The Cosmological Constant

The cosmological constant holds a singular place in the history of physics: it is the term Einstein wished he had never written and the number the universe insisted on anyway. In 1917 he added $\Lambda$ to his field equations to hold the cosmos still, since gravity acting on ordinary matter cannot support a static universe against collapse. Hubble's expanding sky made the fix look foolish, and Einstein set $\Lambda$ aside — yet the term never actually left. It is the unique modification of the Einstein equations consistent with general covariance and a covariantly conserved source, so nothing in the theory forbids it; the only question was ever its numerical value. The decisive insight is that $\Lambda$ , though written as geometry, is mathematically identical to a source: a perfect fluid with energy density $\rho_\Lambda c^2 = \Lambda c^4/(8\pi G)$ and pressure $p_\Lambda = -\rho_\Lambda c^2$ , an equation of state $w \equiv p/(\rho c^2) = -1$ . That negative pressure is everything. In general relativity it is the combination $\rho c^2 + 3p$ that gravitates, and for $\Lambda$ this quantity is $-2\rho_\Lambda c^2 < 0$ : the vacuum gravitates repulsively. Because a $w=-1$ fluid does not dilute as space grows, it is negligible early on but inevitably comes to dominate, driving the Friedmann equation toward a de Sitter solution with constant Hubble rate and exponential growth. In 1998 two supernova teams found exactly this signature — the expansion is speeding up — and $\Lambda$ returned in triumph as the leading model of dark energy, with $\Omega_{\Lambda,0} \approx 0.69$ of today's cosmic budget. Whether it is literally quantum vacuum energy, a dynamical field, or a crack in general relativity remains open, and the $10^{120}$ gap between its observed value and naive theory is among the deepest puzzles in physics.

Key takeaways

The cosmological constant $\Lambda$ is the one term general covariance allows you to add to Einstein's equations, $G_{\mu\nu}+\Lambda g_{\mu\nu}=8\pi G\,T_{\mu\nu}/c^4$ ; it behaves as a uniform energy filling all of space.
Moved to the source side it is a perfect fluid with $\rho_\Lambda c^2=\Lambda c^4/(8\pi G)$ and $p_\Lambda=-\rho_\Lambda c^2$ — equation of state $w=-1$ .
Because $w=-1$ the vacuum density does not dilute, $\rho_\Lambda\propto a^0$ , so it eventually overwhelms matter ( $\propto a^{-3}$ ) and radiation ( $\propto a^{-4}$ ).
Its negative pressure makes $\rho c^2+3p<0$ , so through the acceleration equation $\Lambda$ drives $\ddot a>0$ — the cosmic acceleration discovered with Type Ia supernovae in 1998.
In the standard $\Lambda$ CDM model $\Omega_{\Lambda,0}\approx0.69$ , the deceleration parameter is $q_0\approx-0.53$ , and the far future is a de Sitter phase, $a(t)\propto e^{H_\Lambda t}$ .
The measured $\rho_\Lambda\sim6\times10^{-27}\ \mathrm{kg\,m^{-3}}$ is roughly $10^{120}$ times smaller than Planck-scale vacuum estimates — the cosmological constant problem.
Key idea: A positive cosmological constant is a form of energy whose density does not dilute as space expands; its negative pressure ( $w=-1$ ) makes $\rho c^2+3p<0$ , driving accelerated expansion and eventually an exponential (de Sitter) phase.

Where this lives today

The cosmological constant is no longer a theoretical curiosity but the largest measured slice of the universe, and pinning down whether it is truly constant is the central program of observational cosmology. The Dark Energy Spectroscopic Instrument (DESI) is mapping tens of millions of galaxies to trace the expansion history $H(z)$ through baryon acoustic oscillations; its 2024–2025 data releases hint tantalizingly that dark energy may be evolving rather than a pure $w=-1$ constant, a result that, if it holds, would rule out a plain $\Lambda$ . The Euclid space telescope, launched in 2023, and the Nancy Grace Roman Space Telescope are being flown expressly to measure $w$ and its possible time dependence $w_a$ to percent precision by combining weak gravitational lensing, galaxy clustering, and supernovae. Meanwhile the "Hubble tension" — the $5\sigma$ disagreement between the locally measured $H_0\approx73$ and the CMB-inferred $\approx67$ km/s/Mpc — may signal that the dark-energy sector, or the early expansion history, is not quite $\Lambda$ CDM. And theory still has no accepted answer to the $10^{120}$ problem: proposals range from anthropic selection within a string-theory landscape, to dynamical quintessence fields that relax the vacuum energy, to modifications of general relativity on cosmic scales. The cosmological constant is simultaneously the best-fitting parameter in cosmology and its least-understood number.

