Stellar wind accretion in high-mass X-ray binaries

Recent discoveries have confirmed the existence of a large population of X-ray sources fuelled by accretion from the stellar wind of an OB supergiant. Such systems are powerful laboratories to study many aspects of astrophysics. Over the last decades, the physics of accretion in these systems has been the subject of extensive research, mainly through numerical methods. In spite of this effort, large uncertainties remain in our understanding, reflecting the complexity of the physical situation. A crucial issue that remains open is the possible formation of accretion discs. Though the spin evolution of neutron stars in these systems suggests that angular momentum is, at least occasionally, accreted, and many observational facts seem to require the existence of discs, computational results do not favour this possibility. In this brief review, I will summarise some of the open questions in this area.

💡 Research Summary

**
High‑mass X‑ray binaries (HMXBs) consisting of a compact object—most often a neutron star—and an OB supergiant have emerged as a prolific class of X‑ray sources powered by the capture of the massive star’s stellar wind. The wind is characterized by high mass‑loss rates (10⁻⁶–10⁻⁵ M⊙ yr⁻¹) and terminal velocities of several thousand kilometres per second, following a β‑law acceleration profile that yields a density falling off roughly as r⁻². The wind is not smooth; it contains dense clumps produced by line‑driven instabilities, radiative cooling, and surface perturbations.

When the compact object orbits within this outflow, its gravitational sphere of influence (the capture radius) draws in material. The inflowing gas passes through a sonic point and, if the neutron star possesses a strong magnetic field, an Alfvén radius where magnetic pressure balances ram pressure. A shock forms in this region, heating the gas to X‑ray emitting temperatures. The resulting X‑ray luminosity can be highly variable, reflecting changes in wind density, clump encounters, and the geometry of the magnetosphere.

A central, unresolved issue is whether the captured wind can deliver enough angular momentum to form a persistent or transient accretion disc around the neutron star. Observational clues—spin‑up episodes, quasi‑periodic oscillations, infrared excesses, and certain spectral signatures—suggest that at least occasionally angular momentum is transferred efficiently enough to build a disc‑like structure. Yet state‑of‑the‑art three‑dimensional radiation‑hydrodynamic (RHD) and magnetohydrodynamic (MHD) simulations consistently show that, on average, the specific angular momentum supplied by a clumpy, supersonic wind is insufficient to sustain a long‑lived disc. The simulations reveal that clumps can momentarily increase the angular momentum budget, but the excess is typically accreted within seconds to minutes, preventing the accumulation of material into a stable Keplerian disc.

Additional physical mechanisms complicate the picture. Magnetospheric gating (or “propeller” effects) can intermittently halt inflow when the magnetosphere expands beyond the corotation radius, leading to abrupt drops in X‑ray flux. Shadowing by the stellar wind or by structures in the magnetosphere can produce anisotropic inflow, further modulating the accretion rate. Wind–wind collisions between the stellar outflow and any outflow from the compact object generate hot plasma that enhances X‑ray emission but also increases radiative cooling, which tends to suppress disc formation by keeping the gas temperature high and the density low.

The tension between observations that imply disc presence and simulations that favour direct wind accretion points to missing physics in current models. Key uncertainties include the size distribution and filling factor of wind clumps, the topology and strength of the neutron star’s magnetic field, the role of radiative pressure on the inflowing gas, and the coupling between small‑scale turbulence and large‑scale flow geometry. Moreover, most simulations have limited spatial resolution and often treat the wind as a steady background rather than a fully time‑dependent, multi‑phase medium.

Future progress will require a multi‑pronged approach. High‑resolution, global MHD‑RHD simulations that resolve clump formation, magnetic reconnection, and radiative transfer simultaneously are essential to quantify angular momentum transport accurately. Coordinated multi‑wavelength campaigns—combining radio, optical/IR, soft and hard X‑ray, and γ‑ray observations—will help disentangle the signatures of disc‑like emission from those of direct wind accretion. Precise timing of neutron‑star spin changes, together with spectral modeling of cyclotron lines and Fe Kα fluorescence, can provide indirect measurements of the inner accretion flow geometry. Finally, systematic studies of wind–wind interaction zones and magnetospheric “shadow” regions will clarify how these non‑linear processes shape the observed variability.

In summary, while the prevailing theoretical picture supports direct wind capture as the dominant accretion mode in HMXBs, a number of observational phenomena remain unexplained without invoking transient or partial accretion discs. Bridging this gap demands more sophisticated simulations and comprehensive observational diagnostics, which together will illuminate the complex interplay of wind dynamics, magnetic fields, and angular momentum in these extraordinary astrophysical laboratories.