Abstract
Analysis of γ-rays in solar flares has suggested a distinct continuum component dominating at megaelectronvolt energies, which differs from the well-studied X-ray continuum produced by flare-accelerated electrons, with spectra steeply falling with energy. The origin, precise spatial location and extent of this mysterious megaelectronvolt component have been unknown up to now. If it is produced by bremsstrahlung, such a γ-ray component requires an unusual population of electrons peaked at a few million electron volts. Here we report a joint study of this megaelectronvolt-peaked electron population in the 2017 September 10 solar flare with Fermi megaelectronvolt γ-ray data and spatially resolved microwave imaging spectroscopy data obtained by the Expanded Owens Valley Solar Array. We demonstrate that the microwave spectrum from the megaelectronvolt-peaked distribution has a distinctly different shape from that produced by the electrons with a falling energy spectrum. We inspected microwave maps of the flare and identified an evolving area where the measured microwave spectra matched the theoretically expected ones for the megaelectronvolt-peaked population, thus pinpointing the site where this megaelectronvolt component resides. The locations are in a coronal volume adjacent to the region where prominent release of magnetic energy and bulk electron acceleration were detected. The results imply that transport effects play a key role in forming this population of high-energy particles, which is crucial for building a complete picture of the multifaceted solar flare phenomena.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$119.00 per year
only $9.92 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
All original EOVSA data are maintained in the EOVSA website at http://www.ovsa.njit.edu. Original EOVSA data used for this study are available at http://www.ovsa.njit.edu/fits/IDB/20170910/IDB20170910155625/. Fully processed EOVSA spectral imaging data in IDL save format can be downloaded from http://ovsa.njit.edu/publications/fleishman_ea_science_2019/data/ for the 4-min peak phase and https://ovsa.njit.edu/publications/fleishman_ea_natastro_2025/data/ for a longer, 18-min, duration. The Fermi/GBM spectrum data file (.PHA) can be accessed at https://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/daily/2017/09/10/current/ and the Fermi/GBM response matrix for the event of interest can be accessed at http://hesperia.gsfc.nasa.gov/fermi/gbm/rsp/. Alternatively, data files can be retrieved through OSPEX.
Code availability
All codes we use in this study are based on publicly available software packages: GSFIT is available in the community-contributed SolarSoftWare repository, under the packages category, at https://github.com/Gelu-Nita/GSFIT; the open-source MCMC code is documented in ref. 25; OSPEX spectral analysis software documentation is available at https://hesperia.gsfc.nasa.gov/ssw/packages/spex/doc/ospex_explanation.htm.
References
Yokoyama, T. & Shibata, K. Magnetic reconnection as the origin of X-ray jets and Hα surges on the Sun. Nature 375, 42–44 (1995).
Fleishman, G. D. et al. Decay of the coronal magnetic field can release sufficient energy to power a solar flare. Science 367, 278–280 (2020).
Krucker, S. et al. Measurements of the coronal acceleration region of a solar flare. Astrophys. J. 714, 1108–1119 (2010).
Krucker, S. & Battaglia, M. Particle densities within the acceleration region of a solar flare. Astrophys. J. 780, 107 (2014).
Ackermann, M. et al. Fermi detection of γ-ray emission from the M2 soft X-ray flare on 2010 June 12. Astrophys. J. 745, 144 (2012).
Kurt, V. G., Yushkov, B. Y., Galkin, V. I., Kudela, K. & Kashapova, L. K. CORONAS-F observation of gamma-ray emission from the solar flare on 2003 October 29. Nat. Astron. 56, 102–112 (2017).
Murphy, R. & Share, G. Compton scattering of deexcitation-line and continuum gamma rays in solar flares. Sol. Phys. 293, 163 (2018).
Lysenko, A. L., Anfinogentov, S. A., Svinkin, D. S., Frederiks, D. D. & Fleishman, G. D. Gamma-ray emission from the impulsive phase of the 2017 September 6 X9.3 flare. Astrophys. J. 877, 145 (2019).
Share, G. H., Murphy, R. J., Dennis, B. R. & Finke, J. D. Solar gamma-ray evidence for a distinct population of >1 MeV flare-accelerated electrons. Astrophys. J. 981, 11 (2025).
Fleishman, G. D., Nita, G. M., Chen, B., Yu, S. & Gary, D. E. Solar flare accelerates nearly all electrons in a large coronal volume. Nature 606, 674–677 (2022).
Dulk, G. A. Radio emission from the sun and stars. Annu. Rev. Astron. Astrophys. 23, 169–224 (1985).
Fleishman, G. D. & Toptygin, I. N. Cosmic Electrodynamics (Springer, 2013).
Kozlovsky, B., Lingenfelter, R. E. & Ramaty, R. Positrons from accelerated particle interactions. Astrophys. J. 316, 801–818 (1987).
Kozlovsky, B., Murphy, R. J. & Share, G. H. Positron-emitter production in solar flares from 3He reactions. Astrophys. J. 604, 892–899 (2004).
Silva, A. V. R. et al. Evidence that synchrotron emission from nonthermal electrons produces the increasing submillimeter spectral component in solar flares. Sol. Phys. 245, 311–326 (2007).