The physical picture

Picture an energy that belongs to space itself rather than to anything sitting in it. Take a box of empty space and double its volume; you now hold twice as much vacuum energy, but the energy per unit volume is exactly what it was. Ordinary matter does the opposite — spread the same particles through a larger box and the density falls. This single difference, constant density versus diluting density, is the entire dynamical story of $\Lambda$ : as the universe grows, matter and radiation thin out while the vacuum term holds fixed, so a component that is utterly negligible in the fireball of the early universe is guaranteed to inherit the cosmos in the end. The counterintuitive twist is that this constant-density stuff pushes. In general relativity pressure gravitates alongside energy, and the source of gravitational acceleration is the sum $\rho c^2 + 3p$ , not $\rho$ alone. A vacuum with $p = -\rho_\Lambda c^2$ makes that sum equal to $-2\rho_\Lambda c^2$ , a negative number — so its net gravity is repulsive. A helpful image is a stretched rubber sheet under uniform tension: tension is simply negative pressure, and Einstein's equations say that a uniform tension filling space curves spacetime so that free-falling galaxies drift apart ever faster. The first figure fixes the bookkeeping — volume doubles, density stays put, pressure is a tension, and the gravitational source term comes out negative.

Figure

A comoving box of space doubling in volume: the vacuum energy density ρ_Λ stays fixed so the total energy rises in step with the volume, while the pressure is negative, p_Λ = −ρ_Λc², a tension whose gravitational source term ρc² + 3p = −2ρ_Λc² is negative, so the vacuum gravitates repulsively and accelerates the expansion

Which component rules the universe therefore changes over cosmic history, and $\Lambda$ is always the last to win. In the radiation era its constant density is drowned by $\rho_r \propto a^{-4}$ ; through the long matter era it trails $\rho_m \propto a^{-3}$ ; but both of those fall while the vacuum term does not, so there is a crossover, and after it the vacuum dominates forever. The second figure tracks all three densities against the scale factor and shows the crossings: matter overtakes radiation early, and the flat vacuum line overtakes matter only recently, near $a \approx 0.77$ (redshift $z \approx 0.3$ ) — startlingly close to today, the fact behind the "cosmic coincidence."

Figure

Densities versus scale factor on log axes for a flat ΛCDM universe with Ω_m = 0.31, Ω_Λ = 0.69: radiation ρ ∝ a⁻⁴ and matter ρ ∝ a⁻³ dilute as space grows while the vacuum term ρ_Λ stays constant, so Λ inevitably overtakes matter near a ≈ 0.77 (z ≈ 0.3) and dominates today at a = 1, with matter–radiation equality far earlier near a ≈ 3×10⁻⁴ (z ≈ 3400)

The core equations

The starting point is the field equation itself. Einstein's equations relate spacetime curvature to its matter–energy content; the cosmological constant is the single extra term that respects general covariance and leaves the source covariantly conserved, so it can be attached to the geometry side with a free coefficient $\Lambda$ .

G_{\mu\nu}+\Lambda g_{\mu\nu}=\frac{8\pi G}{c^{4}}\,T_{\mu\nu} \qquad (1)

where: $G_{\mu\nu}$ — Einstein tensor $R_{\mu\nu}-\tfrac12 R g_{\mu\nu}$ , encoding curvature; $\Lambda$ — cosmological constant, units $\mathrm{m^{-2}}$ ; $g_{\mu\nu}$ — metric tensor; $T_{\mu\nu}$ — stress–energy of matter and radiation; $G,c$ — Newton's constant and the speed of light. Reading: $\Lambda g_{\mu\nu}$ is the only term you may add without spoiling $\nabla^\mu G_{\mu\nu}=0$ , which is why it is not optional but merely undetermined in size. Move that term to the right-hand side and it is no longer geometry but a source. Matching it to a perfect fluid identifies a vacuum energy density and a vacuum pressure of equal magnitude and opposite sign.

\rho_\Lambda c^{2}=\frac{\Lambda c^{4}}{8\pi G},\qquad p_\Lambda=-\rho_\Lambda c^{2} \qquad (2)

where: $\rho_\Lambda$ — mass-equivalent vacuum energy density; $p_\Lambda$ — vacuum pressure, negative. Reading: the equation of state is $w \equiv p_\Lambda/(\rho_\Lambda c^2) = -1$ exactly, and the negative pressure — a tension — is the feature that will drive acceleration. Inserting the FLRW metric and a fluid source into (1) yields the Friedmann equation, now carrying an explicit $\Lambda$ term. It ties the square of the expansion rate to the energy density, the spatial curvature, and the vacuum contribution.

