Explaining the X-ray Precursor, Ultra-long Prompt Emission, and Week-long Decay of GRB250702B with a Jetted Micro-TDE

The longest detected gamma-ray burst, GRB 250702B, exhibited seven hours of prompt gamma-ray emission, preceded by a soft X-ray precursor (∼ 1 day earlier) and followed by a weeks-long fading X-ray tail. Lacking an established progenitor for all three phases, we propose that this ultra-long GRB (ULGRB) is powered by a jetted micro-tidal disruption event (micro-TDE), in which a spinning stellar-mass black hole (BH) disrupts a Sun-like star and launches a relativistic jet via the Blandford–Znajek mechanism. Micro-TDE debris disks have hours-to-days viscous timescales, naturally explaining ULGRB durations. Using 3D hydrodynamic arepo simulations of a 1 M⊙ star disrupted by a 10 M⊙ BH, we show that within ∼ 1 day the debris forms a quasi-steady envelope with a low-density polar funnel (rho ∝ r −2, half-opening angle ≈ 15◦). Applying an analytic jet-stability framework to these profiles, we find that the r −2 funnel keeps the jet below the kink-instability threshold, enabling stable propagation and breakout for jet powers, L jet ≳ 1047 erg s−1. We attribute the X-ray precursor to pre-disk stream-fed accretion; the prompt GRB to a tightly beamed jet (theta b ≲ 1◦, L gamma,iso ∼ 1051 erg s−1) escaping the funnel, launched by a rapidly spinning BH (a • ∼ 0.9); and the weeks-long X-ray decline to disk-wind mass loss (L jet ∝ t −2) combined with jet widening (theta b ∝ t, initially steepening the decay to L X,iso ∝ L jet /theta b2 ∝ t −4). Our model reproduces the multi-phase evolution of GRB 250702B and establishes jetted micro-TDEs as a physically motivated ULGRB engine.

The paper is organized as follows. Section 2 describes the hydrodynamic simulations of micro-TDEs. Section 3 presents the micro-TDE disk evolution model based on viscous accretion and wind-driven mass loss. Section 4 presents the observational signatures of jetted micro-TDEs and assesses jet stability during propagation through the debris envelope. Section 5 applies our micro-TDE accretion model to GRB 250702B, constraining the black hole spin and beaming angle required to reproduce the observed prompt gamma-ray luminosity and the X-ray light curve. We summarize our results and discuss their implications in Section 6.

We perform 3D hydrodynamic simulations using the moving-mesh code arepo. We adopt the Helmholtz equation of state, which includes radiation pressure under the assumption of local thermodynamic equilibrium with composition inherited from the MESA progenitor model. We model the black hole as a point particle of mass M• = 10 M⊙ that interacts with the gas only gravitationally and neither accretes any mass nor produces any radiative feedback. We then model in post-processing the subsequent (unresolved) disk evolution and compute the time-dependent accretion rate onto the black hole, using the fallback rate M¤ fb (t) computed from the arepo simulations. We create a 1 M⊙ MESA model with hydrogen core mass fraction of 0.5 (corresponding to an age of 2.5 Gyr) and metallicity of Z = 1 Z ⊙. We consider two initial orbits, both leading to the full disruption of the star: a grazing encounter with the penetration factor beta = 1.4 and semi-major axis a = 3 R ⊙, and a deep encounter with beta = 5 and a = 1 R ⊙, where beta ≡ r T /r p > 1, r T is the tidal disruption radius and r p ≡ a(1 − e) is the pericenter distance. In both cases the orbital eccentricity is e = 0.5 and the star begins at apocenter r apo ≡ a(1 + e). The bound orbit with our adopted moderate eccentricity (e = 0.5) results in about 80% of the stellar mass being captured onto the black hole.

Figure 1 shows face-on and edge-on density slices of the debris envelope formed after the disruption of a 1M⊙ star by a 10M⊙ black hole at t = 0.5 day, for two penetration factors: a grazing encounter (beta = 1.4; left panels) and a deep encounter (beta = 5; right panels). The deep encounter produces a compact, dense inner disk (r d ≈ 1R ⊙), while the grazing encounter forms a more extended configuration (r d ≈ 3R ⊙). Both encounters develop a well-defined bipolar funnel with half-opening angle theta f ≈ 15◦ that can serve as a convenient escape funnel for black hole powered relativistic jets. The density contrast between the polar tunnel and the equatorial plane spans more than three orders of magnitude at r ∼ R ⊙, from rho ∼ 10−4 g cm−3 at theta = 0◦ to rho ≳ 10−1 g cm−3 at theta = 90◦. The polar density profile is well described by a power-law rho ∝ r −2, while at intermediate and equatorial angles the profiles steepen towards rho ∝ r −3. The polar r −2 scaling is the key input to our jet propagation and stability calculations: it provides a steadily declining confining pressure that collimates rather than disrupts the jet.

