Classical Mechanics
Virial Theorem
For a bound system: ⟨T⟩ = −(1/2)⟨V⟩ when V ∝ r^n with n = −1 (gravity, Coulomb)
The virial theorem (Clausius 1870) relates the time-averaged kinetic energy ⟨T⟩ and potential energy ⟨V⟩ of a bound system. For a power-law potential V ∝ r^n: 2⟨T⟩ = n⟨V⟩. For gravity or Coulomb (n = −1): 2⟨T⟩ = −⟨V⟩, so ⟨T⟩ = −⟨V⟩/2 and total energy E = ⟨T⟩ + ⟨V⟩ = ⟨V⟩/2 = −⟨T⟩. Astrophysics applications: galaxy mass estimation (the Virial Theorem applied to a virialized cluster gives total mass = 5σ²R/3G, where σ is velocity dispersion — the original detection of dark matter by Zwicky 1933 in Coma Cluster used this; missing mass factor ~5×). Also: stellar interiors (≈ same kinetic = potential for hydrostatic equilibrium), self-gravitating systems, plasmas.
- AuthorClausius 1870
- Power law V ∝ r^n2⟨T⟩ = n⟨V⟩
- Gravity n=−1⟨T⟩ = −⟨V⟩/2
- Total energyE = −⟨T⟩ (gravity)
- Zwicky 1933Dark matter detected
- Galaxy massM ≈ 5σ²R/3G
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why virial matters
- Galaxy mass estimation. Velocity dispersions σ from spectroscopy plus radii R from imaging give total mass M ≈ 5σ²R/(3G). Used routinely on clusters, dwarf galaxies, and globular clusters.
- Dark matter. Zwicky's 1933 Coma Cluster paper was the first quantitative detection of dark matter — visible mass was about an order of magnitude too small to keep the cluster bound at the measured velocity dispersion.
- Stellar structure. A star in hydrostatic equilibrium satisfies a virial relation: gravitational potential energy is twice the thermal energy. Explains why contracting protostars heat up and why white dwarfs and neutron stars settle at characteristic radii.
- Real gas equations. Clausius's original 1870 motivation; the virial expansion B(T), C(T), ... corrects the ideal gas law for intermolecular interactions.
- Atomic physics. Hydrogen and other Coulomb-bound systems obey ⟨T⟩ = −⟨V⟩/2 to numerical precision in the ground state — handy as a wavefunction sanity check.
- Plasmas. Self-gravitating hot gas in galaxy clusters has thermal pressure balancing gravity; the virial temperature kT_v ≈ G M m_p / R sets the X-ray emitting temperature.
- Cosmology. Virialization is the signature that a halo has settled into a quasi-equilibrium configuration — N-body simulations declare a halo virialized when 2⟨T⟩/⟨|V|⟩ ≈ 1.
Common misconceptions
- "Instantaneous equality." The virial theorem is a time-average statement. At pericenter ⟨T⟩ > |⟨V⟩|/2 momentarily; at apocenter the reverse. Only the long-time average is fixed.
- "Always a 1:2 ratio." Only for n = −1. For a harmonic oscillator (n = +2) the relation is ⟨T⟩ = ⟨V⟩. For a hard-wall infinite well, surface terms dominate and the simple form does not apply.
- "Pure theory, not measurable." Galaxy masses, stellar core temperatures, X-ray cluster temperatures, and atomic binding energies are all routinely derived from virial relations.
- "Requires bound orbits." Strictly, periodicity or boundedness is needed to drop the surface terms. Steady-state systems (stars, plasmas) satisfy a quasi-stationary version.
- "Velocity dispersion = bulk velocity." No — σ is the spread of velocities about the cluster's mean, not the cluster's overall motion through space. Zwicky measured spread in line-of-sight redshifts.
- "Dark matter was the only explanation possible." Modified gravity (MOND, TeVeS) was a competing hypothesis for decades; only with Bullet Cluster lensing and CMB peaks did dark matter become the consensus.
Frequently asked questions
What does the virial theorem say for gravitational systems?
For a bound system held together by 1/r forces (gravity or Coulomb), the time-averaged kinetic and potential energies satisfy 2⟨T⟩ = −⟨V⟩, equivalently ⟨T⟩ = −⟨V⟩/2. The total mechanical energy is E = ⟨T⟩ + ⟨V⟩ = −⟨T⟩ = ⟨V⟩/2, always negative for bound states. This works in the long-time average regardless of orbit shape — circular orbit, eccentric ellipse, chaotic N-body — as long as the system stays bound and you average long enough that the surface terms in d/dt(p · r) vanish.
Why is total energy negative for bound systems?
Bound means the particle (or cluster of particles) cannot escape to infinity. At infinity, by convention, the gravitational potential is zero and the kinetic energy is zero, so the threshold total energy for escape is zero. Anything below that — anything bound — has E < 0. The virial theorem then makes this quantitative: E equals minus the average kinetic energy, so the more tightly bound you are, the faster you orbit. Doubling speed (4× kinetic) means a deeper bound state by exactly the same factor — a relationship exploited to weigh galaxies.
How did Zwicky use it to find dark matter?
In 1933 Fritz Zwicky measured the redshifts of galaxies in the Coma Cluster and found a velocity dispersion σ ≈ 1000 km/s. The virial theorem for a uniform self-gravitating sphere gives total mass M ≈ 5σ²R/(3G). Plugging in σ and the cluster radius R, Zwicky got a total mass of about 4.5 × 10¹⁴ M_sun — yet the visible (luminous) mass was only about a tenth of that. The cluster could not stay bound at the observed σ unless something dark dominated the mass. He coined the term dunkle Materie (dark matter); the missing-mass problem was confirmed by galaxy rotation curves in the 1970s.
Does it work for non-conservative forces?
The general theorem comes from time-averaging d/dt(Σ p_i · r_i) = 0 over a bound, periodic, or steady-state motion. The right-hand side splits into 2⟨T⟩ + ⟨Σ F_i · r_i⟩. For conservative power-law forces F = −∇V with V ∝ r^n, the second term reduces to −n⟨V⟩. With non-conservative dissipation the relation breaks unless an equivalent steady-state energy balance is set up. Quasi-equilibrium plasmas and stars in hydrostatic equilibrium do satisfy a virial relation despite radiating, because they are in a slow-evolving steady state.
What is the virial coefficient in real-gas thermodynamics?
Different beast, similar lineage. Clausius's original 1870 paper used the virial as a correction to the ideal gas equation pV = NkT to account for intermolecular forces. The modern virial expansion is pV/NkT = 1 + B(T)/V + C(T)/V² + ..., where B, C are virial coefficients computed by integrating molecular potentials. They directly map onto deviations from ideality at high density. The same word, applied to particles in a box rather than orbits in a gravitational well, but mathematically descended from the same time-averaging trick.
How does it apply to atom binding energies?
For a hydrogen atom bound by Coulomb attraction, n = −1 again, so ⟨T⟩ = −⟨V⟩/2. Plug in the energy E_n = −13.6/n² eV: ⟨T⟩ = +13.6/n² eV and ⟨V⟩ = −27.2/n² eV. The same partition holds for any Coulomb-bound system — positronium, muonic atoms, Cooper pairs in superconductors, and the average behavior of multi-electron atoms in the Hartree-Fock approximation. It's a quick consistency check: if your computed ⟨T⟩ and ⟨V⟩ don't satisfy the virial relation, your wavefunction probably is not a stationary state.