Failed Jets in Astrophysics & Heavy-Ion Collisions

• Failed jets are phenomena where launched jets lose collimation or are choked by ambient material, preventing canonical observable signatures.
• They appear in diverse contexts including compact merger outflows, choked stellar-collapse jets, and AGN jets disrupted by dense media.
• In heavy-ion collisions, failed jets show as quenched recoil signals, where energy is redistributed into softer, broader emissions.

Searching arXiv for relevant papers on failed jets across astrophysical and heavy-ion contexts. “Failed jets” denotes a heterogeneous class of phenomena in which a jet either does not emerge as an observable prompt outflow, loses its collimation before reaching large scales, is quenched by interaction with ambient matter, or becomes so strongly modified that the original jet signature is effectively absent. In the literature, the term is applied to at least four distinct settings: low-$\Gamma$ compact-merger outflows that produce no prompt short-$\gamma$-ray burst but do generate orphan afterglows; relativistic stellar-collapse jets choked by circumstellar material; AGN jets disrupted by dense media or by accretion-state changes; and, in heavy-ion collisions, recoil jets whose correlated high-$p_T$ signature is strongly suppressed by the quark–gluon plasma (Lamb et al., 2016, Lamb et al., 2017, Martin-Carrillo et al., 8 Jun 2026, Tseng et al., 2024, Lalakos et al., 2023, Ohlson, 2012). A broader feedback-oriented usage treats “failed jets” as cases where the negative jet-feedback mechanism does not self-regulate accretion, so the jet does not couple effectively to the ambient reservoir (Soker, 2015).

1. Terminological scope and unifying concept

The phrase does not identify a single physical mechanism. In compact-object mergers, a “failed GRB” or “failed jet” is an outflow whose Lorentz factor is too low for prompt $\gamma$-rays to escape, because dissipation occurs below the photosphere (Lamb et al., 2016, Lamb et al., 2016). In stripped-envelope supernovae, failure often means jet choking: the engine launches a relativistic jet, but the jet head stalls in dense circumstellar material and deposits its energy into a cocoon or envelope rather than producing a classical breakout signature (Hamidani et al., 20 Mar 2025, Martin-Carrillo et al., 8 Jun 2026). In Galactic-center and radio-galaxy contexts, “failed jets” refers to jets that lose collimation through entrainment, turbulence, precession, or kink-driven disruption, so that they inflate bubbles or remain compact rather than forming long-lived large-scale radio structures (Tseng et al., 2024, Lalakos et al., 2023). In relativistic heavy-ion collisions, the term is used descriptively for recoil jets that are so quenched in the QGP that they effectively disappear as well-defined high-$p_T$ jets (Ohlson, 2012).

A unifying interpretation is that “failure” generally refers to a mismatch between jet launching and jet survival. The engine or hard scattering may occur, but the system does not preserve the conditions needed for a canonical, directly observable jet signature. This suggests that the relevant control parameters are not only launch power, but also optical depth, angular structure, entrainment, ambient density, engine duration, and the degree of coupling to the surrounding medium.

2. Low-$\Gamma$ merger jets and failed short GRBs

In the compact-binary merger literature, a successful short GRB requires prompt $\gamma$-rays to be produced above the photosphere $R_p$, but below the dissipation radius $R_d$ (Lamb et al., 2016). One formulation gives

$$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$

so if $\gamma$0 is sufficiently low that $\gamma$1, prompt high-energy photons are scattered and no prompt $\gamma$2-ray pulse escapes (Lamb et al., 2016). A related criterion is

$\gamma$3

which implies a lower limit

$\gamma$4

for prompt escape (Lamb et al., 2016). Outflows below that threshold remain optically thick at the dissipation site, photons are trapped and reconverted to kinetic energy, and the event is termed a “failed GRB” or “failed jet” (Lamb et al., 2016).

