Unruh Effect and Horizon Temperature

Empty space is not the same for everyone. The no-particle Minkowski vacuum $|0_M\rangle$ — the state an inertial physicist calls "nothing there" — appears, to an observer moving with constant proper acceleration $a$ , as a warm bath of quanta at the Unruh temperature $T_U=\hbar a/(2\pi c\,k_B)$ . Fulling saw the seed in 1973, Davies sharpened it in 1975, and Unruh gave it operational form in 1976. The result is one of the cleanest lessons quantum field theory teaches: the word particle is not an invariant. Different families of positive-frequency modes define different vacua, and the inertial vacuum is a many-particle — indeed exactly thermal — state with respect to the modes natural to an accelerated frame. The geometry that forces this is the Rindler wedge. A uniformly accelerated worldline is a hyperbola hugging the light cone; signals from beyond the null line $x=t$ can never reach it. That light line is a causal horizon, as impassable to the accelerated observer as a black hole's horizon is to the outside world. Cut off from half of spacetime, the pure state $|0_M\rangle$ , restricted to the accessible right wedge, must be a reduced density matrix obtained by tracing over the modes hidden in the left wedge — and that reduced state is a Gibbs ensemble governed by the boost generator, the Hamiltonian conjugate to proper (Rindler) time, at temperature $T_U$ . The Unruh effect is also the flat-space Rosetta Stone for black-hole thermodynamics. Near any Killing horizon the metric collapses to Rindler form, the surface gravity $\kappa$ plays the role of a redshifted acceleration, and $T_U$ with $a\to\kappa$ becomes the Hawking temperature $\hbar\kappa/(2\pi c\,k_B)$ . Horizon area, entropy, and the four laws of black-hole mechanics descend from one observation: horizons are thermal.

Key takeaways

A uniformly accelerated (Rindler) observer immersed in the Minkowski vacuum detects a thermal bath at the Unruh temperature $T_U=\hbar a/(2\pi c\,k_B)$ , set entirely by the invariant proper acceleration $a$ .
The mechanism is a horizon. The accelerated worldline is confined to a wedge bounded by a null causal horizon; tracing the global vacuum over the modes behind that horizon leaves a thermal density matrix.
Three independent derivations agree: Bogoliubov mixing of Minkowski and Rindler modes (a Planck spectrum), KMS periodicity of the worldline Wightman function in imaginary proper time with period $\beta=2\pi c/a$ , and the detailed-balance response of an Unruh–DeWitt detector.
The effect is the local prototype of Hawking radiation: near a horizon the geometry is Rindler, and $a\to\kappa$ turns $T_U$ into $T_H=\hbar\kappa/(2\pi c\,k_B)$ , which for Schwarzschild is $\hbar c^3/(8\pi GM k_B)$ .
The temperature is astonishingly small: $a=2.47\times10^{20}\,\mathrm{m/s^2}$ — about $2.5\times10^{19}$ Earth gravities — buys only $T_U\approx1\,\mathrm{K}$ , which is why direct thermometry is hopeless and analogue systems carry the experimental load.
Key idea: an accelerated trajectory crosses a causal horizon; tracing the global Minkowski vacuum over the field modes hidden behind that horizon yields a thermal density matrix whose temperature, $T_U=\hbar a/2\pi c\,k_B$ , is fixed by the boost (Rindler-time) Hamiltonian — the flat-space template for Hawking radiation.

Where this lives today

The Unruh effect is not yet measured directly — the accelerations required are twenty orders of magnitude beyond mechanical reach — but it is a live target rather than a curiosity. The oldest candidate signature lives in particle storage rings: the Sokolov–Ternov effect polarizes electrons via spin flips in a bending field, and Bell and Leinaas argued in the 1980s that the residual depolarization carries an Unruh-like thermal imprint from the enormous centripetal acceleration — still debated, because circular motion is not the idealized linear case. Modern proposals are more surgical: accelerated electrons in ultra-intense laser fields, where the Larmor radiation should carry an Unruh signature; superconducting circuit-QED experiments mimicking a rapidly modulated boundary (the sibling dynamical Casimir effect has already been seen this way); and interferometric schemes that read an accelerated qubit's thermal response without waiting for equilibrium. In parallel, Unruh's analogue-gravity program uses sonic horizons in Bose–Einstein condensates and optical fibres to observe the Bogoliubov mode-mixing underlying both Unruh and Hawking radiation, with thermal spectra and cross-horizon entanglement reported in 2016–2019. On the theory side the effect anchors the rigorous Bisognano–Wichmann theorem — that the boost generator is the modular Hamiltonian of the wedge, with the $2\pi$ KMS period baked into axiomatic QFT — now central to entanglement-entropy and modular-flow calculations across holography.

