No Arabic abstract
Bright and eclipsing, the high-mass X-ray binary Vela X-1 offers a unique opportunity to study accretion onto a neutron star from clumpy winds of O/B stars and to disentangle the complex accretion geometry of these systems. In Chandra-HETGS spectroscopy at orbital phase ~0.25, when our line of sight towards the source does not pass through the large-scale accretion structure such as the accretion wake, we observe changes in overall spectral shape on timescales of a few kiloseconds. This spectral variability is, at least in part, caused by changes in overall absorption and we show that such strongly variable absorption cannot be caused by unperturbed clumpy winds of O/B stars. We detect line features from high and low ionization species of silicon, magnesium and neon whose strengths and presence depend on the overall level of absorption. They imply a co-existence of cool and hot gas phases in the system that we interpret as a highly variable, structured accretion flow close to the compact object such as has been recently seen in simulations of wind accretion in high-mass X-ray binaries.
Vela X-1 is an eclipsing high mass X-ray binary (HMXB) consisting of a 283s accreting X-ray pulsar in a close orbit of 8.964 days around the B0.5Ib supergiant HD77581 at a distance of just 2.4 kpc. The system is considered a prototype of wind-accreting HMXB and it has been used as a baseline in different theoretical or modelling studies. We discuss the observational properties of the system and the use of the observational data as laboratory to test recent developments in modelling the accretion process in High-Mass X-ray Binaries (e.g., Sander et al. 2018; El Mellah et al. 2018), which range from detailed descriptions of the wind acceleration to modelling of the structure of the flow of matter close to the neutron star and its variations.
We have analyzed 3 observations of the High Mass X-ray Binary A0535+26 performed by the Rossi X-ray Timing Explorer (RXTE) 3, 5, and 6 months after the last outburst in 2011 February. We detect pulsations only in the second observation. The 3-20 keV spectra can be fit equally well with either an absorbed power law or absorbed thermal bremsstrahlung model. Re-analysis of 2 earlier RXTE observations made 4 years after the 1994 outburst, original BeppoSAX observations 2 years later, re-analysis of 4 EXOSAT observations made 2 years after the last 1984 outburst, and a recent XMM-Newton observation in 2012 reveal a stacked, quiescent flux level decreasing from ~2 to <1 x 10^{-11} ergs/cm2/s over 6.5 years after outburst. Detection of pulsations during half of the quiescent observations would imply that accretion onto the magnetic poles of the neutron star continues despite the fact that the circumstellar disk may no longer be present. The accretion could come from material built-up at the corotation radius or from an isotropic stellar wind.
We present an X-ray spectral analysis of the high-mass binary 4U~1700-37 during its hard-soft state evolution. We use the BeppoSAX, Suzaku and RXTE (Rossi X-ray Timing Explorer), Suzaku and BeppoSAX observations for this investigation. We argue that the X-ray broad-band spectra during all spectral states can be adequately reproduced by a model, consisting of a low-temperature Blackbody component, two Comptonized components both due to the presence of a Compton cloud (CC) that up-scatters seed photons of $T_{s1}$~< 1.4 keV, and $T_{s2}<$1 keV, and an iron-line component. We find using this model that the photon power-law index is almost constant, $Gamma_{1}sim 2$ for all spectral states. However, $Gamma_{2}$ shows a behavior depending on the spectral state. Namely, $Gamma_{2}$ is quasi-constant at the level of $Gamma_{2}sim 2$ while the CC plasma temperature $kT^{(2)}_e$ is less than 40 keV; on the other hand, $Gamma_{2}$ is in the range of $1.3<Gamma_{2}<2$, when $kT^{(2)}_e$ is greater than 40 keV. We explain this quasi-stability of $Gamma$ during most of hard-soft transitions of 4U~1700-37 in a framework of the model in which the resulting spectrum is described by two Comptonized components. We find that these Comptonized spectral components of the HMXB 4U~1700-37 are similar to those previously found in NS sources. This index dependence versus both mass accretion rate and $kT_e$ revealed in 4U~1700-37 is a universal observational evidence for the presence of a NS in 4U 1700-37.
Context: Vela X-1 is one of the best studied X-ray binaries. Frequently though, specific values for its parameters have been used in subsequent studies without considering alternatives. Aims: We aim to provide a robust compilation and synthesis of the accumulated knowledge about Vela X-1 as a solid baseline for future studies and identify specific avenues of possible future research. Methods: We explore the literature for Vela X-1 and on modelling efforts, describing the evolution of the system knowledge. We also add information derived from public data, especially the Gaia EDR3 release. Results: We update the distance to Vela X-1, the spectral classification for HD 77518 and find that the supergiant may be very close to filling its Roche lobe. Constraints on the clumpiness of the stellar wind have improved. The orbit is very well determined, but the uncertain inclination limits information on the neutron star mass. Estimates for the stellar wind have evolved towards lower velocities, supporting the idea of transient wind-captured disks around the neutron star. Hydrodynamic models and observations are consistent with an accretion wake trailing the neutron star. Conclusions: Vela X-1 is an excellent laboratory, but a lot of room remains to improve. Well-coordinated multi-wavelength observations and campaigns addressing the intrinsic variability are required. New opportunities will arise through new instrumentation. Models of the stellar wind should account for the orbital eccentricity and the non-spherical shape of HD 77581. Realistic multi-dimensional models of radiative transfer in the UV and X-rays are needed, but remain very challenging. Improved MHD models covering a wide range of scales would be required to improve understanding of the plasma-magnetosphere coupling. A full characterization of the accretion column remains another open challenge. (Abbreviated for arXiv)
We present a $sim$130 ks observation of the prototypical wind-accreting, high-mass X-ray binary Vela X-1 collected with XMM-Newton at orbital phases between 0.12 and 0.28. A strong flare took place during the observation that allows us to investigate the reaction of the clumpy stellar wind to the increased X-ray irradiation. To examine the winds reaction to the flare, we performed both time-averaged and time-resolved analyses of the RGS spectrum and examined potential spectral changes. We focused on the high-resolution XMM-Newton RGS spectra and divided the observation into pre-flare, flare, and post-flare phases. We modeled the time-averaged and time-resolved spectra with phenomenological components and with the self-consistent photoionization models calculated via CLOUDY and XSTAR in the pre-flare phase, where strong emission lines due to resonant transitions of highly ionized ions are seen. In the spectra, we find emission lines corresponding to K-shell transitions in highly charged ions of oxygen, neon, magnesium, and silicon as well as radiative recombination continua (RRC) of oxygen. Additionally, we observe potential absorption lines of magnesium at a lower ionization stage and features identified as iron L lines. The CLOUDY and XSTAR photoionization models provide contradictory results, either pointing towards uncertainties in theory or possibly a more complex multi-phase plasma, or both. We are able to demonstrate the existence of a plethora of variable narrow features, including the firm detection of oxygen lines and RRC that RGS enables to observe in this source for the first time. We show that Vela X-1 is an ideal source for future high-resolution missions, such as XRISM and Athena.