Monte Carlo studies adopt a power-law Lorentz-factor distribution,

$\gamma$5

with $\gamma$6 and $\gamma$7 or $\gamma$8, $\gamma$9, motivated by analogy with blazar and AGN jets (Lamb et al., 2016, Lamb et al., 2016). Under this assumption, the local merger-jet population is dominated by low-$p_T$0 outflows, and approximately $p_T$1 of on-axis merger jets result in failed GRBs (Lamb et al., 2016, Lamb et al., 2016). One numerical threshold quoted for prompt nondetection is

$p_T$2

for the adopted parameter set (Lamb et al., 2016).

The observational consequence is not the absence of emission, but the replacement of a prompt short GRB by a broadband afterglow once the outflow decelerates in the ambient medium. The deceleration time is written as

$p_T$3

and the synchrotron afterglow is characterized by standard $p_T$4, $p_T$5, and $p_T$6 scalings (Lamb et al., 2016). Simulations for sources within $p_T$7 Mpc yield optical peaks $p_T$8 mag, optical peak times $p_T$9 days, X-ray fluxes $\gamma$0 at $\gamma$1 day, $\gamma$2 GHz radio fluxes $\gamma$3 at $\gamma$4 days, and $\gamma$5 MHz fluxes $\gamma$6 at $\gamma$7 days (Lamb et al., 2016). An earlier formulation gives $\gamma$8 GHz peaks of $\gamma$9 mJy and $p_T$0 MHz peaks of $p_T$1 mJy, with radio peak times narrowly clustered around $p_T$2 days (Lamb et al., 2016). The differing quoted ranges reflect different presentation choices in the supplied summaries.

A common misconception is that a failed short GRB is electromagnetically silent. The merger studies instead identify such jets as candidate electromagnetic counterparts to gravitational-wave sources precisely because the prompt $p_T$3-ray channel is suppressed while the afterglow remains observable (Lamb et al., 2016, Lamb et al., 2016).

3. Structured jets, orphan afterglows, and angular dependence

Structured-jet models generalize the failed-GRB picture by allowing $p_T$4 and $p_T$5 to vary with polar angle rather than remaining constant inside a top-hat core (Lamb et al., 2017). In two-component, power-law, and Gaussian structured jets, low-$p_T$6 wings suppress prompt $p_T$7-rays in the same manner as a failed homogeneous jet, but later contribute a fainter, more slowly peaking afterglow (Lamb et al., 2017). This makes the concept of a failed jet intrinsically angle-dependent: some angular sectors can be prompt-dark even when the core is prompt-bright.

For off-axis observers in the slow-cooling regime, the peak afterglow flux and time can be approximated by

$p_T$8

$p_T$9

with $\Gamma$0 (Lamb et al., 2017). These scalings imply that wider jets can produce orphan afterglows brighter than the macronova for the typical GW-detected inclination of $\Gamma$1 (Lamb et al., 2017).

Monte Carlo calculations out to $\Gamma$2 Mpc give the following fractions of NS–NS mergers with peak $\Gamma$3-band counterparts brighter than $\Gamma$4 (Lamb et al., 2017):

Jet model	Parameters	Fraction with $\Gamma$5
Homogeneous	$\Gamma$6	$\Gamma$7
Two-component	$\Gamma$8 core + $\Gamma$9 wing	$\gamma$0
Power-law	$\gamma$1 outside $\gamma$2 core to $\gamma$3	$\gamma$4
Gaussian	$\gamma$5, $\gamma$6	$\gamma$7

The same calculations decompose these into successful short GRBs and orphan counterparts. For homogeneous jets, $\gamma$8 are orphans and $\gamma$9 are successful short GRBs; for two-component jets, $R_p$0 are orphans and $R_p$1 are GRBs; for power-law jets, $R_p$2 are orphans and $R_p$3 are GRBs; for Gaussian jets, $R_p$4 are orphans and $R_p$5 are GRBs (Lamb et al., 2017).

Light-curve morphology becomes a diagnostic of failed-jet structure. Two-component jets can show an early modest peak from the sheath followed by a re-brightening when the core becomes visible; power-law jets produce a gradual single-peaked light curve; Gaussian jets viewed far off-axis can exhibit a soft slowly rising orphan peaking at $R_p$6 weeks (Lamb et al., 2017). This suggests that “failure” in structured jets is not binary, but distributed over angle and encoded in multi-band temporal evolution.