We develop an analytic model for the subsequent viscous accretion by incorporating fallback supply, viscous drainage onto the black hole, and wind-driven mass loss. The disrupted star forms the fallback debris and accretion disk of the characteristic size set by the circularization radius, r c ≃ 2r p = 2r T /beta ≈ 2 × 105 r g beta −1 m −2/3 ∗,1 m 2/3 •,10. Adopting a standard alpha-prescription we calculate the viscous accretion timescale of a geometrically thick disk with aspect ratio h ≡ 0.5h0.5 ≡ H/R for the fiducial parameter choices as tv ≃ alphav−1 h −2 (r c /r g) 3/2 r g /c ≈ 2 day beta −3/2 alphav−1,−1 h −2 0.5 m ∗,1. From our arepo simulations, we see that the two encounters produce different aspect ratios near the circularization radius: the deep encounter (beta = 5) yields h ≈ 0.5, characteristic of geometrically thick, advection-dominated accretion flows at highly super-Eddington rates, which are expected to eject most of the disk mass as winds. The grazing encounter (beta = 1.4) instead settles into a thinner, more rotationally supported configuration with h ≈ 0.2. Because the viscous timescale scales as tv ∝ h −2, it is approximately six times longer than in the grazing case than in the deep case, at the same radius. This slower viscous evolution shifts the accretion peak to later times and lower amplitudes, delaying the onset of the asymptotic M¤ acc ∝ t −2 decline. In what follows, we find that the beta = 5 configuration provides a better match to the late-time X-ray decline of GRB 250702B.

Following Metzger et al. (2008); Kremer et al. (2019, 2023), we approximate the disk mass distribution as a single ring at radius r d where the surface density peaks, and track its mass and angular momentum evolution under fallback supply M¤ fb, viscous accretion onto the black hole, and wind-driven mass loss. The disk mass evolves as M¤ d = − f Md /tv + M¤ fb. Following Blandford & Begelman (1999), we include mass loss from disk winds by adopting a radius-dependent inflow rate M¤ in (r) = (r/r d) s Md /tv, where 0 ≤ s ≤ 1 parametrizes the radial slope of the wind mass outflow. We assume that the wind starts at the radius r acc so that the inflow rate at this radius, M¤ acc = (r acc /r d) s Md /tv, gives the black hole accretion rate. The wind carries away the remainder, M¤ out = [1 − (r acc /r d) s] Md /tv. The disk radius evolves as r¤d = 2r d /tv [1 − C (1 − (r acc /r d) s)] + (r c /r d) −1/2 M¤ fb tv /Md, where C = 2s/(2s + 1) is the torque coefficient assuming the wind carries away angular momentum with its local specific value but exerts no net torque. We solve these equations numerically to compute Md (t) and r d (t), where we take M¤ fb (t) directly from our arepo simulations. As the disk radius r d increases, the viscous accretion timescale increases accordingly t v ∝ r d3/2. We account for this evolving timescale when solving for the coupled evolution of the disk mass and radius. Inserting these time-dependent quantities into the equation for M¤ acc then yields the time-dependent accretion rate onto the black hole. At late times, the system asymptotes to the following scalings: r d ∝ t 2/3, M¤ d ∝ t − (2s+4)/3 = t −5/3, M¤ acc ∝ t −4(s+1)/3 = t −2, where the decay of M¤ acc, steeper than the canonical t −5/3 TDE fallback rate, is the direct consequence of viscous disk spreading combined with wind-driven mass loss. We identify this M¤ acc ∝ t −2 scaling in Section 5 as the origin of the late-time X-ray decay of GRB 250702B. The s = 0.5 value is supported by recent simulations of radiatively inefficient accretion, including neutron star post-merger accretion disks, advection-dominated flows and Bondi accretion with weak cooling. Recent GRMHD simulations have revealed that s is not a universal constant of accretion but depends on the state of the accretion flow. When large-scale vertical magnetic flux fills the black hole to capacity and becomes as strong as the gravitational force acting on the inner disk, it erupts from the black hole, rips through the disk, and buoyantly rises away from the hole. This magnetically arrested disk (MAD) state features ordered accretion, strong magnetically-powered jets, and s ≈ 0.66 ± 0.03. MAD-powered outflows can scramble the angular momentum of the system and lead to a rocking accretion disk (RAD) state. The RAD state features chaotic accretion, continuously reorienting weak jets, and s ≈ 0.87 ± 0.05.

