Relativistic Jets and Launching Mechanisms

Relativistic jets are the most far-reaching structures black holes make. From a launching region only a few gravitational radii across ( $r_g = GM/c^2$ ), a narrow beam of magnetized plasma stays coherent out to megaparsecs in giant radio galaxies — nine orders of magnitude in scale. The same phenomenon appears at every black-hole mass: stellar holes in X-ray binaries flicker jets on and off with accretion state, collapsing stellar cores fire gamma-ray-burst jets, and supermassive holes drive the quasar and blazar zoo. The puzzle that unites them is stark: how does a hot, chaotic, inflowing disk produce an outflow collimated to a few degrees, moving at Lorentz factors of tens, and almost free of baryons? The answer, settled over the past half-century, is that jets are launched electromagnetically: energy leaves the engine not as hot gas but as Poynting flux — stored in a large-scale, ordered magnetic field — and only converts to bulk kinetic energy far downstream. Two mechanisms, both from Roger Blandford's group, define the field. Blandford–Znajek taps the black hole's own rotation: field lines threading the spinning horizon are wound up by frame-dragging in the ergosphere, driving a Poynting outflow like a spinning magnet acting as a dynamo. Blandford–Payne instead flings disk gas outward along inclined field lines like beads on rotating wires. General-relativistic magnetohydrodynamic (GRMHD) simulations added the decisive piece — the Magnetically Arrested Disk, in which accumulated flux chokes the inflow and pushes jet efficiency past 100% of the accreted rest-mass energy, the fingerprint that the hole's spin is being drained. Jet launching thus links black-hole spacetime to galaxy-scale feedback, the M–sigma relation, and the blazar sky.

Key takeaways

A relativistic jet is a collimated, magnetically driven outflow launched near an accreting black hole; blazar jets reach $\Gamma \sim 5$ – $50$ and stay coherent from $r_g$ to megaparsecs.
Blandford–Znajek extracts the black hole's spin energy through field lines threading the horizon, powered by ergospheric frame-dragging; the power scales as $P_{\rm BZ} \propto \Phi_{\rm BH}^2\,\Omega_H^2$ .
Blandford–Payne extracts the disk's rotational energy, centrifugally slinging gas along field lines inclined more than $30^\circ$ from the rotation axis.
Both deliver energy as Poynting flux first; the jet starts magnetically dominated ( $\sigma \gg 1$ , slow) and accelerates to $\Gamma_\infty \approx \sigma_0$ only as the field energy converts to motion.
In the Magnetically Arrested Disk state the horizon flux saturates and jet efficiency can exceed $100\%$ of $\dot{M}c^2$ — direct proof that spin, not accretion, powers the jet; beamed toward us the jet becomes a blazar, its flux boosted by $\delta^{3+\alpha}$ with apparent superluminal motion up to tens of $c$ .
Key idea: a jet is a Poynting-flux-dominated outflow — energy stored in an ordered magnetic field threading either the spinning horizon (Blandford–Znajek) or the rotating disk (Blandford–Payne) is extracted electromagnetically; rotation twists the field into a toroidal component that carries energy and angular momentum outward and, through hoop stress, collimates the flow into a beam.

The core insight

A jet is a Poynting-flux-dominated outflow: energy is stored in a large-scale, ordered magnetic field threading either the spinning horizon (Blandford–Znajek) or the rotating disk (Blandford–Payne) and is extracted electromagnetically. Rotation twists the field lines, the resulting toroidal component carries angular momentum and energy outward as an electromagnetic torque, and hoop stress from that toroidal field collimates the flow into a narrow beam.

Where this lives today

Jet launching is now a precision, multi-messenger subject. The Event Horizon Telescope resolved the shadow of M87* in 2019 and, in 2021, mapped an ordered, spiral horizon-scale field — exactly the geometry a magnetically arrested, Blandford–Znajek jet requires — while decades of VLBI track the jet's parabolic-to-conical collimation out to the Bondi radius. Global GRMHD codes (H-AMR, BHAC, KORAL) run three-dimensional, radiative, two-temperature models that reproduce over-unity efficiencies and feed EHT image libraries to constrain spin. The physics scales down: long gamma-ray bursts are modeled as Blandford–Znajek jets from newborn spinning stellar holes; tidal disruption events like Swift J1644+57 launched jets in real time; and X-ray binaries switch jets on and off across the hard-to-soft transition within weeks. Open problems stay sharp — how jets resist kink instabilities over megaparsecs, where the magnetization drops to power the radiation, how much flux builds a MAD, and what loads the baryons that set $\sigma_0$ . Each is a launching question in disguise.