\left(\frac{\dot a}{a}\right)^{2}=\frac{8\pi G}{3}\rho-\frac{kc^{2}}{a^{2}}+\frac{\Lambda c^{2}}{3} \qquad (3)

where: $a(t)$ — scale factor; $\rho$ — density of matter plus radiation (excluding $\Lambda$ ); $k$ — curvature constant $\{-1,0,+1\}$ ; $\Lambda c^2/3$ — the vacuum term. Reading: because $\Lambda$ enters with a fixed coefficient and $\rho$ falls with $a$ , the last term eventually sets the expansion rate all by itself. Differentiating (3) and using the fluid equation gives the acceleration (second Friedmann) equation. It is the one that decides whether the expansion slows or speeds up, and it is where the sign of $\rho c^2 + 3p$ matters.

\frac{\ddot a}{a}=-\frac{4\pi G}{3}\left(\rho+\frac{3p}{c^{2}}\right)+\frac{\Lambda c^{2}}{3} \qquad (4)

where: $\ddot a$ — second time derivative of the scale factor; $p$ — pressure of matter plus radiation. Reading: ordinary content with $\rho + 3p/c^2 > 0$ decelerates; the standalone $+\Lambda c^2/3$ is strictly positive, so a positive $\Lambda$ always pushes toward acceleration. The fluid (continuity) equation governs how each component's density evolves. Feeding in the vacuum equation of state $w=-1$ makes the whole bracket vanish, so the vacuum density is constant in time.

\dot\rho+3\frac{\dot a}{a}\left(\rho+\frac{p}{c^{2}}\right)=0 \;\Longrightarrow\; \rho\propto a^{-3(1+w)},\quad \rho_\Lambda=\text{const} \qquad (5)

where: $\dot\rho$ — rate of change of density; $w$ — equation-of-state parameter. Reading: matter ( $w=0$ ) gives $\rho\propto a^{-3}$ , radiation ( $w=1/3$ ) gives $a^{-4}$ , and vacuum ( $w=-1$ ) gives $a^{0}$ — no dilution at all. Comparing any density to the critical value defines a density parameter; for the vacuum this packages $\Lambda$ into a single dimensionless number, and the components sum to unity through the closure relation.

\Omega_\Lambda\equiv\frac{\rho_\Lambda}{\rho_{\rm crit}}=\frac{\Lambda c^{2}}{3H^{2}},\qquad \Omega_m+\Omega_r+\Omega_\Lambda+\Omega_k=1 \qquad (6)

where: $\rho_{\rm crit}=3H^2/(8\pi G)$ — critical density; $H=\dot a/a$ — Hubble parameter; $\Omega_k=-kc^2/(a^2H^2)$ — curvature parameter. Reading: measurements give $\Omega_{\Lambda,0}\approx0.69$ and $\Omega_{k,0}\approx0$ , so the vacuum is the largest single slice of today's energy budget. When $\Lambda$ dominates and curvature and matter are negligible, (3) has a constant right-hand side, and its solution is exponential — de Sitter space with a fixed Hubble rate.

a(t)=a_0\,e^{H_\Lambda(t-t_0)},\qquad H_\Lambda=c\sqrt{\Lambda/3}=H_0\sqrt{\Omega_{\Lambda,0}} \qquad (7)

where: $H_\Lambda$ — constant vacuum-dominated Hubble rate; $a_0,t_0$ — reference scale factor and time. Reading: the expansion rate stops changing, every separation doubles in a fixed time $\ln 2/H_\Lambda$ , and the Hubble radius $c/H_\Lambda$ freezes into a genuine event horizon. Finally, the deceleration parameter turns the acceleration equation into one observable number, evaluated today from the density parameters.

q_0\equiv-\frac{\ddot a\,a}{\dot a^{2}}\bigg|_0=\frac12\Omega_{m,0}-\Omega_{\Lambda,0} \qquad (8)

where: $q_0$ — present deceleration parameter. Reading: $q_0>0$ is a decelerating universe and $q_0<0$ an accelerating one; the measured $q_0\approx-0.53$ is the compact statement that the cosmos is speeding up.

Derivation AWhy the cosmological constant is a fluid with equation of state minus one

The goal is to show that the purely geometric term $\Lambda g_{\mu\nu}$ is mathematically indistinguishable from a physical source, and to pin down that its equation of state is $w=-1$ exactly, with a positive energy density $\rho_\Lambda c^2 = \Lambda c^4/(8\pi G)$ and an equal-magnitude negative pressure. We do this by relocating $\Lambda$ from the curvature side of Einstein's equations to the stress–energy side and demanding that the result take the standard perfect-fluid form, which forces the equation of state uniquely. The payoff is that everything the rest of the article says about vacuum "energy" and "pressure" is licensed by this one algebraic move.