4. Choked and barely failed jets in stellar collapse

In engine-driven supernovae and long-GRB environments, failure is primarily a breakout problem. The analytic model of jet propagation in circumstellar material introduces a one-hemisphere jet with power $R_p$7, opening angle $R_p$8, engine lifetime $R_p$9, and total energy $R_d$0, propagating into a static $R_d$1 CSM (Hamidani et al., 20 Mar 2025). Defining

$R_d$2

the head velocity satisfies

$R_d$3

with an analytic approximation accurate to $R_d$4 over $R_d$5 to $R_d$6 (Hamidani et al., 20 Mar 2025). Since $R_d$7 is constant in time for the $R_d$8 CSM case,

$R_d$9

The criterion for success or failure compares the breakout time $$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$0 to the time for the engine tail to catch the head. Using the jet radial length

$$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$1

a jet is successful only if

$$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$2

and failed if

$$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$3

(Hamidani et al., 20 Mar 2025). This formalism yields a critical CSM mass,

$$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$4

with jets failing when $$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$5 (Hamidani et al., 20 Mar 2025). In the extended-CSM limit,

$$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$6

To describe the relativistic content of the ejecta, the model defines an energy-weighted proper velocity $$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$7, which separates three regimes: successful jets with $$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$8, barely failed jets with $$R_d\sim 2c\,\delta t\,\Gamma^2, \qquad R_p\simeq \left(\frac{\sigma_T E_{\rm iso}}{4\pi m_p c^2 \Gamma}\right)^{1/2},$$9, and completely failed jets with $\gamma$00 (Hamidani et al., 20 Mar 2025). Barely failed jets correspond to marginal failure with a mildly relativistic inner cocoon core; completely failed jets correspond to deep choking and full mixing into a non-relativistic cocoon (Hamidani et al., 20 Mar 2025). The predicted luminous thermal signature is cocoon cooling emission with characteristic cooling timescale of hours to days, peak luminosity $\gamma$01, and early blackbody temperature $\gamma$02 K (Hamidani et al., 20 Mar 2025).

A concrete observed candidate is SN 2026gzf, associated with the thermal X-ray shock-breakout transient EP260321a at $\gamma$03 (Martin-Carrillo et al., 8 Jun 2026). The event showed a $\gamma$04 s thermal X-ray flash with $\gamma$05 keV and no non-thermal tail, early optical data beginning at $\gamma$06 hr, a two-component light-curve model requiring $\gamma$07 at $\gamma$08 cm, and broad-lined spectra with $\gamma$09 (Martin-Carrillo et al., 8 Jun 2026). Deep VLA upper limits at $\gamma$10 d and $\gamma$11 d rule out any on-axis GRB-like afterglow with typical energies $\gamma$12 erg in an ISM of density $\gamma$13 for $\gamma$14 (Martin-Carrillo et al., 8 Jun 2026). The supplied interpretation is that a jet with $\gamma$15, $\gamma$16 s, and $\gamma$17 is below the critical luminosity for breakout through the shell and is therefore choked (Martin-Carrillo et al., 8 Jun 2026). The paper identifies SN 2026gzf as the first “failed jet” confirmed in an Ic-BL SN (Martin-Carrillo et al., 8 Jun 2026).

A broader implication is that failed or choked jets may be considerably more common than successful GRBs. For SN 2026gzf, the summary quotes a rate $\gamma$18 for EP-like shock breakouts, while GRBs occur at $\gamma$19 (Martin-Carrillo et al., 8 Jun 2026). This suggests that engine-driven but jet-failed explosions may populate a large fraction of the Ic-BL phenomenology.