Large-scale magnetic field threading a rotating black hole can extract its spin energy and launch twin Poynting-flux-dominated jets of electromagnetic luminosity, L jet = etajet (a •) M¤ acc c2, via the Blandford & Znajek (1977, BZ hereafter) process. Here, M¤ acc is the black hole accretion rate, for which we adopt the expression given by Equation (4), and a • is the dimensionless BH spin. The jet energy efficiency, defined by eq. (9) as the ratio of the extracted electromagnetic energy to accreted rest-mass energy, has been calibrated by GRMHD simulations, etajet (a •) ≃ 2.64 (phi • /50) 2 omega2H f (omegaH), where omegaH = a • / 1 + (1 − a 2•) 1/2 is the dimensionless angular frequency of the black hole and f (omegaH) ≃ 1 + 0.35omega2H − 0.58omega4H is a high-spin correction. Here, phi• ≡ Φ• /(M¤ r g2 c) 1/2 is the dimensionless black hole magnetic flux, which saturates at phi• ∼ 50 in the MAD state. Twin relativistic jets focus their radiation into a narrow solid angle Ωb = 4pi(1 − cos theta b) around the disk rotational axis. Here, theta b is the beaming angle, which is set by the larger of the jet opening angle, theta j and the relativistic beaming angle, 1/gammaj: theta b = max(theta j, 1/gammaj). An observer aligned with the jet axis therefore infers an isotropic-equivalent luminosity that is enhanced relative to the true twin jet power by the inverse of the beaming fraction, fb ≡ Ωb /4pi = 1 − cos theta b. The isotropic-equivalent luminosity is therefore L iso (t) = etarad etajet fb−1 M¤ acc (t)c2, where etarad is the fraction of the jet power that is radiated in X-ray or gamma-ray band and convert the efficiency.

In general, the black hole spin axis can be misaligned with the angular momentum axis of the debris disk, depending on the initial spin–orbit misalignment of the BH–star orbit. In such a case, the disk undergoes solid-body Lense–Thirring precession with period PLT ≈ (pi/5a •) (r g /c) (r d /r g) 5/2 (r ISCO /r g) 1/2. Numerically for s = 0.5 and r d ∝ t 2/3, PLT ≈ 1.8 × 10 4 day beta −5/2 a −1 •,0.7 m 5/3 ∗,1 m −2/3 •,10 (t/tv) 5/3. Lense–Thirring torques are therefore not strong enough to cause the precession of the jets over the jet engine lifetime (∼ day). Physically, this is because micro-TDE disks feature significant radial extents (r d ∼ 105 r g) that imply large angular momentum content and long precession periods. Recent GRMHD simulations show that magnetic torques from powerful jets can, in some regimes, reorient the inner disk angular momentum into alignment with the black hole spin direction on timescales comparable to or shorter than the accretion timescale. However, the alignment is limited to the immediate vicinity of the black hole, r ≲ 100r g, which is orders of magnitude smaller than the simulated length scales. As a result, tilted black hole systems with extended accretion disks in micro-TDEs launch jets along the rotational axis of the disk irrespective of the initial black hole spin–disk misalignment. We therefore assume the configuration in which the jets propagate through the low-density polar funnel along the rotational axis of the extended disk.