The physical picture

Picture an observer who fires a rocket forever at constant proper acceleration. Their worldline is a hyperbola asymptoting to the 45° light cone. Any signal emitted from beyond the null line $x=t$ closes on that asymptote but never catches the ship: there is a horizon, a boundary between the events the observer can ever learn about and those they cannot — the same causal structure a black-hole horizon imposes on an exterior observer, and the whole engine of the effect. Whenever a horizon partitions a global quantum state, the accessible half is a mixed density matrix, and here that mixed state is precisely thermal.

Figure

The Rindler wedge: a uniformly accelerated observer traces the hyperbola X² − T² = (c²/a)², confined to the right wedge bounded by the null lines X = ±T that form a causal horizon. For a = 2.47×10²⁰ m/s² the closest approach is ρ = c²/a = 0.36 mm and the bath temperature is T_U = 1 K

Why is the loss thermal and not merely noisy? For every field mode in the right wedge there is an entangled partner in the left, the two locked together like the photons of a squeezed pair. The accelerated observer sees only the right half of each pair, and the missing which-partner information is exactly the "ignorance" a thermodynamic entropy measures: the reduced state carries a Boltzmann weight $e^{-\hbar\omega/k_BT_U}$ for each quantum. It is the same entanglement that ties an outgoing Hawking quantum to an infalling partner behind a black-hole horizon. The temperature $T_U=\hbar a/2\pi c\,k_B$ threads together every deep constant: $\hbar$ (it vanishes as $\hbar\to0$ , so it is quantum), $c$ (relativistic), $k_B$ (thermal), while gravity enters only later, via the equivalence principle that lets $a$ become a surface gravity. A temperature has emerged from pure kinematics plus the quantum vacuum. The second figure shows its shape: the Planck spectrum of Rindler quanta the detector clicks out.

Figure

The thermal spectrum of the Minkowski vacuum in the accelerated frame: the mean Rindler occupation N_ω = 1/(exp(2πcω/a) − 1) is an exact Planck distribution at T_U = ħa/2πck_B, plotted against quantum energy in units of k_B·T_U

The core equations

The Unruh effect is one number supported by a small cluster of relations. We start with the temperature itself; everything after either produces this formula or exploits it.

T_U = \frac{\hbar\, a}{2\pi c\, k_B} \qquad (1)

where: $T_U$ — temperature of the bath registered by the accelerated observer; $a$ — invariant proper acceleration of the worldline; $\hbar$ — reduced Planck constant; $c$ — speed of light; $k_B$ — Boltzmann constant. Read it as a conversion, acceleration in and temperature out, with the factor $2\pi$ carrying the geometric content — it is the angular period of Euclidean Rindler time, the conical-smoothness condition. The observer that measures $T_U$ moves on a definite trajectory: a worldline of constant proper acceleration is a hyperbola, parametrized by proper time.

x = \frac{c^2}{a}\cosh\!\Big(\frac{a\tau}{c}\Big),\qquad t = \frac{c}{a}\sinh\!\Big(\frac{a\tau}{c}\Big) \qquad (2)

where: $\tau$ — the observer's proper time; $a$ — constant proper acceleration; $x,t$ — inertial Minkowski coordinates. At $\tau=0$ the observer sits at $x=c^2/a$ at rest, then asymptotes to the light cone as $\tau\to\pm\infty$ ; the hyperbola satisfies $x^2-c^2t^2=(c^2/a)^2$ and never leaves the right wedge. To do physics in the accelerated frame one adopts coordinates comoving with the fiducial observer. The Rindler metric writes the flat wedge in terms of proper distance $\xi$ and proper time $\tau$ .