The physical picture

Picture the field lines as stiff wires anchored in the event horizon or the disk. As the anchor rotates, each wire winds up, growing a toroidal (azimuthal) component that does two jobs at once: its tension along the axis pinches the flow inward — hoop-stress collimation — while its outward Poynting flux carries energy and angular momentum away. It is the torque of an electric motor run in reverse, as a generator: rotation of the anchor becomes an outgoing electromagnetic energy flux. The first figure shows the geometry and the governing relation.

Figure

Magnetic launching of a jet from a spinning black hole: poloidal flux Φ_BH threading the horizon is wound up by frame-dragging at rate Ω_H into a toroidal field whose hoop stress collimates the flow, while the Poynting power carried outward scales as P_BZ ∝ Φ_BH² Ω_H²

The two mechanisms differ only in where the anchor sits. In Blandford–Payne the wires root in the disk, the rotation is Keplerian, and disk plasma is flung out once a field line leans far enough from vertical. In Blandford–Znajek the wires thread the horizon, the rotation is $\Omega_H$ of the spacetime, and the process works with no accreting gas at all, draining the hole's rotation through the vacuum field. Which dominates is set by how much ordered flux the flow can pin near the hole; when that flux saturates, the disk becomes magnetically arrested and the jet over-efficient. The second figure is the punchline the simulations deliver.

Figure

Jet efficiency η = P_jet/⟨Ṁ⟩c² versus black-hole spin a in the magnetically arrested state: because η ∝ Ω_H²(a) the curve climbs steeply, crosses 100% near a ≈ 0.93, and reaches η ≈ 1.4 at a ≈ 0.98 — far above the maximum accretion efficiency η_acc = 0.42, so the surplus must come from the hole's spin (schematic, GRMHD-calibrated to Tchekhovskoy, Narayan and McKinney 2011)

The core equations

A small set of relations connects spin, flux, magnetization, and boosting. The central one is the Blandford–Znajek power: a spinning, threaded horizon radiates energy at a rate fixed by the flux it carries and how fast it turns.

P_{\rm BZ} = \frac{\kappa}{4\pi c}\,\Phi_{\rm BH}^2\,\Omega_H^2\,f(\Omega_H) \qquad (1)

where: $P_{\rm BZ}$ — power extracted from the hole; $\Phi_{\rm BH}$ — poloidal flux through one horizon hemisphere; $\Omega_H$ — horizon angular velocity; $\kappa \approx 0.053$ — geometric constant (split-monopole); $f(\Omega_H) \approx 1$ — higher-spin correction. Read it as: power grows with the square of both flux and spin, so a jet needs a rapidly rotating hole and a strong ordered field. The horizon's angular velocity is a pure spacetime quantity, set by the mass and spin of the Kerr solution; it vanishes for a non-rotating hole and saturates near maximal spin.

\Omega_H = \frac{a\,c^3}{2\,G M}\,\frac{1}{1+\sqrt{1-a^2}} \qquad (2)

where: $a = Jc/(GM^2)$ — dimensionless spin, $0 \le a \le 1$ ; $M$ — mass; $J$ — angular momentum. For $a \ll 1$ , $\Omega_H \approx a c/(2 r_g)$ so $P_{\rm BZ} \propto a^2$ ; as $a \to 1$ , $\Omega_H \to c/(2 r_g)$ — a Schwarzschild hole drives no BZ jet. To compare jets across many decades of mass, power is written as an efficiency — jet power per unit rest-mass energy supplied by accretion.

