Ultra-long MeV transient from a relativistic jet: a tidal disruption event candidate

TL;DR

This work reports GRB 250702D/B/E (DBE), three MeV transients lasting >3 hours with overlapping sky positions, interpreted as a single relativistic jet from a tidal disruption event. By combining Fermi GBM and LAT data with Swift/XRT and NuSTAR observations, the authors derive non-thermal spectra with photon indices around 1.5–1.64, find no strong high-energy cutoff within 10 keV–40 MeV, and infer a LAT-implied cutoff between 10–100 MeV; the X-ray afterglow declines steeply, challenging classical GRB models. The emission is inconsistent with the Amati and Yonetoku relations for long GRBs and aligns with a jetted TDE interpretation, with synchrotron radiation from sub-TeV electrons as the likely mechanism. Jet properties are constrained to a bulk Lorentz factor of roughly 10–40 and a distance of 100 Mpc–2 Gpc, placing DBE in the relativistic TDE family and marking the first MeV-emitting member of this class.

Abstract

On July 2, 2025, the Gamma-ray Burst Monitor (GBM) onboard the Fermi Gamma-ray space telescope detected three short-duration MeV transients with overlapping sky locations. These events, named as GRB 250702D, B, and E (collectively referred to as DBE), triggered the detector with delays of approximately 1-2 hours between each burst. Follow-up observations of this unusually long MeV transient (lasting >3 hours) by the Neil Gehrels Swift Observatory and the Nuclear Spectroscopic Telescope Array over a period of 10 days revealed a steep temporal decline in soft X-rays ($\propto t^{-1.9 \pm 0.1}$). The time-averaged spectra during the outbursts are well described by a single power law $dN_γ/dE \propto E^{-1.5}$, while upper limits above 100 MeV imply a spectral cutoff between 10 MeV and 100 MeV. Using standard gamma-ray transparency arguments, we derive a lower limit on the bulk Lorentz factor. Combined with the steep decline in X-rays, these constraints point to a relativistic jet origin. The properties of DBE are inconsistent with established GRB spectral-energy correlations, disfavoring classical long GRB progenitors. Instead, the basic characteristics of DBE resemble those of previously reported jetted tidal disruption events (TDEs), though alternative progenitor channels cannot be excluded. In the relativistic TDE scenario, DBE is the first one with detected MeV gamma-ray emission. We argue that the observed emission is most likely produced by synchrotron radiation from sub-TeV electrons.

Figures (8)

• Figure 1: Upper panel: Background-subtracted $\rm 5\,s$-binned light curve of the hard X-rays/MeV emission detected by Fermi/GBM ($50-900$ keV). Gray shaded regions indicate time bins during which the source was occulted by the Earth. The yellow shaded region marks $1 \sigma$ noise level in the light curve. Lower panel: Light curve of the soft X-ray counterpart obtained from time-resolved spectral analysis of Swift/XRT (0.3–10 keV, yellow circles). The flux at $0.3-10$ keV at 10 days (pink square) was estimated by extrapolating the spectral model fitted to the NuSTAR data (fitted in the $3-30$ keV range).
• Figure 2: Spectral energy distributions of events D, B, and E. The derived data points for the time-integrated spectra of D, B, and E, along with the best-fit power-law models (with $1\sigma$ uncertainties) are shown using the blue, orange, and green symbols and shaded regions, respectively. The dashed orange and green lines represent average cutoff power-law models for the time-integrated spectra of events B and E. Upper limits between 100 MeV and 1 GeV ($3\sigma$) are shown as inverted triangles. The combined D and B spectrum above 200 keV, along with its best-fit model ($1\sigma$), is shown using red triangles and a red shaded region. The corresponding spectral slopes ($2-\alpha$) are also indicated.
• Figure 3: Lower limit on the bulk Lorentz factor of the jet as a function of the isotropic-equivalent luminosity of the D and B emission episodes. The color bar indicates the corresponding luminosity distance to DBE. For comparison, we show the luminosity (brightest hard X-ray outbursts) measured for Swift J2058+05 Cenko2012Pasham2015, Swift J1112+82 Brown2015, AT 2022cmc Andreoni2022Pasham2023, and Swift J1644+57 Burrows2011 as dashed blue, dotted green, and solid red vertical lines.
• Figure 4: Light curves for two detectors (Detector nb: top row; Detector b1: bottom row) across three different events (Event 1: left column; Event 2: middle column; Event 3: right column). The estimated background is plotted using three different orbits. For the 'nb' data, the energy range is cut between 40 keV and 900 keV, while for 'b1' data, the cut is between 200 keV and 40 MeV. Each panel shows the light curve integrated over energy for a specific detector and event combination.
• Figure 5: Left: NuSTAR FPMA and FPMB (blue and red, respectively) spectra fitted with a power-law model. Right: The confidence contours (68%, 95%, and 99%) of the photon index versus the $3-30$ keV flux.