\mathrm{d}s^2 = -\Big(1+\frac{a\,\xi}{c^2}\Big)^2 c^2\,\mathrm{d}\tau^2 + \mathrm{d}\xi^2 \qquad (3)

where: $\xi$ — proper distance from the fiducial observer at $\xi=0$ ; $\tau$ — that observer's proper time; $a$ — the acceleration at $\xi=0$ . Read it as a static metric with a redshift factor: the local acceleration $a(\xi)=a/(1+a\xi/c^2)$ falls with distance exactly as surface gravity redshifts near a black hole, and $g_{\tau\tau}$ vanishes at $\xi=-c^2/a$ , the Rindler horizon. The operational heart is the particle count. The mean number of Rindler quanta of frequency $\omega$ in the Minkowski vacuum is a Planck distribution — thermal by inspection.

N_\omega = \langle 0_M|\, a^{R\dagger}_\omega a^{R}_\omega\,|0_M\rangle = \frac{1}{e^{2\pi c\omega/a}-1} \qquad (4)

where: $N_\omega$ — mean occupation of the Rindler mode of frequency $\omega$ ; $a^{R}_\omega$ — Rindler annihilation operator; $|0_M\rangle$ — the inertial Minkowski vacuum. Read the exponent: $2\pi c\omega/a=\hbar\omega/k_BT_U$ once constants are restored, so this is a Bose–Einstein occupation at exactly $T_U$ — the inertial vacuum is a thermal state of Rindler particles. The same thermality can be read off a correlator without counting a particle. Along the accelerated worldline the field's two-point function collapses to a pure function of proper-time separation.

G^+(\tau,\tau') = \langle 0_M|\phi(x(\tau))\phi(x(\tau'))|0_M\rangle = -\frac{\hbar}{16\pi^2 c}\,\frac{a^2/c^2}{\sinh^2\!\big[\tfrac{a}{2c}(\tau-\tau') - i\epsilon\big]} \qquad (5)

where: $G^+$ — positive-frequency Wightman function on the trajectory; $\phi$ — a massless scalar field; $\epsilon$ — the pole-ordering prescription. The $\sinh^2$ is periodic under $\tau\to\tau-i\,2\pi c/a$ , and periodicity of correlators in imaginary time is the Kubo–Martin–Schwinger definition of a thermal state at $\beta=2\pi c/a$ . Finally, the bridge to gravity: replacing the acceleration by a horizon's surface gravity turns the Unruh temperature into the Hawking temperature.

T_H = \frac{\hbar\kappa}{2\pi c\, k_B} = \frac{\hbar c^3}{8\pi G M k_B} \qquad (6)

where: $\kappa$ — surface gravity of the Killing horizon; $M$ — mass of a Schwarzschild black hole, for which $\kappa=c^4/4GM$ ; $G$ — Newton's constant. The payoff: the near-horizon geometry of any static black hole is Rindler, $\kappa$ is the redshifted acceleration, and $T_U$ becomes the hole's radiation temperature.

Derivation AThermality from tracing out the left wedge

The goal is to prove that the Minkowski vacuum, restricted to the right Rindler wedge, is not merely mixed but a Gibbs state at the Unruh temperature — that is, to derive $\rho_R=Z^{-1}e^{-2\pi H_R/a}$ from first principles. The strategy: expand the field in modes localized to the left and right wedges, exploit the analyticity of the Minkowski modes across the horizon to write the vacuum as a two-mode squeezed (thermofield-double) state, then trace over the inaccessible left wedge and read off the temperature. We set $c=\hbar=k_B=1$ and treat a massless scalar, restoring factors at the end.

Step 1. Expand the field as $\phi=\sum_\omega\big(a^L_\omega f^L_\omega+a^R_\omega f^R_\omega+\text{h.c.}\big)$ , where $f^L_\omega,f^R_\omega$ are Rindler modes supported in the left and right wedges, positive-frequency with respect to the boost Killing time $\eta$ . Why: the boost generator is the time-translation of the accelerated frame, so its positive-frequency modes are the natural "particles" a Rindler observer counts, split cleanly into the two wedges.