\eta_{\rm jet} \equiv \frac{P_{\rm jet}}{\langle\dot{M}\rangle\,c^2} \qquad (3)

where: $\eta_{\rm jet}$ — jet efficiency; $\langle\dot{M}\rangle$ — time-averaged accretion rate. Read it as a budget: accretion caps radiative efficiency at $\eta_{\rm acc} \le 0.42$ (extreme Kerr), so any $\eta_{\rm jet} > 1$ is energy the disk cannot supply — it is spin. The efficiency saturates because the flux does. Once enough poloidal flux piles up, its pressure pushes back on the inflow and the disk becomes magnetically arrested at a near-universal flux.

\Phi_{\rm MAD} = \phi_{\rm MAD}\,\sqrt{\dot{M}\,r_g^2\,c}\,,\qquad \phi_{\rm MAD}\approx 50 \qquad (4)

where: $\phi_{\rm MAD} \approx 50$ — dimensionless saturation flux (Gaussian-cgs GRMHD units); $r_g = GM/c^2$ . The flux the hole can hold is set by $\dot{M}$ , which is why $P_{\rm BZ} \propto \Phi^2 \propto \dot{M}$ in the arrested state and efficiency becomes a function of spin alone. Whether the jet is a battery or a bullet is set by the magnetization — field energy over rest-mass energy.

\sigma \equiv \frac{B^2}{4\pi\,\rho\,c^2} = \frac{\text{Poynting flux}}{\text{rest-mass flux}} \qquad (5)

where: $\sigma$ — magnetization; $B$ — comoving field; $\rho$ — comoving rest-mass density. At the base $\sigma \gg 1$ (magnetically dominated, slow); acceleration is the trade of $\sigma$ for bulk motion. That trade is governed by a quantity conserved along each field line — total energy per rest-mass flux — which fixes the terminal speed.

\mu = \Gamma\,(1+\sigma) = \text{const} \;\;\Longrightarrow\;\; \Gamma_\infty \approx \mu \approx \sigma_0 \qquad (6)

where: $\mu$ — conserved energy-to-mass flux; $\Gamma$ — local Lorentz factor; $\Gamma_\infty$ — terminal Lorentz factor; $\sigma_0$ — base magnetization. Read it as a ceiling: the jet reaches at most $\Gamma_\infty = \mu$ , and only if the field energy fully converts ( $\sigma \to 0$ ). Finally, what we observe is not $\Gamma$ but the Doppler-boosted appearance, set by the beaming factor for a flow of speed $\beta$ seen at $\theta$ .

\delta = \frac{1}{\Gamma\,(1-\beta\cos\theta)} \qquad (7)

where: $\delta$ — Doppler factor; $\beta = v/c$ ; $\theta$ — angle to the line of sight. Head-on, $\delta \approx 2\Gamma$ ; at $\cos\theta = \beta$ ( $\theta \approx 1/\Gamma$ ), $\delta = \Gamma$ . Observed flux scales as $\delta^{3+\alpha}$ , so the appearance is dominated by $\delta$ , not $\Gamma$ .

Derivation AMagnetocentrifugal launching — the bead-on-a-wire criterion

The goal is to find the geometric condition under which a cold parcel of disk gas, frozen to a rigidly rotating field line, is flung outward with no energy barrier — the Blandford–Payne criterion. We treat the field line as a stiff wire corotating at the Keplerian rate, write the effective potential a bead feels along it, and ask when the footpoint is a potential maximum rather than a minimum, so the slightest outward nudge slides the bead to infinity. The answer is a single critical inclination angle.

Step 1. In the frame corotating at the footpoint rate $\Omega_0$ , a unit-mass parcel feels gravity plus a centrifugal potential, $\Phi_{\rm eff}(r,z) = -\dfrac{GM}{\sqrt{r^2+z^2}} - \tfrac{1}{2}\Omega_0^2 r^2$ , with $\Omega_0^2 = GM/r_0^3$ (Keplerian footpoint at $r_0$ ). Why: corotation makes the field "rigid," and centrifugal support is the whole point, so it must live in the potential.

Step 2. Parametrize a straight field line by arc length $s$ from the footpoint, inclined at angle $\theta$ above the disk plane: $r = r_0 + s\cos\theta$ , $z = s\sin\theta$ . At the footpoint the net force along the wire vanishes, $\mathrm{d}\Phi_{\rm eff}/\mathrm{d}s\big|_0 = 0$ . Why: the launch is decided not by the slope (zero) but by the curvature — whether the footpoint is a hilltop or a valley.

