorem (2⟨T⟩ = −⟨V⟩ for a gravitationally bound system in equilibrium) relates the velocity dispersion of an elliptical galaxy or star cluster to its total mass: M ≈ 5σ²R/G. Binary-star orbits measure stellar masses directly through Kepler’s third law, and by cross-correlating spectra from both components orbital parameters can be determined precisely.

At the largest scales, peculiar velocities of galaxies in clusters, X-ray temperature profiles of intracluster gas, and the Sunyaev-Zeld’ovich effect (inverse Compton scattering of CMB photons off hot cluster electrons) all independently measure the total mass — and all yield the same dark-matter-dominated answer.

Simulations on This Platform

Explore galactic and dark-matter physics with these interactive tools:

• Galaxy Rotation Curve — face-on spiral with NFW halo, Freeman disk, bulge; v(r) decomposition
• Gravitational Lensing — strong lensing (Einstein rings, arcs), microlensing light curves, Fermat potential
• Binary Stars — two-body Keplerian orbits, mass ratio, eccentricity, radial velocity curves
• Dark Matter — interactive halo models, density profiles, rotation curve decomposition
• Big Bang Nucleosynthesis — primordial element abundances as a function of baryon density
• Cosmic Microwave Background — CMB power spectrum, acoustic peaks, cosmological parameters