Step 2. Relate these to the Minkowski vacuum via analyticity. Because the genuine Minkowski modes extend smoothly across the horizon, the vacuum is $|0_M\rangle\propto\exp\!\big(\sum_\omega e^{-\pi\omega/a}\,a^{L\dagger}_\omega a^{R\dagger}_\omega\big)|0_L\rangle|0_R\rangle$ . Why: this two-mode squeezed state pairs every left-wedge quantum with a right partner at amplitude $e^{-\pi\omega/a}$ — the vacuum is entangled across the horizon.

Step 3. Trace out the inaccessible left wedge: $\rho_R=\mathrm{Tr}_L\,|0_M\rangle\langle0_M|=(1-e^{-2\pi\omega/a})\sum_n e^{-2\pi n\omega/a}\,|n_R\rangle\langle n_R|$ . Why: summing over the unobservable left occupation numbers destroys all coherence, leaving a density matrix diagonal in the right-wedge number basis — a classical mixture, not a superposition.

Step 4. Identify the temperature. The weights $e^{-2\pi n\omega/a}$ are the Boltzmann factors of a Gibbs ensemble $\rho_R=Z^{-1}e^{-2\pi H_R/a}$ , with $H_R$ the boost Hamiltonian carrying energy $\omega$ per quantum, so $\beta=2\pi/a$ and $T_U=a/2\pi$ , restored as $\hbar a/2\pi c\,k_B$ . Why: a diagonal state with geometric weights in energy is by definition thermal, and the ratio fixes $T$ with no adjustable constant. *Verified: the thermofield-double structure and $T_U=a/2\pi$ are derived in Wald (1994), ch. 5, and Crispino–Higuchi–Matsas, Rev. Mod. Phys. 80, 787 (2008), §III.*

Go deeper · Derivation B · 1 minKMS periodicity of the worldline Wightman function

Derivation BKMS periodicity of the worldline Wightman function

The goal is to obtain the Unruh temperature straight from the analytic structure of the two-point function, using the Kubo–Martin–Schwinger criterion as the operational definition of thermality: a state is thermal at inverse temperature $\beta$ precisely when its correlators are periodic under imaginary time translation, $G^+(\tau-i\beta)=G^-(\tau)$ . We take the flat-space Wightman function, restrict its arguments to the accelerated hyperbola, watch the position dependence collapse into a hyperbolic function of proper-time separation, and read the period. Units $\hbar=c=1$ .

Step 1. Start from the massless-scalar correlator in 4D Minkowski space, $G^+(x,x')=-\frac{1}{4\pi^2}\,\frac{1}{(t-t'-i\epsilon)^2-|\mathbf{x}-\mathbf{x}'|^2}$ . Why: this is the universal short-distance vacuum two-point function; $T_U$ will emerge from how its interval behaves along an accelerated path.

Step 2. Insert the hyperbolic trajectory (2); the invariant interval becomes $(t-t')^2-(x-x')^2=-\frac{4}{a^2}\sinh^2\!\big(\frac{a\,\Delta\tau}{2}\big)$ , a function of $\Delta\tau=\tau-\tau'$ alone. Why: uniform acceleration means time-translation invariance in proper time — the correlator depends only on the separation, and the hyperbola turns the flat interval into a $\sinh^2$ .

Step 3. Substitute to get $G^+(\Delta\tau)=-\frac{a^2}{16\pi^2}\,\frac{1}{\sinh^2[\frac{a}{2}(\Delta\tau-i\epsilon)]}$ . Why: all position dependence has cancelled into a single stationary function of proper-time separation — exactly what a worldline detector is sensitive to.