Step 3. Expand to second order, $\Phi_{\rm eff}(s) \approx \Phi_{\rm eff}(0) + \tfrac{1}{2}\Phi''(0)\,s^2$ . Setting $GM=r_0=1$ , differentiation gives $\Phi''(0) = -3\cos^2\theta + \sin^2\theta = 1 - 4\cos^2\theta$ . Why: steep gravitational curvature ( $-3$ ) along the radial direction and gentle centrifugal curvature combine into one expression whose sign is all that matters.

Step 4. Launching requires $\Phi''(0) < 0$ , i.e. $1 - 4\cos^2\theta < 0 \Rightarrow \cos^2\theta > 1/4 \Rightarrow \theta < 60^\circ$ from the disk plane — equivalently, more than $30^\circ$ from the rotation axis. Why: below $60^\circ$ the centrifugal barrier vanishes, so cold gas accelerates from rest with no thermal push, which is why jets are cold and fast rather than hot and slow. *The marginal inclination is $\arccos(1/2)=60^\circ$ from the disk plane (30° from the axis), confirmed by the sign change in $\Phi''(0)=1-4\cos^2\theta$ . Beyond the Alfvén point the flow decouples from rigid corotation and is collimated by toroidal hoop stress.*

Go deeper · Derivation B · 1 minBlandford–Znajek power from horizon impedance matching

Derivation BBlandford–Znajek power from horizon impedance matching

The goal is to recover the $P_{\rm BZ} \propto \Phi^2\Omega_H^2$ scaling from a circuit picture, treating the rotating horizon as a battery of finite internal resistance driving a load. The membrane paradigm gives the horizon a real surface resistivity; frame-dragging sets up an EMF, and the power delivered to the load (the jet) is maximized by impedance matching, as for any generator. This turns a general-relativistic result into Ohm's law and explains why power is quadratic in flux and spin.

Step 1. Frame-dragging rotates the field lines with the horizon, generating a unipolar-inductor EMF, $\mathcal{E} \sim \Omega_H \Phi_{\rm BH}/c$ — the physics of a Faraday disc. Why: relative rotation between the field's anchor and infinity does the work; $\Omega_H$ and $\Phi$ enter linearly, so their product sets the driving voltage.

Step 2. Assign the horizon its membrane surface impedance $Z_H = 4\pi/c \approx 377\,\Omega$ , the impedance of free space — the "internal resistance" of the battery. Why: the horizon behaves as a resistive membrane in the 3+1 split; without finite impedance there is no dissipation and no steady power transfer.

Step 3. Maximum power is delivered when the external field geometry matches this impedance: $P \sim \dfrac{\mathcal{E}^2}{Z_H} \sim \dfrac{c}{4\pi}\left(\dfrac{\Omega_H \Phi_{\rm BH}}{c}\right)^2$ . Why: this is the maximum-power-transfer theorem; the factor $c/4\pi$ becomes $\kappa \approx 0.053$ once the split-monopole geometry is worked out.

Step 4. Collecting factors gives $P_{\rm BZ} = \dfrac{\kappa}{4\pi c}\,\Phi_{\rm BH}^2\,\Omega_H^2$ . Why: the two squares are distinct — $\Phi^2$ because power is voltage-squared over a fixed impedance, $\Omega_H^2$ because the EMF is itself proportional to the spin rate. *The impedance-matching estimate reproduces the $\Phi^2\Omega_H^2$ scaling and the $\kappa\approx0.053$ geometric prefactor of the split-monopole result.*

Go deeper · Derivation C · 1 minPoynting-to-kinetic conversion and the terminal Lorentz factor

Derivation CPoynting-to-kinetic conversion and the terminal Lorentz factor

The goal is to show why a jet carrying enormous energy starts out moving slowly, and what ultimate speed it reaches. For a steady, ideal, axisymmetric magnetized outflow the total energy flux per rest-mass flux is conserved along each field line; split between kinetic energy and Poynting flux, this invariant forces the terminal Lorentz factor to equal the launch-point magnetization once the field energy is fully mined. We track it from base to infinity.