The propagation of a Poynting-dominated relativistic jet through a confining medium is subject to current-driven (kink) instabilities that can disrupt the jet and cause it to stall before breakout. We can quantify the stability of the jet through the stability parameter Λ = 2gammaj theta j /0.03 = K (L 1,jet /(rhoa z2h gammaj2 c3)) 1/6, where gammaj is the bulk Lorentz factor, where L 1,jet = 0.5L jet is one-sided jet power, zh is the distance to the jet head, rhoa is the ambient density, and K = 20 (2pi/9) 1/2 [pi(5−alpha) (3−alpha)/6] 1/3 is a constant numerical prefactor that depends on the radial slope of the ambient density, alpha ≡ −d log rhoa /d log r. Our simulations have alpha = 2 along the polar direction, as seen in Figure 2. We assume the jet is highly magnetized, with magnetization, sigma ≡ b 2 /(4pirhoc2) ≫ 1, and moves at betaj ≡ vj /c ≃ 1, corresponding to Lorentz factor gammaj ≫ 1. Because the polar density follows rhoa ∝ r −2, the factor of rhoa z2h in Equation (14) is independent of the jet-head distance, and so is Λ; the stability result therefore holds at all radii and does not rely on the jet reaching a particular breakout height. Our profile sits precisely at the critical slope alpha = 2 separating the two regimes identified by Bromberg & Tchekhovskoy (2016): in steeper profiles (alpha > 2) the jet gradually opens up and stabilizes as it propagates, whereas in flatter profiles (alpha < 2) it gradually collimates and grows increasingly unstable. Figure 3 shows the jet stability parameter, Λ, evaluated at a fixed jet head distance, zh = 10 R ⊙, as a function of the intrinsic jet luminosity, L jet, for three representative Lorentz factor values, gammaj = 1, 10, 100, and for three representative jet opening angles, theta j = 0.017◦, 0.17◦, and 1.7◦. Reading the stability threshold (Λ > 2) off Figure 3, the most relativistic jet (gammaj = 100), which is most susceptible to the kink instability, requires the largest jet power to remain stable, L jet,min ≃ 3 × 1047 erg s−1. Producing this power from the micro-TDE wind-modified (s = 0.5) accretion rate M¤ acc ∼ 10−7 M⊙ s−1 requires a rapidly spinning black hole, a • ≳ 0.9, and hence a jet efficiency etajet ≳ 1. Because slower, less relativistic jets have progressively lower stability thresholds, a jet launched by a black hole with the same jet power lies in the stable region (Λ > 2) across the full range gammaj ≲ 100 considered here. Jet survival across the range of Lorentz factors and jet geometries considered in this work therefore makes micro-TDEs a viable central engine for ULGRB s.

GRB 250702B/EP 250702a exhibits three temporally and spectrally distinct phases. Einstein Probe detected a soft X-ray precursor approximately one day before the main burst, followed by three hard gamma-ray episodes registered by Fermi-GBM with quasi-regular spacing of ∼ 2, 825 s. The Einstein Probe localization enabled continued monitoring with Swift, NuSTAR, and Chandra, revealing a long-lived X-ray counterpart fading over days to weeks. The late time X-ray light curve of EP 250702a exhibits two observationally distinct phases. The first is the EP-WXT phase, lasting several hours and contemporaneous with the Fermi gamma-ray triggers, during which the X-ray emission peaks at L X,iso ∼ 1049 erg s−1 with a hard spectrum (Γ ≃ 0.2). The second is the EP-FXT/Swift/Chandra phase, beginning after the final Fermi trigger and decaying as L X ∝ t −1.9 with a softer spectrum (Γ ≃ 1.8). We propose that the gamma-ray and soft X-ray emission from GRB 250702B originates from a jetted micro-TDE in which a rapidly spinning stellar-mass black hole disrupts a companion main-sequence star. This interpretation is motivated by three observational features that are difficult to reconcile with a standard collapsar or blue-supergiant scenario: the day-long soft X-ray precursor with no natural analog in collapsar models, the absence of an accompanying supernova signature, and the multi-hour duration of the prompt emission. The 5.7 kpc offset from the host galaxy nucleus disfavors a jetted tidal disruption around a supermassive black hole, leaving open a range of stellar-mass and intermediate-mass black holes. However, the rapid subsecond variability of the prompt gamma-ray emission points to a stellar-mass black hole origin as the minimum variability timescale sets an upper limit on the central engine size ≲ 100 M⊙. We adopt a full-disruption micro-TDE model in which the observed delay (∼ 1 day) between the initial soft X-ray emission and the onset of the gamma-ray burst reflects the timescale required for the stellar debris stream to circularize and form a disk with an evacuated polar funnel (along the disk rotation axis, Figure 1), and for a relativistic jet to drill its way through the funnel and escape.

Our model accommodates the early WXT X-ray precursor as powered by the pre-viscous-accretion phase, during which the stellar debris stream has not yet circularized and settled into a disk. During this phase, a small fraction of the bound material can fall directly into the black hole (through low-angular-momentum stream collision products) and launch an early jet without radiation-driven wind suppression (i.e., s = 0). The observed peak precursor luminosity, L X,iso ∼ 1046 erg s−1, is five orders of magnitude below the prompt gamma-ray luminosity, which we attribute to a highly magnetized jet in the MAD state. This luminosity contrast therefore points to a correspondingly lower fallback rate, M¤ fb ≪ 10−5 M⊙ s−1 during the precursor, consistent with only a small fraction (≲ 1%) of the bound debris falling directly onto the black hole and lower radiation efficiency etaX ≲ 0.01.

