Skip to main content
Sandbox Physics
Galaxies in close encounters

How a galaxy orbit loses energy

Internal motions can gain energy at the expense of the mutual orbit. You can track the transfer by writing the kinetic energy in each halo as a bulk term and a remainder.

Separate bulk and internal motion

Write each velocity as its original halo’s center-of-mass velocity plus a velocity relative to that center of mass. The cross term vanishes when summed over the halo because its mass-weighted relative velocities add to zero. The remainder belongs to its internal kinetic energy.

Now sort the gravitational pairs according to whether their members started in the same halo. Same-host terms join the internal energy. Cross-host terms join the bulk kinetic energy. When the haloes overlap, follow the original membership tags to keep the two sets of pairs distinct.

E=Kbulk+UAB+Eint,A+Eint,BE=K_\mathrm{bulk}+U_{AB}+E_{\mathrm{int},A}+E_{\mathrm{int},B}
The bulk term uses the centers of mass. The chart in the Lab uses separate center tags selected by initial binding energy.

Galaxy encounter direct-sum runs and archived numerical checks

A drag estimate for a rigid orbit

A moving massive perturber deflects the paths of lighter background particles. Their sideways changes average away in a uniform isotropic bath. The systematic change along the perturber’s path takes momentum from its orbit.

In the local Maxwellian approximation, use the background mass density and its one-dimensional velocity dispersion at the perturber’s position. X is the speed divided by the dispersion times the square root of two in the denominator. The bracket is the fraction of field particles slower than the perturber. The Coulomb logarithm summarizes an assumed range of encounter distances.

adf=−4πG2Mρln⁡Λ[erf⁡(X)−2Xπe−X2]vv3\boldsymbol{a}_\mathrm{df}=-4\pi G^2 M \rho\ln\Lambda\left[\operatorname{erf}(X)-\frac{2X}{\sqrt{\pi}}e^{-X^2}\right]\frac{\boldsymbol{v}}{v^3}
This is the acceleration of one massive perturber in a local bath. The equal-mass rigid comparison combines the two opposite accelerations and uses a Coulomb logarithm of one.
dEorbdt=μ v⋅adf,rel≤0\frac{\mathrm{d}E_\mathrm{orb}}{\mathrm{d}t}=\mu\,\boldsymbol{v}\cdot\boldsymbol{a}_{\mathrm{df,rel}}\leq 0
Relative velocity and relative drag point in opposite directions. Multiplying by the reduced mass gives the rate of orbital energy change.

Chandrasekhar, Dynamical Friction I (1943)

Check the two velocity limits

Expand the error function and the exponential near zero. Their linear terms cancel and the bracket starts at order speed cubed. After dividing by speed cubed, the coefficient approaches a constant, so the acceleration is proportional to velocity near zero.

Far above the background dispersion, almost everybody in the bath is slower than the perturber. The bracket approaches one. The acceleration magnitude then falls as inverse speed squared. Those two limits give independent checks on an implementation of the middle range.

Chandrasekhar, Dynamical Friction I (1943) · Galaxy encounter direct-sum runs and archived numerical checks

Exercise

An encounter gains 0.12 energy units in one halo and 0.08 in the other. The matched isolated runs show no changes, and total energy is conserved. How much do the bulk and cross-pair terms together lose?

Read the explanation

They lose 0.20 units. That statement accounts for both bulk kinetic energy and the energy of pairs from opposite haloes. It does not assign a separate loss to a fitted friction coefficient.

Try it in the Lab

Follow the recorded prograde pass and compare the no-drag curve with the drag curve. See whether slow approach and post-passage retreat are equally well captured.

Sources

Keep reading

What this page links to directly

See this series on the content map →

Galaxies in close encounters · Reading path

Galaxies in close encounters

  1. ArticleWhy galaxies merge even though their stars rarely meet
  2. ArticleTidal acceleration across a galaxy disk
  3. You are hereHow a galaxy orbit loses energy
  4. ArticleCompare a drag orbit with a direct galaxy simulation
  5. ArticleRead how black-hole binaries shrink in a galaxy merger
Something broken, confusing, or missing on this page?Tell us
What kind of feedback?

Sent with this page’s address, its Lab ID, the site language and your device type. Please leave out personal details. Feedback is kept for up to 12 months.Privacy policy