5. Failed AGN jets, bubbles, and compact radio sources

In galactic and AGN environments, failed jets are often disrupted jets rather than prompt-dark jets. Three-dimensional special relativistic hydrodynamic simulations with cosmic-ray advection show that dense clumpy gas in the Galactic disk can disrupt oblique jets launched at angles $\gamma$20 to the disk normal, causing the jets to lose collimation and inflate hot bubbles (Tseng et al., 2024). The simulations use bipolar jets with injected four-velocity $\gamma$21 ($\gamma$22), $\gamma$23 MeV, $\gamma$24, $\gamma$25, and duration $\gamma$26 Myr (Tseng et al., 2024). Once the narrow channels are destroyed, the thermalized energy produces hot plasma reservoirs with $\gamma$27 keV and $\gamma$28, which rise buoyantly and form nearly vertical bubbles despite the initially tilted jet axes (Tseng et al., 2024).

The resulting forward shock corresponds to the eROSITA-bubble edge and the contact discontinuity to the Fermi-bubble edge (Tseng et al., 2024). In the fiducial run at $\gamma$29 Myr, the forward shock lies at $\gamma$30 kpc with semiminor radius $\gamma$31 kpc; the shock Mach number $\gamma$32 implies post-shock $\gamma$33 keV (Tseng et al., 2024). Assuming a leptonic model with CR-electron spectral index $\gamma$34, the simulations reproduce the observed gamma-ray bubbles and microwave haze, and the best-fit CRe power-law spectral index is $\gamma$35 (Tseng et al., 2024). Here “failed jets” therefore generate a large-scale bipolar structure not by penetrating the halo ballistically, but by thermalizing close to the disk.

A different AGN failure mode appears in 3D GRMHD simulations of zero-angular-momentum accretion onto a spinning SMBH (Lalakos et al., 2023). The system develops a magnetically arrested disk (MAD), then transitions to a “barely-a-disk” (BAD) state and finally a rocking accretion disk (RAD) state. During MAD, the normalized magnetic flux reaches $\gamma$36, the accretion rate is suppressed to $\gamma$37, and the jet efficiency is $\gamma$38; the jets propagate to $\gamma$39 and remain straight to $\gamma$40 before helical bends grow (Lalakos et al., 2023). In BAD, $\gamma$41, $\gamma$42, and $\gamma$43; in RAD, $\gamma$44, $\gamma$45, and $\gamma$46, but the jets are intermittent, misaligned, and fail inside $\gamma$47 (Lalakos et al., 2023).

The destruction mechanism combines disk tilt and precession with external kink instability. The relevant constraint is

$\gamma$48

so sufficiently large footpoint whipping forces deceleration and destabilization (Lalakos et al., 2023). In BAD, the core-averaged proper velocity $\gamma$49 falls from $\gamma$50 to $\gamma$51, recollimation points develop at $\gamma$52, and strong bends at $\gamma$53 enforce global dissipation (Lalakos et al., 2023). In RAD, disk flips launch jets at widely varying inclinations, causing kink growth on scales of $\gamma$54 and catastrophic loss of collimation well inside the Bondi radius (Lalakos et al., 2023). These compact destroyed jets are proposed as a natural channel for FR0 radio galaxies (Lalakos et al., 2023).

The broader feedback perspective in Soker’s review characterizes failure as a fizzle of the negative jet-feedback mechanism. If the jet is too narrowly collimated, the axis too stable, or the momentum flux too small, only a small fraction of the reservoir is impacted and feedback becomes ineffective (Soker, 2015). In that framework, failed jet feedback can lead to cooling catastrophes in clusters, runaway SMBH growth in young galaxies, BH formation and GRBs in CCSNe, or mergers in CEE/GEE systems (Soker, 2015). This is a more systemic usage: the jet may exist, but it fails as a regulator.

6. Quenched recoil jets in heavy-ion collisions

In high-energy nuclear physics, “failed jets” refers not to hydrodynamic choking or prompt-dark relativistic outflows, but to severe in-medium modification of hard-scattered partons traversing the quark–gluon plasma. STAR measures azimuthal correlations between charged hadrons and reconstructed trigger jets, together with recoil-jet coincidence observables, to quantify this suppression (Ohlson, 2012). The central observables include the nuclear modification factor

$\gamma$55

the per-trigger yield modification

$\gamma$56

the jet-shape observable $\gamma$57, and the energy-balance observables

$\gamma$58

$\gamma$59

(Ohlson, 2012).

