Skip to main content
Sandbox Physics

Dark Matter Relics: Freeze-Out, Freeze-In and Thermal Decoupling

How much dark matter is left after the early universe cools? The amount left is set when annihilation can no longer keep up with expansion. Particles that annihilate weakly fall out of balance early and leave many survivors, while strong annihilation leaves few. Particles too feeble to ever reach balance are instead made slowly from the hot bath, and that freeze-in yield grows with the coupling.

Watch dark particles annihilate and appear in a cooling bath. Drag the reaction rate against the expansion rate and see when the count stops changing, whether the particles start in equilibrium or from nothing.

Interactive modelDark Matter Relics: Freeze-Out, Freeze-In and Thermal Decoupling
Dark matter density today Ωh2\Omega h^20.1200.120
Particles per unit entropy Y∞Y_\infty4.37×10−124.37\times10^{-12}
Fastest reaction against expansion (Γ/H)max⁡(\Gamma/H)_{\max}3.27×1093.27\times10^{9}
Count settles by x=m/Tx=m/T208208

What the equation leaves behind

Offline teaching model; the gold line is the Planck 2018 measurement

Particles per unit entropy Y=n/sY=n/s

this calculationequilibriumamount that gives the observed density

Reaction against expansion Γ/H\Gamma/H

equilibrium reaction raterate for the particles presentexpansion rate

Density today for every annihilation strength Ωh2 vs ⟨σv⟩0\Omega h^2\ \text{vs}\ \langle\sigma v\rangle_0

start in equilibriumstart from nothingPlanck 2018 value

Follow dark particles through the cooling universe

BackgroundIn the first fraction of a second the universe was a hot soup of particles. If dark matter particles touched that soup, they were made and destroyed in pairs. As the universe expanded and cooled, those reactions slowed down.

Why it mattersThe number left over after the reactions stop is the dark matter we see today. The same equation gives very different answers depending on how strongly the particles interact and whether they started in balance.

One-sentence intuition
The count per unit entropy stops changing once the reaction rate falls below the expansion rate. For particles in balance, stronger annihilation keeps them in balance longer and leaves fewer. For particles never in balance, stronger coupling makes more. The two branches meet at a hot-relic plateau, and a faster early expansion can freeze particles out early but let annihilation keep thinning them.

Core mathematical model

The Boltzmann equation per unit entropy

dYdln⁡x=−s⟨σv⟩H(Y2−Yeq2),Y=ns,x=mT\frac{dY}{d\ln x}=-\frac{s\langle\sigma v\rangle}{H}\left(Y^2-Y_{\rm eq}^2\right),\qquad Y=\frac{n}{s},\quad x=\frac{m}{T}

Dividing by the entropy density removes the dilution by expansion. Only annihilation and its inverse change the count.

The equilibrium count

Yeq=45 g4π4g∗s x2K2(x)  →  {90 g4π4g∗s,x≪145 g4π4g∗sπ2 x3/2e−x,x≫1Y_{\rm eq}=\frac{45\,g}{4\pi^4 g_{*s}}\,x^2K_2(x)\;\to\;\begin{cases}\dfrac{90\,g}{4\pi^4g_{*s}}, & x\ll1\\[1ex]\dfrac{45\,g}{4\pi^4g_{*s}}\sqrt{\dfrac{\pi}{2}}\,x^{3/2}e^{-x}, & x\gg1\end{cases}

A constant while the particles are relativistic, then a Boltzmann suppression once the bath is too cold to make them.

The race between reaction and expansion

ΓH=neq⟨σv⟩H,H=π2g∗90 T2MPl1+(TTr)n\frac{\Gamma}{H}=\frac{n_{\rm eq}\langle\sigma v\rangle}{H},\qquad H=\sqrt{\frac{\pi^2 g_*}{90}}\,\frac{T^2}{M_{\rm Pl}}\sqrt{1+\left(\frac{T}{T_r}\right)^n}

While this ratio is large the count tracks equilibrium. Freeze-out happens near where it falls through one. The extra factor in the expansion rate is the faster-expansion option.

Two closed-form answers

Y∞FO≃xfλ,λ=s ⟨σv⟩0H∣T=m;Y∞FI=3π232(45g4π4g∗s)2λY_\infty^{\rm FO}\simeq\frac{x_f}{\lambda},\quad \lambda=\frac{s\,\langle\sigma v\rangle_0}{H}\Big|_{T=m};\qquad Y_\infty^{\rm FI}=\frac{3\pi^2}{32}\left(\frac{45g}{4\pi^4g_{*s}}\right)^2\lambda

Freeze-out leaves an amount inversely proportional to the cross-section. Freeze-in from nothing makes an amount proportional to it; the second result holds for this Lab’s contact interaction.

Common difficulties

Freeze-out is not a sudden stop

Typical misconceptionAll annihilation stops at the freeze-out temperature.

Better mental modelThe count leaves equilibrium near x≈25x\approx25 here, but it keeps falling until xx reaches a few hundred. The readout for when the count settles shows how long the tail lasts.

Stronger is not always more

Typical misconceptionA stronger interaction always makes more dark matter.

Better mental modelFor particles in balance, stronger annihilation destroys more and leaves less. Only below the hot-relic plateau, where particles never reached balance, does a stronger coupling make more.

The starting state can stop mattering

Typical misconceptionStarting from nothing must always give less dark matter than starting in balance.

Better mental modelWhen the reaction rate is large, particles made from nothing reach equilibrium quickly and then freeze out exactly as before. The start only matters below the plateau.

Run the experiment

  1. 01

    Watch freeze-out

    Keep the default and watch the cyan reaction bar fall past the orange expansion bar.

    What to observe: The violet dots first vanish in flashes, then stop vanishing once the reaction loses the race. The final density is the observed 0.120.
  2. 02

    Drag the reaction stronger

    Drag the cyan handle up by two decades, or choose Stronger annihilation.

    What to observe: Equilibrium lasts longer and the survivors drop more than eightyfold.
  3. 03

    Go below the plateau

    Choose Freeze-in from nothing, then switch the start back to equilibrium.

    What to observe: From nothing, a feeble coupling slowly makes just the observed amount. Starting in balance with the same coupling leaves a hot relic a billion times too dense.
  4. 04

    Speed up the early expansion

    Choose Faster expansion: relentless annihilation.

    What to observe: The particles leave equilibrium earlier, yet the count keeps falling until the extra fluid fades near 20 MeV. The same cross-section now leaves about 34 times more.

Related experiments and articles

What this page links to directly

See every article series on the content map →

How dark matter was left over · Reading path

From thermal equilibrium to relentless dark matter

Articles on this path that use this experiment, followed by the other experiments they suggest.

  1. ArticleWhy did any dark matter survive the hot early universe?
  2. ArticleEquilibrium, expansion and the reaction race
  3. ArticleThe Boltzmann equation and freeze-out
  4. ArticleFreeze-in, hot relics and faster expansion
  5. ArticleRecompute a relic abundance with the saved scripts
  6. ArticleHow to read the relentless dark matter paper of D'Eramo and colleagues
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