Omodei, N., Pesce-Rollins, M., Longo, F., Allafort, A. & Krucker, S. Fermi-LAT observations of the 2017 September 10 solar flare. Astrophys. J. Lett. 865, 7 (2018).
Ajello, M. et al. First Fermi-LAT solar flare catalog. Astrophys. J. Suppl. Ser. 252, 13 (2021).
Fleishman, G. D., Altyntsev, A. T. & Meshalkina, N. S. Microwave signature of relativistic positrons in solar flares. Publ. Astron. Soc. Jpn 65, S7 (2013).
Dreicer, H. Electron and ion runaway in a fully ionized gas. I. Phys. Rev. 115, 238–249 (1959).
Fleishman, G. D. et al. Revealing the evolution of non-thermal electrons in solar flares using 3D modeling. Astrophys. J. 859, 17 (2018).
Jejčič, S., Kleint, L. & Heinzel, P. High-density off-limb flare loops observed by SDO. Astrophys. J. 867, 134 (2018).
Melrose, D. B. & Brown, J. C. Precipitation in trap models for solar hard X-ray bursts. Mon. Not. R. Astron. Soc. 176, 15–30 (1976).
Schwartz, R. A. et al. RHESSI data analysis software: rationale and methods. Sol. Phys. 210, 165–191 (2002).
Gary, D. E. et al. Microwave and hard X-ray observations of the 2017 September 10 solar limb flare. Astrophys. J. 863, 83 (2018).
Foreman-Mackey, D., Hogg, D. W., Lang, D. & Goodman, J. emcee: the MCMC Hammer. Publ. Astron. Soc. Pac. 125, 306–312 (2013).
Kontar, E. P. et al. Deducing electron properties from hard X-ray observations. Space Sci. Rev. 159, 301–355 (2011).
Holman, G. D. Acceleration of runaway electrons and Joule heating in solar flares. Astrophys. J. 293, 584–594 (1985).
Aschwanden, M. J. Particle acceleration and kinematics in solar flares—a synthesis of recent observations and theoretical concepts (invited review). Space Sci. Rev. 101, 1–227 (2002).
Acknowledgements
The EOVSA was designed and built and is now operated by NJIT as a community facility. EOVSA operations are supported by NSF grant AGS-2436999 and NASA grant 80NSSC20K0026 to NJIT. We thank G. Share for discussion of several aspects of this study and A. Sirenko for comments. This work was supported in part by NSF grants AGS-2334931 (B.C., S.Y. and I.O.) and AGS-2425102 (G.D.F. and G.M.N.), NASA grants 80NSSC24K1242 (B.C. and S.Y.) and 80NSSC23K0090 (G.D.F. and G.M.N.) to New Jersey Institute of Technology (NJIT) and NASA grant NH240B72A to Triad National Security LLC via subcontract 89233218CNA000001/C5509 to NJIT (B.C. and I.O.).
Author information
Authors and Affiliations
Contributions
G.D.F. performed modelling of the microwave emission for various particle distributions including the megaelectronvolt-peaked component and wrote the draft paper; I.O. analysed Fermi γ-ray data including spectral model fitting and performed modelling of the γ-ray emission taking into account electron–electron bremsstrahlung; G.M.N. performed spectral fitting of the rising microwave spectra and comparative analysis of the steep and shallow spectra; B.C. obtained self-calibration microwave data for the 18-min time range employed in this study and applied the MCMC methodology to constrain the minimal energy in the electron spectra; S.Y. performed time-domain correlation analysis of the microwave and γ-ray emissions; D.E.G. led the construction and commissioning of the EOVSA, developed the observational strategy and calibration for microwave spectroscopy and provided raw input data for the self-calibration. All authors discussed the interpretation of the data, contributed scientific results and helped prepare the paper.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Astronomy thanks Mykola Gordovskyy and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 MCMC-derived probability distributions of the fit parameters for pixel P1.
This pixel is located in ROI 2 at x = 971.0″ and y = − 164″ outlined by small blue triangle in Fig. 1. The upper right panel shows the microwave spectrum from this pixel by the open circles (mean values) and the error bars at ± 1σ level. Blue curves display a set of trial theoretical spectra employed within MCMC fitting. The free parameters of the fitting included (from top to bottom): the magnetic field B, the thermal number density nth, the nonthermal number density nnth, the spectral index δ, the viewing angle θ, and the minimum energy \({E}_{\min }\) in the power-law electron spectrum. Individual panels display solid black horizontal/vertical lines that indicate the best-fit values from the GSFIT minimization reported in10. The median values of the MCMC probability distributions are shown with the dotted blue horizontal/vertical lines. Dashed black lines in the histograms along the diagonal indicate ± 1-σ standard deviation of a given parameter. Correlations between all possible pairs of parameters are shown as two-dimensional histograms of the probability distributions in off-diagonal panels. The contour levels represent 39.3%, 60%, and 80% of the maximum. The outer contour level represents approximately the 1-σ region of a 2D Gaussian distribution (1 − e−0.5).
