Skip to main content
Sandbox Physics
How dark matter was left over

The Boltzmann equation and freeze-out

Sandbox Physics ·

The number of dark particles per unit entropy obeys one equation: it falls at a rate set by annihilation minus inverse production, and nothing else changes it. Solved for a 100 GeV particle, it leaves Y∞=4.37×10−12Y_\infty=4.37\times10^{-12} when the low-temperature cross-section is 2.136×10−26 cm3 s−12.136\times10^{-26}\,\mathrm{cm^3\,s^{-1}}, which is today's measured density Ωh2=0.120\Omega h^2=0.120. A two-line approximation gets within ten per cent of that, and it shows why the leftover is inversely proportional to the cross-section.

This chapter follows Equilibrium, expansion and the reaction race and uses its units and symbols. You need separable differential equations. The aims are to derive the equation for YY, solve it in its two regimes, convert the result into a density today and see why a computer needs an implicit method.

From particle number to particles per entropy

Count dark particles in a box that grows with the universe. Annihilation removes pairs at the rate ⟨σv⟩n2\langle\sigma v\rangle n^2 per unit volume. Inverse production, bath particles colliding to make a pair, adds them at ⟨σv⟩neq2\langle\sigma v\rangle n_{\mathrm{eq}}^2, because detailed balance requires the two to cancel in equilibrium. Dilution by the growing volume removes 3Hn3Hn. So

dndt+3Hn=−⟨σv⟩(n2−neq2).\frac{\mathrm dn}{\mathrm dt}+3Hn=-\langle\sigma v\rangle\left(n^2-n_{\mathrm{eq}}^2\right).

The Particle Data Group review prints the dilution term with a minus sign, dn/dt−3Hn\mathrm dn/\mathrm dt-3Hn. Particle Data Group, 2025, eq. 27.2 That sign would make the density grow without reactions. The plus sign used here is the one in D'Eramo, Fernandez and Profumo's eq. (15) and in Steigman and colleagues' eq. (1), and it follows from na3na^3 being constant when nothing reacts. D'Eramo, Fernandez and Profumo, 2017, eq. 15

Now divide by ss. Because sa3sa^3 is constant, d(n/s)/dt=(n˙+3Hn)/s\mathrm d(n/s)/\mathrm dt=(\dot n+3Hn)/s. With T∝1/aT\propto1/a the clock x=m/Tx=m/T obeys dx/dt=Hx\mathrm dx/\mathrm dt=Hx, so

dYdln⁡x=−s⟨σv⟩H(Y2−Yeq2).\frac{\mathrm dY}{\mathrm d\ln x}=-\frac{s\langle\sigma v\rangle}{H}\left(Y^2-Y_{\mathrm{eq}}^2\right).

In the radiation era s∝T3s\propto T^3 and H∝T2H\propto T^2, so s/Hs/H is a constant times 1/x1/x. Write ⟨σv⟩=⟨σv⟩0F(x)\langle\sigma v\rangle=\langle\sigma v\rangle_0F(x), where F→1F\to1 for slow particles. Then

dYdln⁡x=−λF(x)x(Y2−Yeq2),λ=s⟨σv⟩0H∣T=m=2π1015g∗ mMPl⟨σv⟩0.\frac{\mathrm dY}{\mathrm d\ln x}=-\frac{\lambda F(x)}{x}\left(Y^2-Y_{\mathrm{eq}}^2\right),\qquad \lambda=\frac{s\langle\sigma v\rangle_0}{H}\bigg|_{T=m}=\frac{2\pi\sqrt{10}}{15}\sqrt{g_*}\,mM_{\mathrm{Pl}}\langle\sigma v\rangle_0.

For m=100 GeVm=100\,\mathrm{GeV} and ⟨σv⟩0=1.830×10−9 GeV−2\langle\sigma v\rangle_0=1.830\times10^{-9}\,\mathrm{GeV^{-2}}, λ=5.48×1012\lambda=5.48\times10^{12}. This single number sets the whole problem. Note that λYeqF/x\lambda Y_{\mathrm{eq}}F/x is exactly the Γ/H\Gamma/H of the previous chapter.