Step 1. Rearrange the field equation (1) so that $\Lambda$ sits with the source: $G_{\mu\nu} = \frac{8\pi G}{c^4}\big(T_{\mu\nu} - \frac{\Lambda c^4}{8\pi G} g_{\mu\nu}\big)$ , and define an effective vacuum stress–energy $T^{(\Lambda)}_{\mu\nu} = -\frac{\Lambda c^4}{8\pi G} g_{\mu\nu}$ . Why: this makes the choice explicit — a constant times the metric can be read either as curvature that was always allowed or as a peculiar kind of matter, and the two readings are physically identical.

Step 2. Compare with the perfect-fluid form $T_{\mu\nu} = (\rho c^2 + p)\,u_\mu u_\nu/c^2 + p\,g_{\mu\nu}$ . The vacuum term is proportional to $g_{\mu\nu}$ alone, with no piece proportional to $u_\mu u_\nu$ , which can only happen if $\rho c^2 + p = 0$ . Why: the velocity-dependent part of any real fluid singles out a rest frame; a term built only from the metric has none, so it must look the same to every observer — and that Lorentz invariance is exactly the condition $p=-\rho c^2$ .

Step 3. Match the surviving coefficient of $g_{\mu\nu}$ : $p_\Lambda = -\frac{\Lambda c^4}{8\pi G}$ , hence $\rho_\Lambda c^2 = -p_\Lambda = +\frac{\Lambda c^4}{8\pi G}$ and $w = p_\Lambda/(\rho_\Lambda c^2) = -1$ . Why: a positive $\Lambda$ gives positive energy density and negative pressure of the same magnitude, the defining signature of vacuum energy and the input to every later result.

Go deeper · Derivation B · 1 minExponential expansion of a vacuum-dominated universe

Derivation BExponential expansion of a vacuum-dominated universe

The goal is to solve the Friedmann equation in the limit where the cosmological constant dominates over matter, radiation, and curvature, and to show that the scale factor then grows exponentially — the de Sitter solution $a(t)\propto e^{H_\Lambda t}$ with a strictly constant Hubble rate $H_\Lambda = c\sqrt{\Lambda/3}$ . This is not a special case but the generic late-time attractor of any universe with a positive $\Lambda$ , and the same one-line solution powers cosmic inflation in the very early universe. We drop every diluting term, read off the constant right-hand side, and integrate the resulting elementary equation.

Step 1. In equation (3) set $k\to 0$ and $\rho\to 0$ , leaving only the vacuum term: $(\dot a/a)^2 = \Lambda c^2/3$ . Why: matter and radiation fall as $a^{-3}$ and $a^{-4}$ while the vacuum term is fixed, so at late times the fixed term is all that remains — the approximation becomes exact as $a\to\infty$ .

Step 2. Take the positive root for an expanding universe: $\dot a/a = c\sqrt{\Lambda/3} \equiv H_\Lambda$ , a constant. Why: because $\rho_\Lambda$ does not change, neither does the expansion rate — a constant Hubble parameter is the very definition of de Sitter space, in contrast to matter or radiation eras where $H$ falls with time.

Step 3. Integrate the separable equation $\mathrm{d}(\ln a)/\mathrm{d}t = H_\Lambda$ to get $\ln a = H_\Lambda t + \text{const}$ , i.e. $a(t) = a_0\,e^{H_\Lambda(t-t_0)}$ . Why: exponential growth means every proper distance doubles in the same fixed time $\ln 2/H_\Lambda$ no matter how large it already is, so distant galaxies eventually cross a fixed event horizon at $c/H_\Lambda$ and vanish from view.

Go deeper · Derivation C · 1 minThe redshift at which cosmic acceleration begins

Derivation CThe redshift at which cosmic acceleration begins

The goal is to find when a matter-plus- $\Lambda$ universe switches from decelerating to accelerating, expressing the answer as a redshift in terms of the measured density parameters. The transition is not set by hand but follows from the acceleration equation the moment the vacuum's repulsion outmuscles matter's attraction, and it turns out to occur remarkably recently — well within the reach of supernova surveys, which is why the acceleration was detectable at all. We write each density as a function of redshift, impose $\ddot a = 0$ as the turning point, and solve.

Step 1. Scale the components with $1+z = a_0/a$ : matter dilutes as $\rho_m = \rho_{m,0}(1+z)^3$ while the vacuum stays put, $\rho_\Lambda = \rho_{\Lambda,0}$ . Why: these two scalings are the whole engine — a diluting attractor racing against a fixed repeller guarantees a single crossover in the deep past or near future.

