Identification of radio and gamma-ray pulsars in X-rays using data from the SRG/eROSITA all-sky survey

Using the data from the all-sky survey in soft X-rays performed by the eROSITA telescope onboard the Spectrum-Roentgen-Gamma observatory we identified known radio and $γ$-ray pulsars in the eastern half of the sky. As a result, new candidate counterparts were found for twelve pulsars of different ages and types at a $\gtrsim$ 3$σ$ confidence level. A comparable number had been previously identified in the western half of the sky. In total, this represents about 12% of all known pulsars already detected in X-rays. For the new counterparts, we provide estimates of their X-ray fluxes, preliminary characteristics of their X-ray spectra, and brief descriptions of the pulsars’ properties. In addition, in the eastern half of the sky eROSITA detected 55 pulsars previously identified in X-rays by other telescopes.

💡 Research Summary

This paper presents a systematic search for X‑ray counterparts of known radio and γ‑ray pulsars in the eastern half of the sky using data from the all‑sky survey performed by the eROSITA telescope aboard the Spectrum‑Roentgen‑Gamma (SRG) mission. The authors start from the ATNF pulsar catalogue (version 2.6.0), selecting 2 260 pulsars located in the eastern hemisphere, excluding those with positional uncertainties larger than 10 arcsec, those associated with globular clusters, and those lacking reliable coordinates. They cross‑matched these positions with the eROSITA source catalogue, which includes only detections with a likelihood L ≥ 6 (approximately a 3σ Gaussian confidence level).

The cross‑match yielded twelve new X‑ray candidate counterparts at ≥3σ significance. The sample comprises two isolated middle‑aged pulsars (characteristic age < 1 Myr), three isolated old pulsars (> 1 Myr, one of which is a rotating radio transient, RRAT), and six millisecond pulsars (MSPs), of which only one is isolated. Positional offsets between the pulsar and the eROSITA source range from 2 to 10 arcsec, well within the 98 % eROSITA positional error circles (3–10 arcsec). In addition, the survey recovered 55 pulsars that had already been identified in X‑rays by previous missions, confirming the high completeness of the eROSITA survey in the soft X‑ray band (0.3–10 keV).

For each new candidate the authors extracted eROSITA spectra using a 60 arcsec source aperture and a background annulus of 120–300 arcsec, masking any contaminating sources with 40 arcsec circles. When available, Swift and Chandra observations of the same field were retrieved and fitted simultaneously. Spectral fitting was performed with XSPEC v12.13.1, employing C‑statistics (Cash 1979) and grouping the data to ensure at least one count per energy bin. The primary model applied to all spectra was an absorbed power‑law (PL). When the best‑fit photon index Γ exceeded 3.5, indicating a possible thermal contribution, the authors added either a black‑body (BB) component or a magnetized hydrogen atmosphere model (NSMAXG) with fixed neutron‑star mass (1.4 M⊙) and radius (13 km). Interstellar absorption was modeled with TBABS using Wilms abundances; column densities N_H were estimated from three‑dimensional Galactic extinction maps (Green et al., Lallament et al.) and converted using the empirical N_H–E(B–V) relation of Foight et al. (2016). The derived N_H values were then fixed during fitting.

The spectral results show that most candidates are well described by a simple PL with photon indices Γ≈1.5–2.5, typical for non‑thermal magnetospheric emission. Unabsorbed fluxes in the 0.3–10 keV band lie between ~1×10⁻¹⁴ and ~5×10⁻¹³ erg s⁻¹ cm⁻². For example, PSR J0006+1834, with a fixed Γ = 2, yields an unabsorbed flux of (1.3 +1.3 −1.1)×10⁻¹⁴ erg s⁻¹ cm⁻² (C‑stat = 19 for 18 d.o.f.). PSR J0242+6256, a rotating radio transient, required two fits with different N_H estimates (1.2×10²⁰ cm⁻² and 7.7×10²⁰ cm⁻²), resulting in Γ = 1.4 ± 0.5 and Γ = 1.8 ± 0.5 respectively, illustrating the sensitivity of the spectral shape to the assumed absorption. PSR J0248+6021, previously undetected by Chandra, is confirmed by eROSITA and the 2CXO catalogue with a flux of ~1×10⁻¹⁴ erg s⁻¹ cm⁻².

The authors discuss the significance of these detections: the twelve new counterparts increase the fraction of known pulsars with X‑ray detections to about 12 % of the total pulsar population, consistent with earlier work on the western sky (Mayer & Becker 2024). The detection of 55 previously known X‑ray pulsars demonstrates eROSITA’s capability to recover faint, soft X‑ray sources across the whole sky, despite its moderate angular resolution (≈18 arcsec half‑energy width).

In the concluding section, the paper emphasizes the prospects offered by the forthcoming eROSITA eight‑scan (final) data set, which will improve sensitivity by roughly a factor of two and provide better statistics for faint sources. Combined with multi‑wavelength data from radio facilities (FAST, LOFAR, MeerKAT) and γ‑ray observatories (Fermi‑LAT), the expanded sample will enable more accurate distance estimates (especially for radio‑quiet γ‑ray pulsars using pseudo‑distances), refined measurements of X‑ray efficiencies (L_X/Ė), and deeper investigations of thermal surface emission, magnetic field strengths (via cyclotron features), and magnetospheric emission mechanisms. The authors also suggest targeted follow‑up observations with higher‑resolution X‑ray telescopes (Chandra, XMM‑Newton, NICER) to disentangle thermal and non‑thermal components, constrain neutron‑star radii, and test equation‑of‑state models. Overall, the study demonstrates that eROSITA is a powerful tool for expanding the census of X‑ray pulsars and for probing the diverse high‑energy phenomenology of neutron stars.