Two regimes and a match

While Γ/H≫1\Gamma/H\gg1, any excess Δ=Y−Yeq\Delta=Y-Y_{\mathrm{eq}} is destroyed almost at once. Setting dY/dln⁡x≃dYeq/dln⁡x\mathrm dY/\mathrm d\ln x\simeq\mathrm dY_{\mathrm{eq}}/\mathrm d\ln x gives a small lag, Δ≃x2/(2λF)\Delta\simeq x^2/(2\lambda F) for x≫1x\gg1, so YY tracks equilibrium. At x=20x=20 this is 4.2×10−114.2\times10^{-11}, 6 per cent of YeqY_{\mathrm{eq}}, as the full solution confirms.

Once YeqY_{\mathrm{eq}} has collapsed, drop it. The equation separates:

1Y∞=1Y(xf)+λ∫xf∞F(x)x2 dx  ≃  λxf.\frac{1}{Y_\infty}=\frac{1}{Y(x_f)}+\lambda\int_{x_f}^{\infty}\frac{F(x)}{x^2}\,\mathrm dx\;\simeq\;\frac{\lambda}{x_f}.

The last step uses F≃1F\simeq1 and Y(xf)≫Y∞Y(x_f)\gg Y_\infty. The matching point xfx_f is fixed by asking where the lag Δ\Delta becomes comparable to YeqY_{\mathrm{eq}}. D'Eramo and colleagues set Δ=cYeq\Delta=cY_{\mathrm{eq}} and obtain

exfxf1/2≃c(c+2)c+1 35 g2π5/2g∗ mMPl⟨σv⟩0,\mathrm e^{x_f}x_f^{1/2}\simeq\frac{c(c+2)}{c+1}\,\frac{3\sqrt5\,g}{2\pi^{5/2}\sqrt{g_*}}\,mM_{\mathrm{Pl}}\langle\sigma v\rangle_0,

their eq. (C12) with a constant cross-section, where cc is a number of order one. D'Eramo, Fernandez and Profumo, 2017, appendix C It follows from putting Yeq∝x3/2e−xY_{\mathrm{eq}}\propto x^{3/2}\mathrm e^{-x} into the condition that the lag reaches cYeqcY_{\mathrm{eq}}, and its constant is the same combination as λ\lambda times the prefactor of YeqY_{\mathrm{eq}}. Taking c=2−1c=\sqrt2-1 and iterating xf=ln⁡(…)−12ln⁡xfx_f=\ln(\ldots)-\tfrac12\ln x_f gives xf=21.7x_f=21.7. This matching point sits earlier than the crossing Γ/H=1\Gamma/H=1 at x=25.2x=25.2; both are markers on a gradual process.

Worked example: the observed density

With xf=21.7x_f=21.7 and λ=5.48×1012\lambda=5.48\times10^{12},

Y∞≃xfλ=3.97×10−12.Y_\infty\simeq\frac{x_f}{\lambda}=3.97\times10^{-12}.

To turn this into a density today, multiply by the mass and today's entropy density s0=2891 cm−3s_0=2891\,\mathrm{cm^{-3}}, and divide by the critical density ρc/h2=1.054×10−5 GeV cm−3\rho_c/h^2=1.054\times10^{-5}\,\mathrm{GeV\,cm^{-3}}:

Ωh2=m Y∞ s0ρc/h2=2.74×108 mGeV Y∞=0.109.\Omega h^2=\frac{m\,Y_\infty\,s_0}{\rho_c/h^2}=2.74\times10^{8}\,\frac{m}{\mathrm{GeV}}\,Y_\infty=0.109.