Step 4. Shift imaginary time by $\beta=2\pi/a$ : since $\sinh[\frac{a}{2}\Delta\tau-i\pi]=-\sinh[\frac{a}{2}\Delta\tau]$ , the square is unchanged, so $G^+(\Delta\tau-i\,2\pi/a)=G^+(\Delta\tau)$ . Why: this is the KMS periodicity, and the period $\beta=2\pi/a$ is the inverse temperature — hence $T_U=a/2\pi$ , with no reference to particles at all. *Verified: Birrell & Davies (1982), §3.3 — the $\sinh^2$ form and its imaginary period $2\pi/a$ are standard.*

Go deeper · Derivation C · 1 minThe Unruh-DeWitt detector response is Planckian

Derivation CThe Unruh-DeWitt detector response is Planckian

The goal is to show that a concrete particle detector — a two-level system with energy gap $\mathcal{E}$ carried along the accelerated worldline and weakly coupled to the field — actually clicks with a thermal excitation rate, so that "the observer sees a thermal bath" is a statement about measurable transition probabilities, not just mode bookkeeping. First-order perturbation theory expresses the excitation rate as the Fourier transform of the worldline Wightman function; we insert the $\sinh^2$ correlator, close the contour, and sum the residues. Units $\hbar=c=1$ .

Step 1. Write the excitation rate per unit proper time as $\mathcal{F}(\mathcal{E})=\int_{-\infty}^{\infty}\mathrm{d}s\,e^{-i\mathcal{E}s}\,G^+(s)$ , with $s=\Delta\tau$ . Why: to lowest order the up-transition probability factorizes into the coupling squared times this response function, the power spectrum of the field fluctuations the detector rides through.

Step 2. Insert the correlator: $\mathcal{F}(\mathcal{E})=-\frac{a^2}{16\pi^2}\int\mathrm{d}s\,\frac{e^{-i\mathcal{E}s}}{\sinh^2[\frac{a}{2}(s-i\epsilon)]}$ , whose integrand has double poles at $s=i\,2\pi n/a$ , the $i\epsilon$ pushing the contour below the real axis. Why: the periodic imaginary-time poles from the $\sinh^2$ manufacture a geometric series of Boltzmann factors when the contour is closed.

Step 3. Close in the lower half-plane and sum residues at $s=-i\,2\pi n/a$ , $n\ge1$ : $\sum_{n\ge1}e^{-2\pi n\mathcal{E}/a}=e^{-2\pi\mathcal{E}/a}/(1-e^{-2\pi\mathcal{E}/a})$ . Why: each pole contributes a Boltzmann weight raised to its order in the tower, resumming to the Bose–Einstein denominator.

Step 4. Collect the result: $\mathcal{F}(\mathcal{E})=\frac{\mathcal{E}}{2\pi}\,\frac{1}{e^{2\pi\mathcal{E}/a}-1}$ , with detailed-balance ratio $\mathcal{F}(\mathcal{E})/\mathcal{F}(-\mathcal{E})=e^{-2\pi\mathcal{E}/a}=e^{-\mathcal{E}/k_BT_U}$ . Why: excitation suppressed relative to de-excitation by exactly the Boltzmann factor at $T_U$ is the defining signature of equilibrium with a thermal bath — the detector genuinely thermalizes. *Verified: the contour evaluation reproduces Crispino–Higuchi–Matsas, Rev. Mod. Phys. 80, 787 (2008), Eq. (3.42), and Birrell & Davies (1982), §4.5.*

Limiting cases

In its limits the formula $T_U=\hbar a/2\pi c\,k_B$ carries both the statement that the effect is real and the ones that keep it from contradicting ordinary experience.

Case	Result	Meaning
Classical limit	$\hbar\to0\;\Rightarrow\;T_U\to0$	The effect is intrinsically quantum; a classical field under acceleration shows no bath.
Inertial limit	$a\to0\;\Rightarrow\;T_U\to0$	An unaccelerated observer has no horizon and sees the genuinely empty Minkowski vacuum.
Black-hole correspondence	$a\to\kappa\;\Rightarrow\;T_U\to T_H=\dfrac{\hbar\kappa}{2\pi c\,k_B}$	With surface gravity in place of acceleration, Unruh becomes Hawking.
Numerical scale	$a=2.47\times10^{20}\,\mathrm{m/s^2}\;\Rightarrow\;T_U\approx1\,\mathrm{K}$	Needs $2.5\times10^{19}$ Earth gravities; the tiny scale is why direct detection is so hard.
High-frequency modes	$\hbar\omega\gg k_BT_U\;\Rightarrow\;N_\omega\to e^{-\hbar\omega/k_BT_U}$	Far above the thermal peak the spectrum is Boltzmann-suppressed, as any thermal bath is.