Step 2. The acceleration equation (4) with pressureless matter ( $p=0$ ) and the vacuum contributing $-2\rho_\Lambda$ gives $\ddot a > 0 \iff 2\rho_\Lambda > \rho_m$ . Acceleration begins exactly when matter thins to $\rho_m = 2\rho_{\Lambda}$ . Why: the factor of two comes from the $3p$ term — the vacuum's negative pressure counts three times over in the source, so it beats matter while still only half as dense.

Step 3. Solve $\rho_{m,0}(1+z_{\rm acc})^3 = 2\rho_{\Lambda,0}$ for the transition redshift: $1+z_{\rm acc} = (2\Omega_{\Lambda,0}/\Omega_{m,0})^{1/3}$ . With $\Omega_{m,0}=0.31$ and $\Omega_{\Lambda,0}=0.69$ this is $(4.45)^{1/3}=1.64$ , so $z_{\rm acc}\approx0.64$ . Why: the equal-density crossover sits even later, at $1+z_{\rm eq}=(\Omega_{\Lambda,0}/\Omega_{m,0})^{1/3}\approx1.31$ or $z_{\rm eq}\approx0.31$ — both within the last few billion years, the essence of the coincidence problem.

Limiting cases

How the vacuum term stacks up against the other components across cosmic history:

Case	Result	Meaning
Radiation era	$\rho_\Lambda/\rho_r\propto a^{4}\to 0$	Radiation ( $\rho\propto a^{-4}$ ) utterly dominates the constant vacuum; $\Lambda$ is dynamically irrelevant at high redshift.
Matter era	$\rho_\Lambda/\rho_m\propto a^{3}$	Matter dilutes as $a^{-3}$ ; equality $\rho_m=\rho_\Lambda$ occurs at $1+z_{\rm eq}=(\Omega_{\Lambda,0}/\Omega_{m,0})^{1/3}\approx1.31$ , i.e. $z\approx0.3$ .
Late time / de Sitter	$a(t)\propto e^{H_\Lambda t},\ H_\Lambda=c\sqrt{\Lambda/3}$	Once $\Lambda$ dominates, $H$ is constant and expansion is exponential — the asymptotic de Sitter attractor.
Empty universe ( $\rho\to0$ , $\Lambda>0$ )	$\ddot a/a=\Lambda c^2/3>0$	Pure vacuum still accelerates; a positive $\Lambda$ needs no matter at all to drive $\ddot a>0$ .
Einstein static (historical)	$\Lambda_{\rm E}=4\pi G\rho/c^{2}=1/R^{2}$	The fine-tuned balance Einstein sought — unstable to any perturbation and ruled out by the observed expansion.

Common trapA constant energy density means Λ has no gravitational effect

The opposite is true. It is tempting to reason that because $\rho_\Lambda$ never changes, the vacuum is some inert backdrop that does nothing — but a $w=-1$ fluid has both a genuine positive energy density and a large negative pressure, and in general relativity the pressure gravitates. The source term for acceleration is $\rho c^2 + 3p$ , which for the vacuum equals $-2\rho_\Lambda c^2$ : not zero, but strongly negative, producing repulsion. The related trap is to think $\rho_\Lambda \propto a^0$ violates energy conservation, since the total vacuum energy in a comoving volume grows as space expands. It does not: the fluid equation is satisfied precisely because the work done by the negative pressure, $-p\,\mathrm{d}V = +\rho_\Lambda c^2\,\mathrm{d}V$ , exactly supplies the energy of the new volume. Constant density is a consequence of $w=-1$ , not a violation of anything.

Worked examples

1 · The measured value of Λ from the cosmic budget (fully worked)

Given $H_0=70\ \mathrm{km\,s^{-1}\,Mpc^{-1}}$ and $\Omega_{\Lambda,0}=0.69$ , find the vacuum mass density $\rho_\Lambda$ , the cosmological constant $\Lambda$ , and the equivalent energy scale $\rho_\Lambda^{1/4}$ . Then compare with the Planck-scale expectation.

Step 1. Convert the Hubble constant and form the critical density. $H_0 = 70\times10^3/(3.086\times10^{22})\ \mathrm{m\,s^{-1}\,m^{-1}} = 2.27\times10^{-18}\ \mathrm{s^{-1}}$ , so $\rho_{\rm crit,0} = 3H_0^2/(8\pi G) = 3(2.27\times10^{-18})^2/(1.68\times10^{-9}) = \mathbf{9.2\times10^{-27}\ kg\,m^{-3}}$ — a few hydrogen atoms per cubic metre.

Step 2. Scale by $\Omega_{\Lambda,0}$ : $\rho_\Lambda = 0.69\times9.2\times10^{-27} = \mathbf{6.4\times10^{-27}\ kg\,m^{-3}}$ , about four hydrogen atoms per cubic metre. The corresponding energy density is $\rho_\Lambda c^2 = 5.7\times10^{-10}\ \mathrm{J\,m^{-3}}$ .