The numerical solution with a constant cross-section gives Y∞=4.11×10−12Y_\infty=4.11\times10^{-12}, so the matching estimate is 3.5 per cent low. With the Lab's F(x)F(x), which is 0.86 to 0.9 near freeze-out, annihilation is slightly weaker and the full answer is 4.37×10−124.37\times10^{-12}, or Ωh2=0.120\Omega h^2=0.120. Planck measures 0.120±0.0010.120\pm0.001. Planck Collaboration, 2018, abstract

Table 3.1 The full solution for the default Lab settings.
xxYYY/YeqY/Y_{\mathrm{eq}}Y/Y∞Y/Y_\infty
10105.69×10−65.69\times10^{-6}1.001.001.3×1061.3\times10^{6}
20207.14×10−107.14\times10^{-10}1.061.06163163
25.225.23.42×10−113.42\times10^{-11}6.766.767.837.83
50507.81×10−127.81\times10^{-12}4×10104\times10^{10}1.781.78
1001005.55×10−125.55\times10^{-12}6×10316\times10^{31}1.271.27
2082084.81×10−124.81\times10^{-12}2×10782\times10^{78}1.101.10
100010004.37×10−124.37\times10^{-12}1.001.00

Freeze-out is a process spread over a factor of ten in temperature. At Γ/H=1\Gamma/H=1 there are still almost eight times the final number, and the last ten per cent disappears only by x≃200x\simeq200.

Why the leftover goes as one over the cross-section

Y∞≃xf/λY_\infty\simeq x_f/\lambda and λ∝⟨σv⟩0\lambda\propto\langle\sigma v\rangle_0, while xfx_f grows only logarithmically. A larger cross-section therefore leaves proportionally less. The saved scan gives Ωh2=2.23\Omega h^2=2.23, 0.2480.248 and 0.02730.0273 at 10−2710^{-27}, 10−2610^{-26} and 10−25 cm3 s−110^{-25}\,\mathrm{cm^3\,s^{-1}}: each decade costs a factor of nine rather than ten, because xfx_f rises by about 2.3 per decade.

The required cross-section barely depends on the mass. In this model it is 1.92×10−261.92\times10^{-26} for 10 GeV, 2.14×10−262.14\times10^{-26} for 100 GeV and 2.36×10−26 cm3 s−12.36\times10^{-26}\,\mathrm{cm^3\,s^{-1}} for 1 TeV. The careful calculation of Steigman, Dasgupta and Beacom, with the full temperature dependence of g∗g_*, gives 2.2×10−262.2\times10^{-26} within about 5 per cent above 10 GeV for Ωh2=0.11\Omega h^2=0.11; their fit, rescaled to 0.120, gives 1.98×10−261.98\times10^{-26}. Steigman, Dasgupta and Beacom, 2012, section II.D The Lab is 8 per cent higher, mostly because it holds g∗g_* fixed.

Why a computer needs an implicit step

At x=1x=1, Γ/H=3.3×109\Gamma/H=3.3\times10^{9}: the equation tries to restore equilibrium a billion times per e-fold of cooling. An explicit step of any practical size overshoots and blows up. The saved model therefore uses the second-order backward differentiation formula in ln⁡x\ln x. Each step asks for the new YY on both sides of the equation, which turns it into a quadratic,

aY2+Y−c=0,Y=2c1+1+4ac,aY^2+Y-c=0,\qquad Y=\frac{2c}{1+\sqrt{1+4ac}},

with aa and cc built from the step size, λF/x\lambda F/x, the two previous values and Yeq2Y_{\mathrm{eq}}^2. This root is never negative and never loses precision by subtraction. The method guide Recompute a relic abundance with the saved scripts shows the explicit step failing and checks the implicit one against an independent time-domain solution.

Lab tasks

Open the dark matter relic Lab.

  1. On the default run, read Y∞Y_\infty and the settling point. Then compare them with the last two rows of Table 3.1.
  2. Move the cross-section slider up by exactly one in the logarithm. Predict Ωh2\Omega h^2 from the factor-of-nine rule, then read it.
  3. Switch the start to "From nothing". Explain why the final density does not change.