Common trapthe misconception

The Unruh effect does not say an inertial observer sees any radiation. In the Minkowski frame the vacuum is empty, the particle number is zero, and the renormalized stress-energy tensor vanishes identically — thermality is strictly observer-relative, a consequence of which modes you call "positive frequency," not a property of the state alone. Three further traps: (i) the $a$ in the formula is the invariant proper acceleration, frame-independent even though the effect is observer-dependent. (ii) A detector clicking in the bath gets no free energy — it does work against the accelerating agent and radiates real quanta into Minkowski space; the agent pays. (iii) "Seeing a thermal bath" is a rate, not "equilibrium forever": maintaining it requires maintaining the acceleration.

Worked examples

1 · The acceleration behind a one-kelvin bath (confidence builder)

What proper acceleration produces an Unruh temperature of exactly $T_U=1\,\mathrm{K}$ , and what does the answer tell us about detecting the effect?

Step 1. Invert equation (1) for the acceleration: $a=\dfrac{2\pi c\,k_B\,T_U}{\hbar}$ . This is the whole physics; the rest is arithmetic.

Step 2. Insert numbers with $T_U=1\,\mathrm{K}$ : $a=\dfrac{2\pi(2.998\times10^{8})(1.381\times10^{-23})(1)}{1.055\times10^{-34}}=\mathbf{2.47\times10^{20}\ m/s^2}$ . Compared to Earth's gravity this is $a/g=2.47\times10^{20}/9.81=\mathbf{2.5\times10^{19}\ g}$ .

Step 3. Sanity-check the geometry: the closest approach to the horizon is $\rho=c^2/a=(2.998\times10^{8})^2/2.47\times10^{20}=\mathbf{0.36\ mm}$ . A bath barely "one degree above absolute zero" demands an acceleration that carries an object from rest to light speed across a fraction of a millimetre.

Step 4. Draw the lesson: no sustained mechanical process reaches $10^{20}\,\mathrm{m/s^2}$ . This is why the Unruh effect has never been thermometrically measured, and why the field relies on analogue systems and indirect signatures such as storage-ring spin depolarization.

2 · From Unruh to Hawking: the temperature of real black holes

Use the $a\to\kappa$ correspondence to compute the Hawking temperature of a solar-mass black hole and of the supermassive hole Sgr A\ ( $M=4.3\times10^{6}\,M_\odot$ ), and comment on the trend.*

Step 1. The surface gravity of a Schwarzschild hole is $\kappa=c^4/4GM$ . For $M=M_\odot=1.99\times10^{30}\,\mathrm{kg}$ : $\kappa=\dfrac{(2.998\times10^{8})^4}{4(6.674\times10^{-11})(1.99\times10^{30})}=\mathbf{1.52\times10^{13}\ m/s^2}$ — a genuinely enormous "acceleration," yet far below the $10^{20}$ of Example 1.

Step 2. Feed it through the Unruh formula with $a\to\kappa$ : $T_H=\dfrac{\hbar\kappa}{2\pi c\,k_B}=\dfrac{1.52\times10^{13}}{2.47\times10^{20}}\,\mathrm{K}=\mathbf{6.2\times10^{-8}\ K}$ . Equivalently, using the closed form (6), $T_H=\hbar c^3/8\pi GM_\odot k_B$ gives the same $6.2\times10^{-8}\,\mathrm{K}$ — sixty nanokelvin, far colder than the CMB.

Step 3. Scale to Sgr A\*. Because $T_H\propto1/M$ , the temperature of a hole $4.3\times10^{6}$ times heavier is $4.3\times10^{6}$ times colder: $T_H=6.2\times10^{-8}/4.3\times10^{6}=\mathbf{1.4\times10^{-14}\ K}$ .

Step 4. Read the trend: bigger holes are colder — a stellar hole is a hundred-millionth of a kelvin, a galactic-centre monster ten femtokelvin. Since $T_H\propto1/M$ , astrophysical black holes are thermodynamically almost perfectly cold; evaporation matters only for hypothetical primordial holes of asteroid mass or below.