Figure 4 shows the isotropic-equivalent jet power, Piso ≡ L iso /etarad, as a function of the intrinsic jet power, L jet, and beaming angle, theta b = 1/gammaj, where gammaj is the jet Lorentz factor. Here, we identify the engine configurations that reproduce the peak gamma-ray isotropic equivalent luminosity of GRB 250702 (L gamma,iso ∼ 1051.6 erg s−1) for a range of gamma-ray efficiency, eta gamma (solid and dashed red lines). The allowed parameter space is the wedge with Λ > 2 and eta gamma ≤ 1: the jet must be kink-stable (Λ > 2, cyan line) and powerful enough to produce the observed luminosity (eta gamma ≤ 1, solid red line). The stability and energetics boundaries meet near our minimum power fiducial central engine, marked with the magenta star: it features high-power and strongly beamed jets, L jet ≈ 3 × 1047 erg s−1 and theta b = 0.7◦, respectively. Our model achieves this solution at the peak black hole accretion rate, M¤ acc ∼ 10−7 M⊙ s−1, for a rapidly spinning black hole, a • = 0.9, and fiducial disk wind mass loss slope, s = 0.5. Note that higher black hole mass accretion rates (e.g., for weaker disk-driven winds, s < 0.5) can allow lower spin and/or radiative efficiency values.

Having established the wind-modified viscous accretion model in Section 3, we now compare it to the light curve of GRB 250702. Figure 5 shows the gamma-ray and X-ray isotropic-equivalent luminosity light curves, computed by numerically solving the coupled disk evolution equations (Equations 3–5), with the fallback rate, M¤ fb (t), taken directly from our arepo simulations. We adopt the parameters a • = 0.9 (yielding jet efficiency, etajet ≈ 1 for phi• = 50), radiative efficiency etaX = 0.01 and eta gamma = 1, and wind exponent s = 0.5, accretion radius r acc = 10 r g and initial beaming angle theta b = 0.7◦ which together set the peak isotropic X-ray and gamma-ray luminosity to L X,iso ∼ 1049 erg s−1 and L gamma,iso ∼ 1051 erg s−1, respectively. The horizontal axis shows time since disruption of the star; T0 ≃ 1.5 × 104 s (the black dashed vertical line marking the last Fermi-GBM trigger) is placed at the epoch in the model where a significant fraction of the bound debris has returned to pericenter and the evolution becomes dominated by viscous accretion. The EP-WXT X-ray light curve (green circles) decays as L X,iso ∝ t − alpha with alpha ≈ 3.8, significantly steeper than the intrinsic engine decay L jet ∝ t −2 predicted by our wind-modified accretion model for s = 0.5 (Eq. 8). We propose that this steepening reflects a time-dependent jet beaming geometry. In our model, at times T0 < t < ttransition ∼ 4 × 104 s, the jet opening angle widens linearly with time from theta b,i = 0.7◦ to theta b,f = 2.8◦, reducing the degree of beaming and luminosity by a factor, (theta b,f /theta b,i) 2 ≈ 10. Combined with the intrinsic engine decay scaling from wind-modified accretion, L jet ∝ t −2, this geometric widening produces the steep L X,iso ∝ L jet /theta b2 ∝ t −4 observed during the EP-WXT phase. Such widening can arise from several physical mechanisms, including the reconnection-driven dissipation of the toroidal magnetic field, which weakens the magnetic hoop stress that helps to to collimate the jets, and the lateral expansion at the comoving sound speed once the cocoon disperses. In the latter process, just before exiting out of the confining envelope, jet beaming angle is set by the relativistic beaming angle, theta b,i = 1/gammaj. After the jets break out, they expand sideways to a larger opening angle, which sets the four times larger beaming angle, theta b,f ≡ theta j,f = 4/gammaj, as seen in the simulations of idealized GRB jets. A more detailed physical treatment is beyond the scope of this work; here we use this prescription to demonstrate that the observed L X,iso ∝ t −4 decline is naturally reproduced by opening-angle growth combined with the intrinsic L jet ∝ t −2 engine decay from our wind-modified accretion model.

After the beaming angle saturates at theta b ∼ 2.8◦, the beaming correction becomes constant and the observed light curve tracks the intrinsic L jet ∝ t −2 decay from wind-modified viscous accretion, in agreement with the EP-FXT data (blue diamonds) in Figure 5. The FXT emission remains detectable for ∼ weeks, far exceeding the viscous timescale (Equation 2) evaluated at the initial circularization radius r c ≃ 1R ⊙ for the deep encounter (beta = 5), which is t v ≈ 4 hr for alphav = 0.1, h = 0.5. At times t ≫ t v, the disk enters a spreading phase in which the accretion rate onto the black hole decays as M¤ acc ∝ t −2 while the disk radius grows as r d ∝ t 2/3 (Equation 6). By t ∼ 1–3 weeks post-disruption, the disk has spread by r d /r c ∼ (t/tv) 2/3 ∼ 10–25, prolonging the viscous time at the outer edge, t v (r d) = tv (r c) (r d /r c) 3/2, to the observed duration of the FXT emission. The long-lived FXT X-ray emission is therefore powered by the drainage of the viscously spreading disk, with the emission timescale set by the viscous accretion time at the disk’s outer edge.