The azimuthal correlation function is written as

$\gamma$60

with a background modulated by flow harmonics,

$\gamma$61

(Ohlson, 2012). STAR fits the raw distribution to two Gaussians plus this Fourier-modulated background to extract peak yields and widths (Ohlson, 2012).

The measurements most directly associated with “failed jets” are the away-side suppression results. Jet–hadron correlations with trigger-jet $\gamma$62 and $\gamma$63 GeV/$\gamma$64 show that $\gamma$65 is positive at low $\gamma$66 GeV/$\gamma$67 and large-negative at high $\gamma$68 GeV/$\gamma$69, while the integrated imbalance $\gamma$70 is small compared to the trigger-jet energy (Ohlson, 2012). Recoil-jet coincidence measurements show that the per-trigger spectrum of recoil jets in Au+Au is suppressed by a factor of $\gamma$71 relative to $\gamma$72 for both constituent cuts $\gamma$73 GeV/$\gamma$74 and $\gamma$75 GeV/$\gamma$76 (Ohlson, 2012). The quoted $\gamma$77 values are $\gamma$78 GeV/$\gamma$79 for trigger jets of $\gamma$80 GeV/$\gamma$81 and $\gamma$82 GeV/$\gamma$83 for $\gamma$84 GeV/$\gamma$85 (Ohlson, 2012).

The interpretation is that the recoil partner in central Au+Au is either strongly broadened out of the cone or softened into the bulk, so that the jet “fails” as a well-defined high-$\gamma$86 object (Ohlson, 2012). This usage differs substantially from the astrophysical one: the parent hard process occurs, but the medium erases the recognizable recoil-jet signature.

7. Cross-domain patterns and major distinctions

Across the supplied literature, failed jets arise from several recurring failure channels.

Optical-depth failure: in merger jets, low $\gamma$87 places the dissipation radius below the photosphere, suppressing prompt $\gamma$88-ray escape while preserving the external-shock afterglow (Lamb et al., 2016, Lamb et al., 2016).

Breakout failure: in GRB-SN and Ic-BL systems, dense or extended CSM causes the engine tail to catch the head before breakout, yielding barely failed or completely failed jets and transferring energy to a cocoon (Hamidani et al., 20 Mar 2025, Martin-Carrillo et al., 8 Jun 2026).

Collimation failure: in AGN or Galactic-center environments, entrainment by clumpy gas, buoyant bubble inflation, or kink-driven destabilization destroys narrow jet channels and replaces them with bubbles or compact radio structures (Tseng et al., 2024, Lalakos et al., 2023).

Feedback failure: in the JFM framework, jets exist but do not couple strongly enough to halt the inflow, so negative feedback “fizzles” and the system evolves toward runaway or over-energetic states (Soker, 2015).

Medium-induced disappearance: in heavy-ion collisions, the recoil jet is so quenched that it is dramatically softened or effectively disappears from high-$\gamma$89 observables (Ohlson, 2012).

A common misconception is that a failed jet is simply a weak jet. The supplied literature does not support such a uniform definition. Some failed jets are initially relativistic and energetic but are choked by CSM; some are powerful enough to inflate kpc-scale or Galactic-scale bubbles after losing collimation; some are observationally “failed” only in the prompt band; and some are defined entirely by analysis observables rather than launch physics (Hamidani et al., 20 Mar 2025, Martin-Carrillo et al., 8 Jun 2026, Tseng et al., 2024, Lamb et al., 2016, Ohlson, 2012). A plausible implication is that “failed jet” is best understood as an outcome class rather than an engine class.

The current literature therefore uses the term as a diagnostic of hidden energy release. Whether in orphan afterglows from compact mergers, cocoon-powered transients from jet-choked supernovae, Galactic bubbles from disrupted AGN outflows, FR0-like compact radio sources, or quenched recoil jets in QGP, the central problem is the same: a jet-launching or jet-producing event has occurred, but the canonical jet signature has been transformed, obscured, or erased by the surrounding medium and the dynamical history of the flow.