3 · An electron in an intense field

Estimate the Unruh temperature of an electron subjected to a static field $E=10^{13}\,\mathrm{V/m}$ , of the order reachable at the focus of a petawatt laser.

Step 1. The proper acceleration is $a=eE/m_e=(1.602\times10^{-19})(10^{13})/(9.109\times10^{-31})=\mathbf{1.76\times10^{24}\ m/s^2}$ .

Step 2. Then $T_U=\hbar a/2\pi c\,k_B=1.76\times10^{24}/2.47\times10^{20}\approx\mathbf{7.1\times10^{3}\ K}$ . A few thousand kelvin from one of the strongest fields we can make — which is why detection proposals chase resonant and interferometric signatures rather than a literal temperature reading.

Physics of the portrait

The Unruh effect carries three names on its founding papers, but took its cleanest form in a single 1976 article whose central device — an idealized particle detector on a worldline — reframed what "a particle" means.

Origins · 1 minWilliam Unruh and how this was discovered

William Unruh

b. 1945 · Canadian theoretical physicist · quantum field theory in curved and accelerated spacetimes

William George Unruh trained under John Wheeler at Princeton in the early 1970s, when Bekenstein's black-hole entropy and Hawking's radiation were turning geometry into thermodynamics. His 1976 paper "Notes on black-hole evaporation" (Phys. Rev. D 14, 870) did something deceptively simple: instead of arguing about field modes, it asked what an actual device — a two-level "particle detector" — would register while accelerating through empty space, and showed the answer is a thermal spectrum at $T=\hbar a/2\pi c\,k_B$ , making the observer-dependence of particle content operational and inescapable. The seed was Stephen Fulling's (1973), who found Rindler quantization unitarily inequivalent to the Minkowski one, sharpened by Paul Davies (1975), who tied the thermality to horizons; Unruh's detector fused these into a physical prediction and the template for Hawking radiation. He is also a founder of analogue gravity — his 1981 "dumb hole," a sonic horizon in flowing fluid, launched the program that now probes horizon physics in Bose–Einstein condensates where the literal Unruh temperature is unreachable but the mechanism is not. He remains at the University of British Columbia, a clarifying voice on the foundations of quantum theory and gravity.

History

1973

— Fulling shows Rindler quantization is unitarily inequivalent to Minkowski quantization

1974

— Hawking predicts that black holes radiate thermally

1975

— Davies links the accelerated-frame thermality to the presence of a horizon

1976

— Unruh introduces the particle detector and derives T = ħa/2πck_B

2008

— Crispino, Higuchi and Matsas publish the definitive review of the effect and its applications

Going further

Texts. Wald, General Relativity (1984), §14.4 — Unruh effect and Bogoliubov coefficients · Wald, QFT in Curved Spacetime and Black Hole Thermodynamics (1994), ch. 5 — thermofield-double and modular structure · Birrell & Davies, Quantum Fields in Curved Space (1982), §3.3, §4.5 — the $\sinh^2$ correlator and detector rate · Carroll, Spacetime and Geometry (2004), §9.5 · Crispino, Higuchi & Matsas, Rev. Mod. Phys. 80, 787 (2008) · Unruh, Phys. Rev. D 14, 870 (1976).

Provenance. The Unruh temperature, the Rindler worldline and metric, the Planck occupation, the KMS period $2\pi c/a$ , the thermofield-double reduced density matrix, the detector detailed balance, and the $a\to\kappa$ correspondence are all standard. A verifier should: (i) reproduce the $\sinh^2$ worldline Wightman function and its imaginary period $2\pi/a$ (Birrell & Davies §3.3); (ii) check the Bogoliubov ratio $|\beta_\omega/\alpha_\omega|^2=e^{-2\pi c\omega/a}$ and its Bose–Einstein occupation (Wald 1984 §14.4); (iii) confirm the detector rate by contour integration (Crispino–Higuchi–Matsas Eq. 3.42); (iv) verify $a\approx2.47\times10^{20}\,\mathrm{m/s^2}$ for $T_U=1\,\mathrm{K}$ , $\rho=c^2/a=0.36\,\mathrm{mm}$ , and $T_H=6.2\times10^{-8}\,\mathrm{K}$ at one solar mass.