Step 1. The relativistic Bernoulli invariant for a cold magnetized wind is $\mu = \Gamma(1+\sigma) = \text{const}$ along a field line. Why: energy is conserved in ideal MHD, so what the jet gains in $\Gamma$ it loses in $\sigma$ — acceleration is literally magnetic energy being spent.

Step 2. At the base the flow is nearly static and field-dominated, $\Gamma_0 \approx 1$ , $\sigma_0 \gg 1$ , so $\mu \approx \sigma_0$ . Why: this resolves the apparent paradox — the jet carries a huge flux $\mu$ , all stored magnetically, so the plasma barely moves.

Step 3. As magnetic energy converts, $\sigma$ falls and $\Gamma$ rises; complete conversion ( $\sigma \to 0$ ) gives $\Gamma_\infty \to \mu \approx \sigma_0$ . Why: the field sets a hard speed ceiling — you cannot accelerate past the energy you started with — and real jets convert incompletely, retaining $\sigma \sim 0.1$ – $1$ .

Step 4. The equipartition milestone $\sigma = 1$ gives $\Gamma \approx \mu/2$ : half the energy is now kinetic, and this fast-magnetosonic point is where acceleration slows and dissipation (shocks, reconnection) lights up the blazar emission. Why: it marks the handoff from a launching problem to a radiation problem — what telescopes actually detect. *Here $\Gamma_\infty = \mu = \Gamma(1+\sigma)$ is the conserved Bernoulli flux of ideal MHD for relativistic magnetized winds.*

Limiting cases

The mechanisms and regimes, read off the equations above:

Case	Result	Meaning
No spin (BZ)	$P_{\rm BZ} \propto \Omega_H^2 \propto a^2 \to 0$	No ergosphere, no frame-dragging EMF, no Blandford–Znajek jet — rotation is essential.
Near-maximal spin (BZ)	$\Omega_H \to c/(2 r_g)$ ; MAD $\eta_{\rm jet} \gtrsim 1$	Horizon spin saturates; with arrested flux the jet outpowers accretion — spin dominates.
Launching angle (BP)	$\theta < 60^\circ$ from disk plane ( $>30^\circ$ from axis)	Field lines must lean far enough from vertical; steeper lines keep a centrifugal barrier.
Magnetically dominated base	$\sigma \gg 1,\ \Gamma \approx 1$	Energy stored in the field; powerful but slow, and $\mu \approx \sigma_0$ sets the ultimate speed.
Fully converted jet	$\sigma \to 0,\ \Gamma_\infty \to \mu$	All magnetic energy became bulk motion; real jets stall near $\sigma \sim 0.1$ – $1$ .
Beaming cone	$\theta \lesssim 1/\Gamma,\ \delta \sim \Gamma$	Inside the cone the jet is a blazar; outside, the same source looks like a faint radio galaxy.

Common trapJets are not thermal exhaust, and the two mechanisms are not interchangeable

Three traps recur. First, a jet is not "rocket exhaust" — thermal and radiation pressure can neither collimate the flow to a few degrees nor keep it baryon-free; collimation and low mass loading come from the ordered field, not heat. Second, do not conflate the mechanisms: BZ taps the black hole's spin through horizon field lines and works even for a non-accreting hole, while BP taps the disk's rotation and needs gas loaded onto disk field lines — and they often operate together, a fast BZ spine inside a slower BP sheath. Third, never equate the bulk Lorentz factor $\Gamma$ (how fast the plasma moves) with the Doppler factor $\delta$ (how boosted it appears): they coincide only near $\theta \approx 1/\Gamma$ , and fluxes scale as a high power of $\delta$ , so a modest $\Gamma$ aimed at us outshines a faster jet aimed away.

Worked examples

1 · The spin reservoir and spin-down time of M87*

M87 hosts a black hole of $M = 6.5\times10^9\,M_\odot$ with spin $a \approx 0.9$ . How much rotational energy is available to a Blandford–Znajek jet, and how long could a $10^{44}\,\mathrm{erg\,s^{-1}}$ jet run on it?

