Backwards Gamma-Ray Bursts: Searching for Exploding Primordial Black Holes in Short-Duration GRB Catalogs

TL;DR

The paper tackles the question of whether terminal primordial black hole (PBH) evaporation can be detected as a distinctive backwards burst in short GRB data. It develops a forward-modeling framework that folds Hawking-emitted photon spectra through detector responses (via BlackHawk and PYTHIA) and compares the PBH template to conventional GRB pulse forms using AIC and BIC. An analysis of 35 Swift short GRBs with no detected afterglows shows that all events favor standard FRED or ERCA shapes over the PBH template, yielding no PBH candidates and an empirical local explosion-rate bound of $R_{ m PBH} \lesssim 10^5\ \mathrm{pc}^{-3}\ \mathrm{yr}^{-1}$. The work establishes a robust, scalable methodology for template-based PBH searches across larger, multi-instrument catalogs, enabling deeper probes into quantum gravity and early-Universe physics.

Abstract

We present a systematic search for signatures of terminal black-hole evaporation in short gamma-ray burst (sGRB) catalogs. An exploding primordial black hole (PBH) undergoing final-stage Hawking radiation is predicted to produce a distinctive "backwards burst"-a very short, spectrally hard transient with monotonically increasing flux and little or no longer-wavelength afterglow. We develop a forward-modeling framework that directly compares theoretical PBH evaporation light curves, computed with full Standard Model particle content and detector response folding, against empirical GRB pulse templates. Analyzing 39 well-characterized Swift sGRBs with non-detected or extremely faint afterglows, we find that all events exhibit fast-rise, slow-decay temporal profiles inconsistent with the PBH prediction. Model comparison via Akaike and Bayesian information criteria decisively favors conventional FRED or ERCA fits over the PBH template for every burst. No candidates for terminal PBH evaporation are identified. The null result yields an upper bound on the local PBH explosion rate density $R_{\mathrm{PBH}} \lesssim 10^5~\mathrm{pc}^{-3}~\mathrm{yr}^{-1}$, comparable to constraints from dedicated TeV $γ$-ray searches. Our methodology establishes a robust template-matching approach that can be scaled to larger multi-instrument catalogs, providing a foundation for future searches targeting this unique signature of quantum gravity and early-Universe physics.

Figures (4)

• Figure 1: Representative short-duration GRB light-curve fits. Observed Swift/BAT count rates (solid blue) are fitted with four models: the physically motivated PBH evaporation template (dashed green), a Fast-Rise–Exponential-Decay (FRED) profile (dashed orange), an Exponential-Rise–Constant-Afterglow (ERCA) form (dashed red), and a PBH with Constant-Afterglow (PBHCA) curve (dashed purple). For all four examples the FRED or ERCA models reproduce the observed pulse asymmetry and extended decay far better than the PBH template, whose universal monotonic brightening shape cannot mimic the data. Residuals (bottom panels) highlight the systematic deviation of the PBH model, particularly its underprediction of the rising phase and overprediction of the decay tail.
• Figure 2: Comparison of light curves for the GRB events in Table 1 observed by Swift/BAT and Fermi/GBM. Four GRB events detected by both Swift/BAT and Fermi/GBM at different energy bin are compared and fitted with four models.
• Figure 3: Analysis of GRB events potentially close to Earth ( Fermi data). Four GRB events identified by the IPN as compatible with a location close to Earth are modeled. The estimated minimum distance from Earth for these events is less than $10^{14}$cm.
• Figure 4: sGRB light-curve analysis summary. For each GRB we report the best-fit parameters of the FRED and ERCA models, the AIC and BIC values for all four models (FRED, PBH, PBHCA, ERCA), and the preferred model according to AIC and BIC.