Prerequisites. Killing Horizons and Surface Gravity · Schwarzschild Black Hole Structure · Event Horizon and Apparent Horizon. Leads to. Four Laws of Black Hole Mechanics · Hawking Radiation · Trapped Surfaces and Closed Trapped Regions.

Ask the tutor: Ask me to walk through any of the three derivations line by line, or to show exactly how the near-horizon Schwarzschild metric becomes Rindler space so that $a\to\kappa$ turns Unruh into Hawking.

Is the Unruh effect real, or just a formal artifact of choosing weird coordinates?

It is real in the only sense that matters operationally: a physical detector carried along the accelerated worldline genuinely clicks, with excitation and de-excitation rates in the exact Boltzmann ratio $e^{-\mathcal{E}/k_BT_U}$ — a statement about measurable probabilities, not coordinates. What is observer-dependent is the word "particle": inertial and accelerated observers disagree about how many quanta the same state contains, because they define positive frequency differently, and both are right in their own framework. The invariant quantities — the renormalized stress tensor, the correlators — are unambiguous; only the particle count is frame-relative, and the Unruh effect is the clean demonstration of that fact.

How can the vacuum be empty for one observer and thermal for another at the same time?

Because "number of particles" is not invariant under the change from inertial to accelerated frames. A particle is a quantum of a positive-frequency mode, and positive frequency is defined relative to some notion of time. The inertial observer uses Minkowski time and finds the vacuum annihilated by every annihilation operator — zero particles. The accelerated observer uses Rindler time; those modes relate to the Minkowski ones by a Bogoliubov transformation mixing creation and annihilation operators, so the same state teems with Rindler quanta. No contradiction: the two count genuinely different things, and neither is "the" true count.

Where does the energy of the detected quanta actually come from?

From the agent doing the accelerating. A detector excited in the Unruh bath has absorbed energy, and by conservation it must come from somewhere: the detector also radiates real quanta into Minkowski space, and the reaction force means the engine maintaining its acceleration does extra work. Nothing is created from nothing — the vacuum is not a free energy source. The thermal character describes the statistics of the excitations; the energy budget is paid by whatever keeps the worldline accelerating.

Why exactly does the factor 2π appear in the temperature?

It is the angular period of Euclidean time. Continue Rindler time to imaginary values and the $(\rho,\theta)$ plane near the horizon becomes ordinary flat space in polar coordinates; flatness — no conical singularity at the origin — requires the angle to have period exactly $2\pi$ . Back in real proper time this is an imaginary period $\beta=2\pi c/a$ , and $\beta$ is the inverse temperature. So the $2\pi$ is the smoothness condition of the horizon itself, which is why the identical factor appears in the Hawking temperature, enforcing regularity of the Euclidean black hole.

Is the Unruh temperature the same thing as the Hawking temperature?

They are the same formula wearing two hats. Unruh is $\hbar a/2\pi c\,k_B$ for an accelerated observer in flat space; Hawking is $\hbar\kappa/2\pi c\,k_B$ for a black hole, with $\kappa$ the surface gravity. The bridge is that the near-horizon region of any static black hole looks, to a hovering observer, exactly like Rindler space, and the acceleration that observer sustains — redshifted to infinity — is $\kappa$ . So Hawking radiation is the Unruh effect for a static observer just outside a real horizon. Not merely an analogy: it is why the same $2\pi$ and the same Planck spectrum govern both.

Can we ever hope to measure it directly in the lab?

As a thermometer reading, almost certainly not with steady mechanical acceleration: one kelvin needs $2.5\times10^{19}$ g. The realistic paths are indirect. Electrons in ultra-intense laser fields reach $10^{24}\,\mathrm{m/s^2}$ and thousands of kelvin, where modifications to their radiation could betray the effect; superconducting circuits with modulated boundaries have already shown the sibling dynamical Casimir effect. Complementing these, analogue-gravity experiments in Bose–Einstein condensates and optical fibres directly observe the Bogoliubov mode-mixing that is the mathematical core of both Unruh and Hawking radiation.