Step 1. The rest energy is $Mc^2 = (6.5\times10^9)(1.989\times10^{33}\,\mathrm{g})(2.998\times10^{10})^2 = \mathbf{1.16\times10^{64}\ erg}$ . Only the reducible part above the irreducible mass can be extracted.

Step 2. The extractable fraction for spin $a$ is $1 - \sqrt{\tfrac{1}{2}(1+\sqrt{1-a^2})}$ . For $a=0.9$ this is $1 - \sqrt{\tfrac12(1.436)} = \mathbf{0.153}$ , so $E_{\rm rot} = 0.153\,Mc^2 = \mathbf{1.8\times10^{63}\ erg}$ . (At extremal $a=1$ it rises to $0.29\,Mc^2 \approx 3.4\times10^{63}$ erg.)

Step 3. The spin-down time is $t_{\rm spin} \sim E_{\rm rot}/P_{\rm jet} = (1.8\times10^{63})/(10^{44}) = \mathbf{1.8\times10^{19}\ s \approx 6\times10^{11}\ yr}$ — roughly 40 times the age of the universe. The spin is effectively a permanent battery; jet activity is limited by fuel supply and duty cycle, not by draining the reservoir. (Nuclear fusion liberates only $0.7\%$ of rest mass, so spin at $\approx 15\%$ here, up to $29\%$ , is a vastly larger tank.)

2 · Doppler beaming and superluminal motion in a blazar

A jet with $\Gamma = 10$ points $\theta = 5^\circ$ from our line of sight. Find its Doppler factor, flux boost, and apparent transverse speed.

Step 1. The bulk speed is $\beta = \sqrt{1-1/\Gamma^2} = \mathbf{0.99499}$ . With $\cos 5^\circ = 0.99619$ , the beaming denominator is $1 - \beta\cos\theta = 1 - (0.99499)(0.99619) = \mathbf{0.00880}$ .

Step 2. The Doppler factor is $\delta = 1/[\Gamma(1-\beta\cos\theta)] = 1/[10\times0.00880] = \mathbf{11.4}$ — larger than $\Gamma$ itself because we sit inside the beaming cone ( $\theta < 1/\Gamma = 5.7^\circ$ ); head-on the maximum would be $\delta_{\max}\approx2\Gamma=20$ .

Step 3. The flux is boosted by $\delta^{3+\alpha}$ ; a moving-blob $\delta^4$ gives $11.4^4 \approx \mathbf{1.7\times10^{4}}$ . An intrinsically ordinary jet is amplified ten-thousand-fold — which is why blazars dominate the extragalactic gamma-ray sky.

Step 4. The apparent transverse speed is $\beta_{\rm app} = \dfrac{\beta\sin\theta}{1-\beta\cos\theta} = \dfrac{(0.99499)(0.0872)}{0.00880} = \mathbf{9.9}$ . The blob appears to move at $9.9c$ — superluminal motion, and no violation of relativity: a light-travel-time illusion of a sub-luminal flow aimed nearly at us.

3 · Blandford–Znajek power of M87* in the arrested state

With accretion rate $\dot{M} \approx 10^{-3}\,M_\odot\,\mathrm{yr^{-1}}$ and MAD efficiency $\eta \approx 0.8$ at $a=0.9$ , estimate the jet power and compare with observations.

Step 1. Convert the rate: $\dot{M} = 10^{-3}\times(1.989\times10^{33})/(3.156\times10^7) = \mathbf{6.3\times10^{22}\ g\,s^{-1}}$ , so the supply is $\dot{M}c^2 = (6.3\times10^{22})(8.99\times10^{20}) = \mathbf{5.7\times10^{43}\ erg\,s^{-1}}$ .

Step 2. The jet power is $P_{\rm jet} = \eta\,\dot{M}c^2 = 0.8\times5.7\times10^{43} = \mathbf{4.5\times10^{43}\ erg\,s^{-1}}$ , squarely in the observed range $10^{43}$ – $10^{44}\,\mathrm{erg\,s^{-1}}$ from M87's radio lobes and X-ray cavities.

Step 3. Push the spin to $a \to 1$ , where the same MAD flux gives $\eta \approx 1.4$ : then $P_{\rm jet} = 1.4\times5.7\times10^{43} = \mathbf{8\times10^{43}\ erg\,s^{-1}} > \dot{M}c^2$ . This exceeds everything accretion can supply — the surplus is spin energy, the smoking gun of Blandford–Znajek.