Step 3. Get $\Lambda$ from equation (6): $\Lambda = 3\Omega_{\Lambda,0}H_0^2/c^2 = 3(0.69)(2.27\times10^{-18})^2/(9.0\times10^{16}) = \mathbf{1.1\times10^{-52}\ m^{-2}}$ . In natural units the energy scale is $\rho_\Lambda^{1/4}\approx\mathbf{2.3\times10^{-3}\ eV}$ — a whisper above zero.

Step 4. Compare with theory. Quantum field theory expects vacuum energy near the Planck density, $\sim M_{\rm Pl}^4$ ; the ratio of that to the observed value is $(M_{\rm Pl}/\rho_\Lambda^{1/4})^4 \approx (1.2\times10^{28}\,\mathrm{eV}/2.3\times10^{-3}\,\mathrm{eV})^4 \approx \mathbf{10^{120}}$ . Even cutting off at the TeV electroweak scale overshoots by some $10^{58}$ . The observed $\Lambda\approx1.1\times10^{-52}\ \mathrm{m^{-2}}$ is real, tiny, and $\sim10^{120}$ below naive expectation — the cosmological constant problem.

2 · Deceleration parameter and the onset of acceleration (fully worked)

For a flat $\Lambda$ CDM universe with $\Omega_{m,0}=0.31$ and $\Omega_{\Lambda,0}=0.69$ , compute the present deceleration parameter, the redshift at which acceleration began, and the redshift of matter– $\Lambda$ equality.

Step 1. Apply equation (8): $q_0 = \tfrac12\Omega_{m,0} - \Omega_{\Lambda,0} = 0.5(0.31) - 0.69 = 0.155 - 0.69 = \mathbf{-0.535}$ . The negative sign is the direct statement that the expansion is accelerating today.

Step 2. Acceleration turned on when $\rho_m = 2\rho_\Lambda$ , i.e. $1+z_{\rm acc} = (2\Omega_{\Lambda,0}/\Omega_{m,0})^{1/3} = (2\times0.69/0.31)^{1/3} = (4.45)^{1/3} = 1.64$ , so $z_{\rm acc}=\mathbf{0.64}$ — roughly six billion years ago.

Step 3. Matter and vacuum reached equal density more recently, at $1+z_{\rm eq} = (\Omega_{\Lambda,0}/\Omega_{m,0})^{1/3} = (2.23)^{1/3} = 1.31$ , so $z_{\rm eq}=\mathbf{0.31}$ . The universe began accelerating at $z\approx0.64$ and vacuum overtook matter at $z\approx0.31$ — both within the last few billion years, near enough to "now" to make the coincidence striking.

3 · The de Sitter future — doubling time and event horizon

In the far future the universe is $\Lambda$ -dominated. With $H_0=70$ km/s/Mpc and $\Omega_{\Lambda,0}=0.69$ , find the asymptotic Hubble rate, the time for all distances to double, and the radius of the de Sitter event horizon.

Step 1. The vacuum-dominated rate is $H_\Lambda = H_0\sqrt{\Omega_{\Lambda,0}} = 70\sqrt{0.69} = \mathbf{58\ km\,s^{-1}\,Mpc^{-1}}$ , so $1/H_\Lambda \approx 16.8$ Gyr. Note this is smaller than $H_0$ because today's expansion is still boosted by the residual matter.

Step 2. Exponential growth doubles every separation in $t_{\rm double} = \ln 2/H_\Lambda = 0.693\times16.8\ \mathrm{Gyr} = \mathbf{11.7\ Gyr}$ , independent of how large the separation already is.

Step 3. The event horizon freezes at $d_{\rm EH} = c/H_\Lambda = (c/H_0)/\sqrt{\Omega_{\Lambda,0}} = 4283/0.83\ \mathrm{Mpc} = \mathbf{5160\ Mpc} \approx 16.8$ billion light-years. Galaxies beyond this radius recede faster than light forever; their light infinitely redshifts, and in the de Sitter future the observable universe empties out.

Physics of the portrait

The cosmological constant is where a symmetry principle, a discarded fudge factor, and a Nobel-winning measurement all meet: one number that Einstein put in by hand, took out in embarrassment, and the sky put back.