Extended Data Fig. 2 MCMC-derived probability distributions of the fit parameters for pixel P2.
This pixel is located in ROI 3 at x = 961.0″ and y = − 154″ and marked by the small blue cross in Fig. 1. The upper right panel shows the microwave spectrum from this pixel by the open circles (mean values) and the error bars at ± 1σ level. Blue curves display a set of trial theoretical spectra employed within MCMC fitting. The free parameters of the fitting included (from top to bottom): the magnetic field B, the thermal number density nth, the nonthermal number density nnth, the spectral index δ, the viewing angle θ, and the minimum energy \({E}_{\min }\) in the power-law electron spectrum. Individual panels display solid black horizontal/vertical lines that indicate the best-fit values from the GSFIT minimization reported in10. The median values of the MCMC probability distributions are shown with the dotted blue horizontal/vertical lines. Dashed black lines in the histograms along the diagonal indicate ± 1-σ standard deviation of a given parameter. Correlations between all possible pairs of parameters are shown as two-dimensional histograms of the probability distributions in off-diagonal panels. The contour levels represent 39.3%, 60%, and 80% of the maximum. The outer contour level represents approximately the 1-σ region of a 2D Gaussian distribution (1 − e−0.5).
Extended Data Fig. 3 MCMC-derived probability distributions of the fit parameters for pixel P3.
This pixel is located in ROI 3 at x = 959.0″ and y = − 162″ and marked by the small blue plus sign in Fig. 1. The upper right panel shows the microwave spectrum from this pixel by the open circles (mean values) and the error bars at ± 1σ level. Blue curves display a set of trial theoretical spectra employed within MCMC fitting. The free parameters of the fitting included (from top to bottom): the magnetic field B, the thermal number density nth, the nonthermal number density nnth, the spectral index δ, the viewing angle θ, and the minimum energy \({E}_{\min }\) in the power-law electron spectrum. Individual panels display solid black horizontal/vertical lines that indicate the best-fit values from the GSFIT minimization reported in10. The median values of the MCMC probability distributions are shown with the dotted blue horizontal/vertical lines. Dashed black lines in the histograms along the diagonal indicate ± 1-σ standard deviation of a given parameter. Correlations between all possible pairs of parameters are shown as two-dimensional histograms of the probability distributions in off-diagonal panels. The contour levels represent 39.3%, 60%, and 80% of the maximum. The outer contour level represents approximately the 1-σ region of a 2D Gaussian distribution (1 − e−0.5).
Extended Data Fig. 4 Fermi/GBM spectra of two one-minute accumulations during the 2017 September 10 flare and their spectral fits.
Panels (a, b) show background-subtracted counts spectrum (black points) near the peak of the MeV electron emission (a), and at a time when there is still strong power-law electron emission, but the emission from the MeV-peaked component is weak (b). Vertical error bars correspond to the 1σ count uncertainty estimated from Poisson statistics, while horizontal bars indicate the width of each energy bin. The spectral fits are composed from three components: (i) power-law extension of the standard HXR component (red); (ii) the standard OSPEX template of combined spectrum of broad and narrow nuclear de-excitation lines (green); and (iii) thin-target bremsstrahlung component from the new MeV population modeled with a broken power-law energy spectrum. Gray lines on panels (a, b) show the sum of all three fitting components. (c) The electron energy spectrum (blue lines) needed to generate thin-target bremsstrahlung shown in blue in panels (a, b). The light gray polygons show 1σ uncertainties in the slopes of this spectrum reported in9. For reference, the red line shows the normalized spectrum of electrons responsible for the standard power-law X-ray / γ-ray component shown in red in panels (a, b).
Extended Data Fig. 5 Time histories of the power-law and MeV-peaked electron populations in the 2017 September 10 flare observed by the Fermi/GBM.
Red points show the mean values of photon flux above 300 keV of the standard power-law component of the flare-accelerated electrons, while the blue circles show the mean values of the electron flux above 300 keV of the MeV component derived from the thin-target bremsstrahlung fit; the vertical error bars show the corresponding 1σ uncertainties. The horizontal error bar corresponds to the time interval over which the initial photon spectrum was integrated to perform the spectral fitting. The red line shows microwave emission at 7.92 GHz integrated over ROI 1 and ROI 2. The blue line shows microwave emission at 15.9 GHz integrated over ROI 3.
Supplementary information
Supplementary Video 1 (download MOV )
Evolving map of the microwave spectral index, demonstrating evolution of β of the spatially resolved microwave flux Ff ∝ fβ at the locations where the microwave brightness temperature is above 109 K at one or more of the observing frequencies. The video demonstrates that the steep spectra are concentrated at/around ROI 3, and they appear and then disappear during the considered time episode.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Fleishman, G.D., Oparin, I., Nita, G.M. et al. Megaelectronvolt-peaked electrons in a coronal source of a solar flare. Nat Astron 10, 363–370 (2026). https://doi.org/10.1038/s41550-025-02754-w
Received:
Accepted:
Published:
Version of record:
Issue date:
DOI: https://doi.org/10.1038/s41550-025-02754-w