Physics of the portrait

Relativistic jets have a single dominant author. Both launching mechanisms carry Roger Blandford's name, and the flat-spectrum theory of their emission is his too — the man who turned a "curious straight ray" in a nearby galaxy into black-hole electrodynamics.

Origins · 1 minRoger Blandford and how this was discovered

Roger Blandford

b. 1949 · relativistic astrophysics and black-hole electrodynamics

Roger Blandford, born in 1949 and trained at Cambridge, built the theoretical scaffolding of nearly everything in this article. In 1977, with his student Roman Znajek, he showed that field lines threading a rotating Kerr horizon set up an electromagnetic circuit that extracts the hole's spin as a Poynting jet — the Blandford–Znajek process, still the leading explanation for the most powerful jets in the universe. In 1979, with Ari Königl, he explained the flat spectra of compact radio cores as a self-absorbed conical relativistic jet, the model still used to read VLBI images. In 1982, with David Payne, he derived the magnetocentrifugal disk-wind mechanism and its bead-on-a-wire criterion, the standard picture for disk-driven outflows from protostars to AGN. Blandford spent decades at Caltech before founding the Kavli Institute for Particle Astrophysics and Cosmology at Stanford, and chaired the 2010 US decadal survey. His hallmark is physical clarity — recasting curved-spacetime electrodynamics as circuits, impedances, and torques — which is why the field's foundations are so durable.

History

1918

— Curtis records a "curious straight ray" — the optical jet of M87

1969

— Penrose shows rotational energy can be extracted from a spinning black hole

1977

— Blandford and Znajek: electromagnetic energy extraction from Kerr holes

1982

— Blandford and Payne: the magnetocentrifugal disk-wind launching criterion

2011

— Tchekhovskoy, Narayan and McKinney: magnetically arrested jets exceed 100% efficiency

Going further

Texts. Carroll & Ostlie, An Introduction to Modern Astrophysics, 2nd ed., Ch. 17 (Kerr) and Ch. 28 (active galaxies) · Binney & Tremaine, Galactic Dynamics, 2nd ed. · Mo, van den Bosch & White, Galaxy Formation and Evolution · Blandford & Znajek (1977), MNRAS 179, 433 · Blandford & Payne (1982), MNRAS 199, 883 · Tchekhovskoy, Narayan & McKinney (2011), MNRAS 418, L79 · Event Horizon Telescope Collaboration (2019, 2021). Key numerical anchors: the $P_{\rm BZ}\propto\Phi^2\Omega_H^2$ scaling carries $\kappa\approx0.053$ . Horizon angular velocity and irreducible mass are standard Kerr results: $a=1\Rightarrow M_{\rm irr}=M/\sqrt{2}$ , $E_{\rm rot}\approx0.29\,Mc^2$ ; $a=0.9\Rightarrow E_{\rm rot}=0.153\,Mc^2$ . The Blandford–Payne launching angle follows from $\Phi''(0)=1-4\cos^2\theta$ , giving $\arccos(1/2)=60^\circ$ from the disk plane (30° from the axis) — note this corrects a common misstatement of $54.7^\circ$ . The terminal Lorentz factor $\Gamma_\infty=\mu=\Gamma(1+\sigma)$ is the ideal-MHD Bernoulli invariant; $\phi_{\rm MAD}\approx50$ and $\eta\sim1.4$ at high spin. Worked-example numbers: $Mc^2=1.16\times10^{64}$ erg, $\delta=11.4$ , $\beta_{\rm app}=9.9$ , $\dot{M}c^2=5.7\times10^{43}\,\mathrm{erg\,s^{-1}}$ .

Builds on: Black Hole Spin and the Ergosphere · Thin Accretion Disks and the Shakura–Sunyaev Model · Magnetorotational Instability and Accretion Turbulence · Advection-Dominated Accretion Flows. Leads to: Blazars and Relativistic Beaming · AGN Feedback · The M–sigma Relation and Black Hole–Galaxy Coevolution · Active Galactic Nuclei: the Unified Model.

