A Universal 1.5 GeV Gamma-Ray Line in Active Galactic Nuclei

Abstract

We report the detection of a gamma-ray spectral line at approximately 1.5 GeV in three active galactic nuclei (AGN) using 17 years of Fermi-LAT observations. The sample includes both blazars (with relativistic jets directed toward Earth) and a radio galaxy (with a misaligned jet, free from significant beaming effects). The line is detected with local significances of $\sim$4.1$σ$, $\sim$3.9$σ$, and $\sim$2.8$σ$ in the individual sources. A joint likelihood analysis yields a combined test statistic TS $\simeq$ 57.77, corresponding to a significance well above 5$σ$. The line flux remains stable over the full observation period, in contrast to the variable continuum emission from the AGN. The appearance of an identical spectral feature in astrophysically distinct environments is difficult to reconcile with standard jet-based emission mechanisms. While a conventional astrophysical explanation remains elusive, the signal's characteristics are consistent with predictions for dark matter annihilation. This finding motivates further investigation into the nature of this spectral feature and its possible connection to particle dark matter.

Figures (8)

• Figure 1: Fermi-LAT spectral energy distribution (SED) with a spectral line signal in the 100 MeV–1 TeV energy range.(a) Left panel: the LAT spectrum of 4FGL J0250.2$-$8224 (a blazar with its relativistic jet aligned toward Earth), using 17 years of data from 2008 August 4 to 2025 August 4. The green colour map indicates the $\Delta$LogLikelihood of each SED point. Inset (a1) shows the SED fitted with a pure power-law model (null hypothesis, red dashed line) and with a power-law plus Gaussian model (signal hypothesis, blue dash-dotted line). Blue points and grey downward arrows represent SED points with TS values above and below 4/9, respectively. The blue dotted line denotes the Gaussian excess component. The light‑blue band marks the 95% confidence interval of the continuum baseline. Inset (a2) displays the model‑fitted count spectrum. All three Gaussian parameters ($N_0$, $E_{\rm signal}$, $\sigma_{\rm signal}$) were kept free and scanned over a wide range (e.g., the full energy interval from 100 MeV to 1 TeV) in pyLikelihood analysis. Black dots and the solid black line show the total count spectrum and the total model from the pyLikelihood analysis. Contributions of individual background sources in the model are drawn as light‑grey lines. Red dot‑dashed lines correspond to model‑fitted count spectra of the aimed sources. The blue solid line and the red dashed line represent the power‑law model for the continuum and the Gaussian model for the line signal, respectively. The corresponding spectral‑fit parameters are listed in Table \ref{['tab1']}. (b) Right panel: the LAT spectrum of 4FGL J2329.7$-$2118 (a radio galaxy with a misaligned jet and no significant beaming effects).
• Figure 2: Schematic diagram illustrating the observational geometries for comparison. This schematic compares the observational geometries of a blazar and a radio galaxy to investigate the origin of a 1.5 GeV gamma‑ray line signal. Blazar (upper left): Its relativistic jet is nearly aligned with the observer's line of sight (red arrow). The jet emission is Doppler‑boosted and dominates the observed gamma‑ray flux. Radio galaxy (lower right): Its relativistic jet is inclined at a large angle to the observer's line of sight. The jet emission shows no significant Doppler boosting. Key insight: An identical 1.5 GeV gamma-ray line (red beams and green unbeamed) comes from the central (e.g., dark‑matter halo, Light Bluish Gray) of both systems, independent of jet orientation. The detection of this line in two distinct astrophysical environments (beamed blazar and unbeamed radio galaxy) excludes an origin related to the jet, supporting the hypothesis that the signal originates from dark‑matter processes in the galaxy central dark matter halo.
• Figure 3: Spectral Fitting Parameters for the line Signal.
• Figure 4: Temporal stability and specificity of the 1.5 GeV line. The energy range of 100 MeV to 1 TeV is adopted for the data from August 4, 2008, to August 4, 2025. The upper limit of the light curve is indicated by a gray downward arrow for data with a TS value below 9. Light curves of the 1.5 GeV line-like signal flux and the total continuum flux for the primary blazars, where (a): 4FGL J0250.2$-$8224, (b): 4FGL J2329.7$-$2118. The flux of the 1.5 GeV line (red data points, left y-axis) remains consistent with a constant value (blue dashed line; $\chi^2$/d.o.f. = 0.81/1.17 for a constant fit respectively) over the 17-year observation period. In stark contrast, the total gamma-ray flux above 100 MeV (grey curve and shaded area, right y-axis), representing the blazar's jet activity, exhibits characteristic, order-of-magnitude variability. The stability of the line flux, despite dramatic changes in the blazar's central engine output, decouples its origin from standard jet processes.
• Figure 5: Model-fitted Count spectra.(a): 4FGL J0749.6$+$1324 and (b): 4FGL J0357.0$-$4955. It covers the energy range of 100 MeV to 1 TeV over a 17-year period - spanning from August 4, 2008, to August 4, 2025. All three Gaussian parameters ($N_0$, $E_{\rm signal}$, $\sigma_{\rm signal}$) were kept free and scanned over a wide range (e.g., the full energy interval from 100 MeV to 1 TeV) in pylikelihood analysis. Black dots and the solid black line show the total count spectrum and the total model from the pyLikelihood analysis. Contributions of individual background sources in the model are drawn as light‑grey lines. Red dot‑dashed lines correspond to model‑fitted count spectra of the aimed sources. The blue solid line and the red dashed line represent the power-law/log-parabola model for the continuum and the Gaussian model for the line signal, respectively. The corresponding spectral‑fit parameters are listed in Table \ref{['tab1']} and Appendix Table \ref{['TabA2']}.