Exercises

  1. Starting from dn/dt+3Hn=…\mathrm dn/\mathrm dt+3Hn=\ldots and s∝T3s\propto T^3, H∝T2H\propto T^2, derive the equation for dY/dln⁡x\mathrm dY/\mathrm d\ln x and the expression for λ\lambda.
  2. At x=25.2x=25.2 the full solution has Y=3.42×10−11Y=3.42\times10^{-11}. Use ∫25.2∞F/x2 dx=0.0374\int_{25.2}^\infty F/x^2\,\mathrm dx=0.0374 to estimate Y∞Y_\infty, and explain the remaining difference from 4.37×10−124.37\times10^{-12}.
  3. A particle of 1 TeV needs ⟨σv⟩0=2.36×10−26 cm3 s−1\langle\sigma v\rangle_0=2.36\times10^{-26}\,\mathrm{cm^3\,s^{-1}} in this model. Show from Y∞≃xf/λY_\infty\simeq x_f/\lambda why the value rises only slightly with mass.
  4. Show that dn/dt−3Hn=0\mathrm dn/\mathrm dt-3Hn=0 would make na−3na^{-3} constant, and explain why that cannot describe dilution.
Answers
  1. Y˙=(n˙+3Hn)/s=−(⟨σv⟩s)(Y2−Yeq2)\dot Y=(\dot n+3Hn)/s=-(\langle\sigma v\rangle s)(Y^2-Y_{\mathrm{eq}}^2). Divide by x˙=Hx\dot x=Hx: dY/dln⁡x=−(s⟨σv⟩/H)(Y2−Yeq2)\mathrm dY/\mathrm d\ln x=-(s\langle\sigma v\rangle/H)(Y^2-Y_{\mathrm{eq}}^2). With s=(2π2/45)g∗T3s=(2\pi^2/45)g_*T^3 and H=π2g∗/90 T2/MPlH=\sqrt{\pi^2g_*/90}\,T^2/M_{\mathrm{Pl}}, s/H=(2π10/15)g∗ MPlT=(2π10/15)g∗ mMPl/xs/H=(2\pi\sqrt{10}/15)\sqrt{g_*}\,M_{\mathrm{Pl}}T=(2\pi\sqrt{10}/15)\sqrt{g_*}\,mM_{\mathrm{Pl}}/x.
  2. 1/Y∞=1/(3.42×10−11)+5.48×1012×0.0374=2.92×1010+2.05×10111/Y_\infty=1/(3.42\times10^{-11})+5.48\times10^{12}\times0.0374=2.92\times10^{10}+2.05\times10^{11}, so Y∞=4.27×10−12Y_\infty=4.27\times10^{-12}, 2.4 per cent below the full value. Just after x=25.2x=25.2 inverse production, the Yeq2Y_{\mathrm{eq}}^2 term, still adds a little.
  3. λ∝m\lambda\propto m, so at fixed cross-section Y∞∝xf/mY_\infty\propto x_f/m and Ωh2∝mY∞∝xf\Omega h^2\propto mY_\infty\propto x_f. The mass cancels except through xfx_f, which grows by ln⁡10≃2.3\ln10\simeq2.3 per decade of mass. Keeping Ωh2\Omega h^2 fixed then needs a cross-section larger by about xf(1 TeV)/xf(100 GeV)≃1.1x_f(1\,\mathrm{TeV})/x_f(100\,\mathrm{GeV})\simeq1.1.
  4. d(na−3)/dt=a−3(n˙−3Hn)=0\mathrm d(na^{-3})/\mathrm dt=a^{-3}(\dot n-3Hn)=0, so n∝a3n\propto a^3 would grow as the volume grows. A fixed number of particles in a growing volume must have n∝a−3n\propto a^{-3}, which is n˙+3Hn=0\dot n+3Hn=0.

Sources

  1. Particle Data Group, 2025, eq. 27.2
  2. D'Eramo, Fernandez and Profumo, 2017, eq. 15
  3. Planck Collaboration, 2018, abstract
  4. Steigman, Dasgupta and Beacom, 2012, section II.D

Keep reading

What this page links to directly

See this series on the content map →

How dark matter was left over · Reading path

From thermal equilibrium to relentless dark matter

  1. ArticleWhy did any dark matter survive the hot early universe?
  2. ArticleEquilibrium, expansion and the reaction race
  3. You are hereThe 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