X-ray Binaries: a potential dominant contributor to the cosmic ray spectral knee structure

``PeVatrons" refer to astrophysical sources capable of accelerating particles to energies $\sim$PeV and higher, potentially contributing to the cosmic ray spectrum in the knee region. Recently, HAWC and LHAASO have discovered a new type PeVatrons – X-ray binaries, allowing us to investigate in greater depth of the contributions of these sources to cosmic rays around the knee region. There are hundreds of X-ray binaries in our galaxy observed, which are potential PeVatrons.In this work, we derive the radial distribution of X-ray binaries in the Galaxy. Then we use the DRAGON package to calculate energy spectrum, anisotropy of cosmic rays as well as the resulting diffuse gamma ray emissions, after considering them as factories of cosmic rays in the knee energy bands. Our findings show that the contributions from X-ray binary PeVatrons may be dominant. More X-ray binary PeVatrons can be observed by LHAASO and HAWC in the future, and will confirm the contribution of X-ray binaries to high energy cosmic rays.

💡 Research Summary

The paper investigates whether X‑ray binaries (XRBs) can dominate the cosmic‑ray (CR) spectrum around the “knee” (∼3 PeV) by treating the hundreds of known Galactic XRBs as a population of PeVatrons. The authors first construct a spatial distribution for XRBs using recent catalogs (Neumann et al. 2023; Avakyan et al. 2023) that list 172 high‑mass X‑ray binaries (HMXBs) and 360 low‑mass X‑ray binaries (LMXBs). They project these sources onto the Galactic plane, adopt a Sun–Galactic‑center distance of 8.5 kpc, and correct for distance‑dependent selection effects by building a concentric‑circle grid centered on both the Sun and the Galactic centre. In each grid cell the observed surface density (sources kpc⁻²) and its Poisson error are calculated, and a weighted average over the grid yields a radial density profile. The profile is fitted with a two‑component exponential function
ρ(R)=A exp(−a R/R⊙)+B exp(−b R²/R⊙²)
with best‑fit parameters a=2.67±0.21, b=39.33±3.88, A=1.54±0.29 kpc⁻², B=9.74±1.99 kpc⁻². This function captures a strong concentration toward the Galactic centre together with a broader disk component, and it is used as the source term in the propagation model.

For CR propagation the authors employ the DRAGON code, which solves the full diffusion‑convection‑reacceleration equation in three dimensions. The source spectrum for each XRB is assumed to be a power law with an exponential cutoff, dN/dE∝E⁻²·² exp(−E/3 PeV), and a CR acceleration efficiency of 10 % of the jet kinetic power. Standard diffusion parameters (Dxx≈3×10²⁸ cm² s⁻¹ β(R/4 GV)^δ with δ≈0.33), a modest convective wind (≈5 km s⁻¹ ẑ), and re‑acceleration are adopted. The Galactic halo height is set to 4 kpc and the disk thickness to 0.2 kpc, with a uniform magnetic field of 5 µG.

The resulting all‑sky CR spectrum shows that XRBs contribute negligibly below ∼1 PeV, where supernova remnants (SNRs) dominate, but they become increasingly important above the knee. Between 2–5 PeV the XRB component accounts for 30–60 % of the total flux, flattening the overall spectrum and naturally reproducing the observed softening at the knee. The model also predicts a modest anisotropy in the 10–30 TeV band that matches current measurements, reflecting the asymmetric distribution of XRBs in the Galactic plane. In the gamma‑ray channel, secondary π⁰ decay photons generated by CR interactions with interstellar gas reproduce the diffuse gamma‑ray flux measured by LHAASO, providing an independent consistency check.

The authors discuss several caveats. The assumed CR acceleration efficiency and the identical spectral shape for HMXBs and LMXBs ignore possible differences in jet composition, magnetic reconnection rates, and environmental densities. The DRAGON magnetic‑field model is simplified and does not capture spiral‑arm or turbulent structures that could affect diffusion locally. The catalogued XRB count is likely incomplete, especially toward the heavily obscured inner Galaxy, which could alter the normalization of the source term by a factor of a few. Moreover, the exponential cutoff energy (3 PeV) is chosen to match the knee but lacks direct observational confirmation for most binaries.

In conclusion, the paper presents a plausible scenario in which X‑ray binaries, acting as a collective population of PeVatrons, can dominate the CR spectrum around the knee and explain associated diffuse gamma‑ray observations. While the model reproduces several key observables, the uncertainties mentioned above mean that the claim remains provisional. Future high‑sensitivity observations by LHAASO, HAWC, and next‑generation gamma‑ray facilities, together with more complete XRB catalogs and detailed jet‑physics simulations, will be essential to confirm or refute the dominant‑XRB hypothesis.