We have studied jet formation and escape in micro-TDEs as a channel for powering ultra-long GRB s. Using 3D arepo hydrodynamic simulations of a 1 M⊙ star disrupted by a 10 M⊙ black hole, we showed that within ≲ 1 day the debris settles into a disk with a low-density polar funnel along the angular momentum direction of the disk, with a steep rho ∝ r −2 polar profile and half-opening angle theta f ≈ 15◦. The two encounter geometries we considered (beta = 1.4 and 5) both produce well-defined funnels, but the grazing encounter (beta = 1.4) develops a thinner, rotationally supported extended disk. Deeper encounter (beta = 5) instead produces more compact, thicker inner disk (h ≃ 0.5), yielding a higher accretion rate and correspondingly a higher jet power. The funnel geometry, combined with the steep r −2 polar profile, provides ideal conditions for jet collimation and stable propagation and escape for both encounters. Coupling our hydrodynamic simulations to a semi-analytic wind-modified disk evolution model, we find that jetted micro-TDEs can match the energetics and timescales of ultra-long GRB s: their hours-to-days viscous accretion times set the engine duration, and a highly collimated jet from a rapidly spinning black hole delivers the required jet power (L jet,min ≳ 1047 erg s−1). These results establish jetted micro-TDEs as a physically motivated ULGRB engine. Applying this framework to GRB 250702B, we account for all three of its observed phases (the soft X-ray precursor, the several-hour prompt gamma-ray emission, and the weeks-long X-ray decay) as a natural progression in the aftermath of a full disruption of a Sun-like companion star. We propose a stream-fed origin for the X-ray precursor and quantitatively reproduce the X-ray light curve, without invoking repeated stripping, multiple engines, or external-shock afterglow physics. Our main results are as follows. We propose the following three emission phases: (1) the X-ray precursor traces pre-disk stream-fed accretion, with only a small fraction (≲ 1%) of the debris reaching the black hole before disk formation; (2) once the disk forms and clears out the polar direction, a narrowly collimated jet with a beaming angle theta b ≲ 1◦ launched by a spinning black hole in the MAD state propagates through the funnel and powers the prompt burst, energetic enough to match the observed L gamma,iso ∼ 1051 erg s−1; and (3) the subsequent X-ray emission arises from wind-modified viscous drainage of the disk (s = 0.5), with the engine power declining as L jet ∝ t −2 while gradual jet widening from theta b = 0.7◦ to 2.8◦ produces the steep L X,iso ∝ t −4 decline observed by EP-WXT. At later times, the beaming angle saturates at 2.8◦ and the intrinsic t −2 decay is recovered, matching the temporal slope of the long-lived EP-FXT tail, whose weeks-long duration follows from the viscous disk spreading (r d ∝ t 2/3). The kink-instability analysis further establishes that the inferred micro-TDE jet power for GRB 250702B sits at or above the jet stability threshold, with the stability criterion (Λ > 2) satisfied at L jet ≳ 1047 erg s−1 for theta b ≲ 1◦ in the deep encounter case and at much lower jet powers for the wider jet opening angles relevant to the FXT phase. The multi-phase evolution therefore proceeds entirely within the stable regime of the Λ–L jet plane, confirming that the polar funnel geometry established during the disruption is favorable for launching stable, well-collimated relativistic jets throughout the emission window. Notably, the r −2 polar profile sits precisely at the critical slope alpha = 2 that separates jet collimation from de-collimation. Because alpha = 2 implies rhoa z2h = constant, Λ is then independent of jet-head distance, so stability holds at all radii without requiring breakout at a particular height.

