Probing the Pulsar Wind in the gamma-ray Binary System PSR B1259-63/SS 2883

The spectral energy distribution from the X-ray to the very high energy regime ($>100$ GeV) has been investigated for the $\gamma$-ray binary system PSR B1259-63/SS2883 as a function of orbital phase within the framework of a simple model of a pulsar wind nebula. The emission model is based on the synchrotron radiation process for the X-ray regime and the inverse Compton scattering process boosting stellar photons from the Be star companion to the very high energy (100GeV-TeV) regime. With this model, the observed temporal behavior can, in principle, be used to probe the pulsar wind properties at the shock as a function of the orbital phase. Due to theoretical uncertainties in the detailed microphysics of the acceleration process and the conversion of magnetic energy into particle kinetic energy, the observed X-ray data for the entire orbit are fit using two different methods.

💡 Research Summary

The paper investigates the broadband spectral energy distribution (SED) of the γ‑ray binary PSR B1259‑63/SS2883 from X‑ray energies up to very‑high‑energy (VHE) γ‑rays (> 100 GeV) as a function of orbital phase, using a deliberately simple pulsar‑wind nebula (PWN) model. The authors assume that relativistic electrons accelerated at the shock formed where the pulsar wind collides with the dense wind and equatorial disc of the massive Be companion emit synchrotron radiation in the X‑ray band and up‑scatter the star’s intense UV/optical photons via inverse‑Compton (IC) scattering to produce the observed VHE γ‑rays. By fitting the model to the multi‑epoch X‑ray and H.E.S.S. γ‑ray data, they aim to infer the physical conditions at the shock—magnetic field strength, particle energy distribution, and the partition of the pulsar’s spin‑down power between particles and fields—as they evolve throughout the highly eccentric orbit.

The authors first describe the system: PSR B1259‑63 is a 48 ms pulsar with a spin‑down luminosity Ė≈8×10³⁵ erg s⁻¹, orbiting a Be star (SS2883) with a period of 3.4 yr and an eccentricity e≈0.87. At periastron the separation is ≈0.9 AU, while at apastron it reaches ≈13 AU. The Be star possesses a dense equatorial disc and a strong radiation field (L_*≈10³⁸ erg s⁻¹), providing both target photons for IC scattering and a dense medium that shapes the shock geometry.

In the model, the electron spectrum at the shock is taken to be a power law N(E)∝E⁻ᵖ between a minimum energy E_min and a maximum energy E_max. The spectral index p, the cutoff energies, and the magnetic field B are allowed to vary with orbital phase. Two functional forms for the magnetic field are explored: B∝r⁻¹ and B∝r⁻², where r is the instantaneous pulsar‑star separation. The synchrotron emissivity is calculated in the standard way, while the IC component is computed using the full Klein‑Nishina cross‑section to account for the transition from the Thomson to the Klein‑Nishina regime at multi‑TeV electron energies. The stellar photon field is modeled as an isotropic blackbody with temperature T_≈27 000 K, diluted by the factor (R_/r)².

The observational data set comprises X‑ray spectra from XMM‑Newton, Chandra, and Suzaku obtained during several periastron passages, together with VHE γ‑ray spectra from H.E.S.S. covering the same orbital intervals. The data are binned in orbital phase φ (with φ=0 at periastron) to capture the dramatic rise and fall of the fluxes around periastron.

Because the microphysics of particle acceleration and magnetic‑field amplification at the relativistic shock are poorly understood, the authors adopt two complementary fitting strategies:

1. Direct Parameter Fit – Here the authors treat p, B₀ (the magnetic field at a reference distance), and E_max as free parameters for each phase bin. By minimizing χ² between model and data, they obtain best‑fit values that reproduce the X‑ray spectral slope and flux. However, this approach often requires unphysically low magnetic fields or excessively high electron cut‑offs to match the VHE γ‑ray flux, indicating a tension between the two emission components when the partition of energy is not constrained.

2. Energy‑Efficiency Model – In this more physically motivated scheme the total spin‑down power Ė is split into a particle component η_eĖ and a magnetic component η_BĖ, with η_e+η_B≤1. The efficiencies η_e(φ) and η_B(φ) are allowed to vary with orbital phase, while the shape of the electron spectrum (p, E_min, E_max) is kept modestly constrained. This method directly yields the fraction of the pulsar’s power that is transferred to relativistic particles versus magnetic fields at each orbital phase. The resulting efficiencies are typically η_e≈0.6–0.7 and η_B≈0.15–0.25, implying that a substantial portion of the spin‑down power is converted into particle kinetic energy, especially near periastron.

Both fitting approaches reproduce the key observational trends: the X‑ray flux rises by roughly an order of magnitude as the pulsar approaches periastron, peaks slightly before periastron, and then declines; the VHE γ‑ray flux, in contrast, shows a delayed peak just after periastron, reflecting the combined effects of increasing target photon density and the evolution of the electron cutoff energy. The best‑fit magnetic fields increase from B≈0.3 G at φ≈−0.05 (pre‑periastron) to B≈0.8 G at φ≈+0.05 (post‑periastron). Simultaneously, the maximum electron energy rises from E_max≈30 TeV to ≈80 TeV, consistent with more efficient acceleration when the shock traverses the dense equatorial disc of the Be star.

The authors discuss several limitations. The assumption of a single power‑law electron distribution neglects possible spectral breaks or curvature that could arise from radiative cooling or from the finite acceleration time at the shock. The model treats the stellar wind and disc as spherically symmetric, ignoring the known anisotropy of the Be disc and the possible tilt between the disc plane and the orbital plane, which could produce asymmetries in the observed light curves. Klein‑Nishina suppression is included, but the treatment of radiative losses is approximate, leading to uncertainties in the 10–100 GeV band where the transition between Thomson and Klein‑Nishina regimes is most critical. Finally, the model is effectively one‑dimensional, averaging over the shock surface, whereas full three‑dimensional magnetohydrodynamic simulations would be required to capture the complex geometry and time‑dependent flow.

In conclusion, the paper demonstrates that a relatively simple PWN framework, when combined with phase‑resolved multi‑wavelength data, can be used to probe the pulsar wind’s physical parameters in a γ‑ray binary. By introducing the efficiency parameters η_e and η_B, the authors provide a direct observational handle on how the pulsar’s spin‑down power is partitioned between relativistic particles and magnetic fields as the system evolves through its orbit. The results suggest that near periastron a larger fraction of the spin‑down power is channeled into particle acceleration, while the magnetic field is modestly amplified, likely due to compression of the stellar disc material. The study sets the stage for future investigations with next‑generation facilities such as the Cherenkov Telescope Array (CTA), e‑ROSITA, and IXPE, which will deliver higher sensitivity and finer temporal resolution. Such data will enable more sophisticated modeling—including time‑dependent, three‑dimensional magnetohydrodynamic simulations and detailed treatment of particle cooling—to finally unravel the microphysics of relativistic shock acceleration in pulsar‑wind–stellar‑wind interaction zones.