Origins · 1 mincosmological constant and how this was discovered

cosmological constant

Einstein's Λ: from static-universe fix to dark energy · relativistic cosmology and the accelerating universe

The cosmological constant is the most famous second thought in physics. In 1917 Einstein applied general relativity to the whole universe and found, to his dismay, that a static cosmos filled with matter is impossible — gravity would collapse it. Committed to a changeless universe, he inserted $\Lambda$ by hand to hold matter in balance, an unstable equilibrium later shown by Eddington to tip at the slightest perturbation. That same year Willem de Sitter found a very different $\Lambda$ solution: an empty universe that nonetheless expands, the first hint that $\Lambda$ drives motion rather than stillness. Alexander Friedmann in 1922 and Georges Lemaître in 1927 derived the expanding solutions and, in Lemaître's case, tied them to the galaxy redshifts, predicting a velocity–distance law. When Edwin Hubble published the expansion in 1929, Einstein abandoned $\Lambda$ ; the widely repeated line, relayed by George Gamow, is that he called it his "greatest blunder." But $\Lambda$ proved impossible to kill. In 1967 Yakov Zel'dovich argued that the quantum vacuum must gravitate exactly like a cosmological constant, which reframed the puzzle as a catastrophic mismatch of scale — the $10^{120}$ problem — rather than a question of whether the term exists. The verdict came in 1998, when the Supernova Cosmology Project led by Saul Perlmutter and the High-z Supernova Search Team led by Brian Schmidt and Adam Riess found distant Type Ia supernovae dimmer than any decelerating model allowed: the expansion was speeding up, and a positive $\Lambda$ with $\Omega_\Lambda\approx0.7$ fit best. The three shared the 2011 Nobel Prize in Physics, and Einstein's blunder became the dominant component of the universe.

History

1917

— Einstein introduces Λ to build a static cosmos; de Sitter finds an empty Λ-driven expanding solution

1927

— Lemaître links Λ and expansion, predicting a velocity–distance relation for galaxies

1967

— Zel'dovich identifies Λ with quantum vacuum energy, exposing the ~10¹²⁰ mismatch

1998

— Type Ia supernovae reveal cosmic acceleration, resurrecting Λ as dark energy

2011

— Perlmutter, Schmidt and Riess share the Nobel Prize for the discovery

Going further

Texts. Ryden, Introduction to Cosmology, 2nd ed., ch. 4–6 (Friedmann equations, single-component solutions, $\Lambda$ ) · Carroll & Ostlie, An Introduction to Modern Astrophysics, 2nd ed., ch. 29 · Dodelson, Modern Cosmology, 2nd ed., ch. 1–2 · Mo, van den Bosch & White, Galaxy Formation and Evolution, ch. 3 · Binney & Tremaine, Galactic Dynamics, 2nd ed., cosmology appendix · Peebles & Ratra, Rev. Mod. Phys. 75, 559 (2003). The key relations to keep in mind are: (1) $\rho_\Lambda c^2=\Lambda c^4/(8\pi G)$ and $w=-1$ from the source-side matching; (2) $\rho\propto a^{-3(1+w)}$ giving $\rho_\Lambda=$ const; (3) the de Sitter solution $a\propto e^{H_\Lambda t}$ with $H_\Lambda=c\sqrt{\Lambda/3}=H_0\sqrt{\Omega_{\Lambda,0}}$ ; (4) $q_0=\tfrac12\Omega_{m,0}-\Omega_{\Lambda,0}$ . For $\Omega_{m,0}=0.31$ , $\Omega_{\Lambda,0}=0.69$ , $H_0=70$ km/s/Mpc the anchor numbers are: $\rho_{\rm crit,0}\approx9.2\times10^{-27}\,\mathrm{kg\,m^{-3}}$ , $\rho_\Lambda\approx6.4\times10^{-27}\,\mathrm{kg\,m^{-3}}$ , $\Lambda\approx1.1\times10^{-52}\,\mathrm{m^{-2}}$ , $\rho_\Lambda^{1/4}\approx2.3\,\mathrm{meV}$ , $q_0\approx-0.53$ , $z_{\rm acc}\approx0.64$ , $z_{\rm eq}\approx0.31$ , $H_\Lambda\approx58$ km/s/Mpc, doubling time $\approx11.7$ Gyr, event horizon $\approx5160$ Mpc. The $\sim10^{120}$ Planck-scale mismatch is the standard order of magnitude for the cosmological constant problem.

Prerequisites. The Friedmann Equations · The Fluid and Acceleration Equations · Equation of State in Cosmology · The FLRW Metric. Leads to. The Expansion of the Universe and the Scale Factor · Cosmological Redshift · Spatial Curvature of the Universe.

Ask the tutor: Ask me to work any $\Omega_\Lambda$ into its $\Lambda$ , deceleration parameter, and acceleration-onset redshift, or to walk through why a $w=-1$ vacuum gravitates repulsively while never diluting.

Is the cosmological constant the same thing as dark energy?