Several progenitor scenarios have been proposed for GRB 250702B, ranging from a single engine to distinct processes driving each phase: tidal disruption of a white dwarf by an intermediate-mass black hole, an external-shock afterglow origin for the late X-rays, self-regulated collapse of a supergiant star, and micro- or milli-TDEs of main-sequence stars. We compare our jetted micro-TDE model to each of these previous studies below, focusing on the physical mechanisms governing the different phases of GRB 250702B. At the highly super-Eddington rates expected in micro-TDEs, the geometrically thick, advection-dominated flow ejects most of the inflowing mass as winds before it reaches the black hole, so that only a small fraction (≲ 1%) of the bound debris is ultimately accreted. A key feature of our model relative to previous TDE interpretations of GRB 250702B is that the accretion rate is evolved self-consistently with a wind-loss prescription: M¤ acc ∝ (r acc /r d) s with s = 0.5. The resulting suppressed accretion rate, M¤ acc ∼ 10−7 M⊙ s−1, sets the scale of the peak luminosity: considering disk accretion alone results in isotropic X-ray luminosities of up to ∼ 1046 erg s−1 for s = 0.5, several orders of magnitude below the isotropic gamma-ray luminosity of GRB 250702B. Reproducing the prompt isotropic equivalent gamma-ray luminosity (L gamma,iso ∼ 1051 erg s−1) at these suppressed accretion rates therefore requires beamed jet emission rather than quasi-isotropic disk radiation: a narrowly collimated jet (theta b ≲ 1◦, beaming factor fb−1 ∼ 103 –104) launched by a spinning black hole via the Blandford–Znajek mechanism. In our model, the ∼hours-long prompt gamma-ray duration is set by the viscous timescale of an extended (∼ R ⊙) micro-TDE disk (Eq. 2), whereas collapsar scenarios invoke extended envelope fallback and the intermediate-mass black hole scenarios tie the timescale to the larger black hole mass through the fallback time of a disrupted white dwarf or main-sequence star. A further distinction of our model is that we incorporate viscous spreading of the disk, which regulates the late-time emission by extending the engine lifetime: the growth of the disk (r d ∝ t 2/3), combined with the wind-modified accretion (M¤ acc ∝ t −2), extends the local viscous time at the disk edge to weeks. The origin of the quasi-regular recurrence of the prompt episodes remains uncertain: An (2025) and Sato et al. (2026) attribute it to Lense–Thirring precession of a compact, misaligned inner torus at r ∼ 102 –103 r g, and Zhang et al. (2026) to the collapse of successive stellar layers. In our model the debris disk is radially extended (∼ 105 r g), and its large angular momentum content makes global Lense–Thirring precession inefficient, with precession periods far exceeding the engine lifetime; we therefore do not expect global disk precession under spin–disk misalignment. Instead, the recurrence may reflect ongoing debris fallback perturbing the disk and modulating the jet. We regard this as a qualitative possibility; confirming it requires coupled GRMHD modeling of the jet–disk coupling, which we defer to future work. Several aspects of our model warrant further investigation. Dedicated GRMHD simulations will be needed to determine under what conditions a MAD state is established, and to self-consistently determine the resulting accretion rate and jet power for a given black hole spin. Likewise, the linear jet-widening prescription, theta j ∝ t, adopted to reproduce the t −4 decay observed by WXT is phenomenological. This assumption can be tested with GRMHD simulations of jet propagation through the micro-TDE debris envelope, which would self-consistently capture the mechanisms governing the lateral jet expansion. In particular, the jet opening angle may evolve through hydrodynamic expansion on the sound-crossing timescale and/or through reconnection-driven dissipation of the magnetic field responsible for collimation. Finally, while the large-scale jet propagation considered here is not expected to depend strongly on the orientation of the black-hole spin, the effects of spin–disk misalignment on the inner accretion flow and jet launching warrant further investigation. In particular, strongly tilted systems (≳ 60◦) may be less likely to reach the MAD state, potentially affecting the accretion rate and jet power. Future GRMHD simulations could therefore test how the degree of misalignment influences MAD formation and the resulting jet properties. Such simulations would also allow us to explore the dynamics of bent jets propagating through the asymmetric debris environment and whether disk–spin misalignment introduces additional time variability into the jet emission. These effects could provide further observational signatures of the micro-TDE scenario beyond those considered in this work.

Improvements to AI Systems:

1. Multi-Phase Transient Classification: Train an AI system to classify gamma-ray bursts (GRBs) into distinct engine scenarios (collapsar, supermassive TDE, micro-TDE) by analyzing multi-wavelength light curves across three phases (precursor, prompt, afterglow) simultaneously, rather than treating each phase independently. The improved system can automatically identify micro-TDE signatures—such as day-long soft X-ray precursors, multi-hour prompt emission with subsecond variability, and weeks-long X-ray tails with specific temporal slopes (t−2 to t−4)—and flag candidates for follow-up observations.

2. Jet Stability Prediction from Density Profiles: Implement a physics-informed neural network that takes 3D density profiles of debris envelopes (from hydrodynamic simulations) as input and predicts jet propagation success, kink-instability thresholds, and optimal jet parameters (power, Lorentz factor, opening angle) using the analytic stability framework (Λ > 2 criterion). The improved system can rapidly evaluate thousands of simulated TDE configurations (varying black hole mass, spin, penetration factor, stellar type) to determine which produce stable, detectable jets, accelerating progenitor identification for future ULGRB detections.