Ask the tutor: Want to see how far a magnetically arrested disk can push jet efficiency at your chosen spin, or how the beaming cone reshapes a jet into a blazar — pick a black hole and let's launch its jet together.

How can a jet be more powerful than everything the accretion disk supplies?

It cannot — if accretion were the only source. Converting infalling rest mass to radiation caps out near $42\%$ even for a maximally spinning hole, so an efficiency above $100\%$ of $\dot{M}c^2$ is energy the disk does not have. The resolution is a second reservoir: the rotational energy of the black hole itself, extracted electromagnetically by the Blandford–Znajek mechanism. GRMHD simulations of magnetically arrested disks routinely reach $\eta_{\rm jet}\sim1.4$ at high spin, and this over-unity result is the cleanest dynamical proof that a jet drains a black hole's spin, spinning it down over cosmological times.

What is the difference between the Blandford–Znajek and Blandford–Payne mechanisms?

Both convert rotation into a Poynting-flux jet through twisted field lines, but they tap different flywheels. Blandford–Znajek threads field lines through the event horizon and extracts the black hole's spin; it needs an ergosphere and frame-dragging, and works even for a magnetically threaded hole with no gas accreting. Blandford–Payne anchors field lines in the accretion disk and extracts the disk's orbital energy, flinging gas centrifugally along inclined lines. In practice the two coexist: BZ powers a fast, low-density, highly magnetized spine along the axis, while BP drives a slower, denser, mass-loaded sheath around it — a structure now seen in both simulations and observations.

Why does the black hole's spin, not just its mass, matter for jet power?

Because the Blandford–Znajek power depends on the horizon's angular velocity, $P_{\rm BZ}\propto\Omega_H^2$ , and $\Omega_H$ is set by spin: zero for a non-rotating (Schwarzschild) hole, rising toward $c/2r_g$ near maximal spin. A non-spinning hole has no ergosphere, no frame-dragging, no induced EMF, and no BZ jet — however massive it is or however fast it accretes. Mass sets the overall scale (through $r_g$ and the sustainable accretion rate), but spin is the on/off switch and the throttle. That is why jet power, with the accretion rate, is one of the few observational handles on black-hole spin.

What is a magnetically arrested disk, and why does it matter?

A magnetically arrested disk (MAD) is a state in which so much ordered poloidal flux has piled up near the horizon that its magnetic pressure pushes back on the inflow, halting it — gas then leaks inward in discrete, bubble-like intervals rather than smoothly. The flux saturates at a near-universal value ( $\phi_{\rm MAD}\approx50$ ), so MAD jets reach a maximum efficiency that depends almost entirely on spin. It matters because it is the regime where jets are most powerful and where Blandford–Znajek efficiency exceeds unity. The 2021 EHT polarization images of M87* show exactly the ordered, spiral horizon-scale field a MAD state predicts, making it the leading model for the most luminous AGN jets.

Why do jets start out barely moving if they are called "relativistic"?

At the launch point almost all of a jet's energy is stored in its magnetic field, not in bulk motion — the magnetization $\sigma = B^2/(4\pi\rho c^2)$ is much greater than one. The conserved flux $\mu = \Gamma(1+\sigma)$ can be enormous while $\Gamma\approx1$ , because the large factor lives in $\sigma$ . As the flow expands, magnetic energy converts to kinetic: $\sigma$ falls, $\Gamma$ climbs, and the jet reaches its terminal $\Gamma_\infty\approx\sigma_0$ only far downstream. So "relativistic" describes the terminal state; the base is a slow, magnetically dominated launcher. Incomplete conversion leaves residual magnetization that later powers the radiation through shocks and reconnection.

What actually collimates a jet into such a narrow beam?

The same toroidal field that carries the energy also collimates it. As the anchor rotates, it winds the field into tight azimuthal loops, and the tension of those loops — the magnetic "hoop stress" — squeezes inward toward the axis, like stretched rubber bands pinching a column of fluid. This self-collimation focuses the flow into an opening angle of a few degrees, far narrower than any thermal outflow, and works over enormous distances because the ordered field is preserved along the jet. External pressure from the disk wind or interstellar medium tunes the profile, but the core mechanism is the jet's own toroidal field.