Not quite — the cosmological constant is the simplest candidate for dark energy, not a synonym. "Dark energy" is the general name for whatever is driving the accelerated expansion, defined observationally by its negative pressure. A cosmological constant is the specific hypothesis that this component has exactly $w=-1$ and a density that never changes, equivalent to vacuum energy. Other candidates exist: quintessence is a dynamical scalar field with $w$ close to but not exactly $-1$ and possibly time-varying, and some models modify gravity itself so that no new energy is needed. Current data are consistent with a pure $\Lambda$ , but experiments like DESI and Euclid are searching for any deviation of $w$ from $-1$ that would prove dark energy is something richer.

How can the total vacuum energy grow as the universe expands without breaking energy conservation?

Because the negative pressure does exactly the right amount of work. Apply the first law $\mathrm{d}U = -p\,\mathrm{d}V$ to a comoving volume of vacuum. The energy is $U = \rho_\Lambda c^2 V$ , so as the volume grows $\mathrm{d}U = \rho_\Lambda c^2\,\mathrm{d}V$ since the density is fixed. The pressure term is $-p_\Lambda\,\mathrm{d}V = +\rho_\Lambda c^2\,\mathrm{d}V$ , using $p_\Lambda=-\rho_\Lambda c^2$ . The two sides match identically — the energy of each newly created parcel of space is precisely paid for by the work done by the vacuum's tension. In an expanding universe there is in any case no global conservation of energy in the Newtonian sense, because spacetime itself is dynamical, but the local statement (the fluid equation) is satisfied exactly.

Why did Einstein need Λ for a static universe, and why is that solution rejected?

Einstein's unmodified equations have no static, matter-filled solution: gravity is purely attractive, so a uniform cloud of matter must either expand or collapse. To hold it still he added $\Lambda$ , tuned so its repulsion exactly cancels matter's attraction — the Einstein static universe with $\Lambda_{\rm E}=4\pi G\rho/c^2$ . Two things doom it. First, it is unstable: nudge the density up a hair and gravity wins, collapsing it; nudge it down and $\Lambda$ wins, running it away. There is no way to keep it balanced. Second, and decisively, Hubble showed in 1929 that the universe is not static at all but expanding, so the whole motivation evaporated. The irony is that $\Lambda$ survived — not as a brake to stop expansion, but as an accelerator that speeds it up.

Does the cosmological constant make galaxies and atoms expand too?

No. The acceleration from $\Lambda$ acts only on systems that are not otherwise bound — the vast, near-empty stretches between galaxy clusters, where nothing but the cosmic flow sets the separation. Anything held together by a force stronger than the local vacuum repulsion simply does not participate: atoms are bound by electromagnetism, stars and planets by gravity, and galaxies and clusters by their own collective gravity, each sitting at an equilibrium size set by its internal physics. The vacuum's repulsive acceleration between two points a metre apart is fantastically tiny, swamped by every ordinary force. Only on scales of tens of megaparsecs, between objects with no binding at all, does $\Lambda$ win and drive them apart — which is exactly why a hydrogen atom has the same size today as in the early universe.

What is the cosmological constant problem, in one paragraph?

It is the worst quantitative prediction in physics. Quantum field theory says empty space is not empty — every field has zero-point fluctuations that carry energy, and that vacuum energy should gravitate exactly like a cosmological constant. Estimating its size by summing modes up to the Planck scale gives an energy density about $10^{120}$ times larger than the value we measure; even a conservative cutoff at the electroweak scale overshoots by some $10^{58}$ . So the puzzle is not why $\Lambda$ exists but why it is so extraordinarily, precisely small without being exactly zero — a cancellation to 120 decimal places with a tiny remainder left over. No accepted mechanism explains it; proposed answers range from anthropic selection in a multiverse to new dynamical fields to modifications of gravity, and it remains one of the deepest open problems in fundamental physics.

If dark energy dominates, what is the ultimate fate of the universe?

For a true cosmological constant, the future is a cold, empty de Sitter expansion. Because $\rho_\Lambda$ never dilutes, the Hubble rate asymptotes to a constant $H_\Lambda=c\sqrt{\Lambda/3}$ and every distance grows exponentially, doubling roughly every 12 billion years. Galaxies beyond our de Sitter event horizon — about 16 billion light-years out — recede faster than light and redshift away, so over hundreds of billions of years the observable universe empties until only our own gravitationally bound Local Group remains visible. Structure formation halts, since the accelerating background tears unbound systems apart faster than gravity can assemble them. This is the "Big Freeze." Note the fate hinges on $w$ : if dark energy is not a constant but has $w<-1$ (phantom energy), the acceleration diverges and ends in a "Big Rip" that shreds even bound systems — which is why measuring $w$ precisely is measuring the fate of the cosmos.