3. Self-Consistent Accretion-Disk Evolution Modeling: Develop an AI surrogate model that couples fallback rates from hydrodynamic simulations with wind-modified viscous accretion physics (including disk spreading, mass-loss slope s, and time-dependent viscous timescales) to predict accretion rates and jet luminosities over weeks. The improved system can invert observed X-ray and gamma-ray light curves to infer black hole spin (a•), beaming angles (thetab), wind efficiency (s), and accretion radius—enabling rapid parameter estimation for transient events without requiring full numerical simulations.

4. Multi-Phase Light Curve Fitting with Geometric Evolution: Build an AI system that jointly fits prompt gamma-ray and X-ray light curves using a model that includes time-dependent jet beaming (e.g., linear widening thetab ∝ t) and intrinsic engine decay (Ljet ∝ t−2). The improved system can automatically distinguish between competing explanations for steep decays (e.g., beaming widening vs. external shock vs. precession) by comparing goodness-of-fit across physical models, reducing false positives in transient classification.

5. Precursor Detection and Engine Sizing: Train a deep learning model on simulated micro-TDE precursor emission (stream-fed accretion, low luminosity ∼1046 erg s−1) to predict the subsequent prompt burst properties (delay time, peak luminosity, duration) from precursor characteristics alone. The improved system can issue early warnings for ULGRB candidates and estimate central engine parameters (black hole mass, spin) hours before the main burst, enabling rapid multi-wavelength follow-up coordination.

6. Cross-Phase Temporal-Spectral Correlation: Implement a transformer-based architecture that ingests time-resolved spectral data (photon index, luminosity) across all phases and learns to map spectral evolution (e.g., hard-to-soft transitions) to physical processes (stream-fed accretion → jet breakout → disk drainage). The improved system can automatically identify the transition point between prompt and afterglow phases, estimate jet breakout times, and validate or reject micro-TDE models for individual events.

7. Simulation-to-Observation Transfer Learning: Create an AI system pre-trained on 3D arepo simulations of micro-TDEs (density, velocity, angular momentum distributions) and fine-tuned on observed GRB light curves to generate synthetic multi-wavelength observations for arbitrary progenitor parameters. The improved system can produce realistic mock datasets for survey design, optimize observing strategies for next-generation missions (e.g., Einstein Probe, SVOM), and quantify detection biases for ULGRB populations.

8. Uncertainty-Aware Parameter Inference: Develop a Bayesian neural network that predicts posterior distributions for black hole spin, beaming angle, wind slope, and disk radius from observed light curves, incorporating systematic uncertainties from jet stability criteria and radiative efficiency models. The improved system can provide robust error bars on engine parameters and flag events where multiple degenerate solutions exist (e.g., low spin with high accretion vs. high spin with low accretion), guiding follow-up theoretical work.

The longest detected gamma-ray burst, GRB250702B, exhibited seven hours of prompt gamma-ray emission, preceded by a soft X-ray precursor (about1 day earlier) and followed by a weeks-long fading X-ray tail. Lacking an established progenitor for all three phases, we propose that this ultra-long GRB (ULGRB) is powered by a jetted micro-tidal disruption event (micro-TDE), in which a spinning stellar-mass black hole (BH) disrupts a Sun-like star and launches a relativistic jet via the Blandford-Znajek mechanism. Micro-TDE debris disks have hours-to-days viscous timescales, naturally explaining ULGRB durations. Using 3D hydrodynamic AREPO simulations of a 1,M star disrupted by a 10,M BH, we show that within about1 day the debris forms a quasi-steady envelope with a low-density polar funnel (rho proportional to r-2, half-opening angle about15). Applying an analytic jet-stability framework to these profiles, we find that the r-2 funnel keeps the jet below the kink-instability threshold, enabling stable propagation and breakout for jet powers, L jet 10 47 erg s-1. We attribute the X-ray precursor to pre-disk stream-fed accretion; the prompt GRB to a tightly beamed jet (theta b 1, L gamma, iso about10 51 erg s-1) escaping the funnel, launched by a rapidly spinning BH (a about0.9); and the weeks-long X-ray decline to disk-wind mass loss (L jet proportional to t-2) combined with jet widening (theta b proportional to t, initially steepening the decay to L X,iso proportional to L jet/theta b squared proportional to t-4). Our model reproduces the multi-phase evolution of GRB250702B and establishes jetted micro-TDEs as a physically motivated ULGRB engine.