No Arabic abstract
We present an analysis of chandra/LETGS observations of the ultracompact X-ray binary (UCXB) 4U 1626$-$67, continuing our project to analyze the existing Chandra gratings data of this interesting source. The extremely low mass, hydrogen-depleted donor star provides a unique opportunity to study the properties and structure of the metal-rich accreted plasma. There are strong, double-peaked emission features of OVII-VIII and Ne IX-X, but no other identified emission lines are detected. Our spectral fit simultaneously models the emission line profiles and the plasma parameters, using a two-temperature collisionally-ionized plasma. Based on our line profile fitting, we constrain the inclination of the system to 25--60$^{circ}$ and the inner disk radius to $sim$1500 gravitational radii, in turn constraining the donor mass to $lesssim$0.026 M_sun, while our plasma modeling confirms previous reports of high neon abundance in the source, establishing a Ne/O ratio in the system of $0.47 pm 0.04$, while simultaneously estimating a very low Fe/O ratio of $0.0042 pm 0.0008$ and limiting the Mg/O ratio to less than 1% by number. We discuss these results in light of previous work.
We report on high-resolution X-ray spectroscopy of the ultracompact X-ray binary pulsar 4U 1626-67 with Chandra/HETGS acquired in 2010, two years after the pulsar experienced a torque reversal. The well-known strong Ne and O emission lines with Keplerian profiles are shown to arise at the inner edge of the magnetically-channeled accretion disk. We exclude a photoionization model for these lines based on the absence of sharp radiative recombination continua. Instead, we show that the lines arise from a collisional plasma in the inner-disk atmosphere, with $Tsimeq 10^7$ K and $n_e sim 10^{17}$ cm^(-3). We suggest that the lines are powered by X-ray heating of the optically-thick disk inner edge at normal incidence. Comparison of the line profiles in HETGS observations from 2000, 2003, and 2010 show that the inner disk radius decreased by a factor of two after the pulsar went from spin-down to spin-up, as predicted by magnetic accretion torque theory. The inner disk is well inside the corotation radius during spin-up, and slightly beyond the corotation radius during spin-down. Based on the disk radius and accretion torque measured during steady spin-up, the pulsars X-ray luminosity is $2times 10^{36}$ erg/s, yielding a source distance of 3.5(+0.2-0.3) kpc. The mass accretion rate is an order of magnitude larger than expected from gravitational radiation reaction, possibly due to X-ray heating of the donor. The line profiles also indicate a binary inclination of 39(+20-10) degrees, consistent with a 0.02 Msun donor star. Our emission measure analysis favors a He white dwarf or a highly-evolved H-poor main sequence remnant for the donor star, rather than a C-O or O-Ne white dwarf. The measured Ne/O ratio is 0.46+-0.14 by number. In an appendix, we show how to express the emission measure of a H-depleted collisional plasma without reference to a H number density.
We present an analysis of the spectral shape and pulse profile of the accretion-powered pulsar 4U 1626-67 observed with Suzaku and NuSTAR during a spin-up state. The pulsar, which experienced a torque reversal to spin-up in 2008, has a spin period of 7.7 s. Comparing the phase-averaged spectra obtained with Suzaku in 2010 and with NuSTAR in 2015, we find that the spectral shape changed between the two observations: the 3-10 keV flux increased by 5% while the 30-60 keV flux decreased significantly by 35%. Phase-averaged and phase-resolved spectral analysis shows that the continuum spectrum observed by NuSTAR is well described by an empirical NPEX continuum with an added broad Gaussian emission component around the spectral peak at 20 keV. Taken together with the observed Pdot value obtained from Fermi/GBM, we conclude that the spectral change between the Suzaku and NuSTAR observations was likely caused by an increase of the accretion rate. We also report the possible detection of asymmetry in the profile of the fundamental cyclotron line. Furthermore, we present a study of the energy-resolved pulse profiles using a new relativistic ray tracing code, where we perform a simultaneous fit to the pulse profiles assuming a two-column geometry with a mixed pencil- and fan-beam emission pattern. The resulting pulse profile decompositions enable us to obtain geometrical parameters of accretion columns (inclination, azimuthal and polar angles) and a fiducial set of beam patterns. This information is important to validate the theoretical predictions from radiation transfer in a strong magnetic field.
We present analysis of 4U 1626-67, a 7.7 s pulsar in a low-mass X-ray binary system, observed with the hard X-ray detector of the Japanese X-ray satellite Suzaku in March 2006 for a net exposure of sim88 ks. The source was detected at an average 10-60 keV flux of sim4 x10^-10 erg cm^-2 s^-1. The phase-averaged spectrum is reproduced well by combining a negative and positive power-law times exponential cutoff (NPEX) model modified at sim 37 keV by a cyclotron resonance scattering feature (CRSF). The phase-resolved analysis shows that the spectra at the bright phases are well fit by the NPEX with CRSF model. On the other hand, the spectrum in the dim phase lacks the NPEX high-energy cutoff component, and the CRSF can be reproduced by either an emission or an absorption profile. When fitting the dim phase spectrum with the NPEX plus Gaussian model, we find that the feature is better described in terms of an emission rather than an absorption profile. The statistical significance of this result, evaluated by means of an F-test, is between 2.91 x 10^-3 and 1.53 x 10^-5, taking into account the systematic errors in the background evaluation of HXD-PIN. We find that, the emission profile is more feasible than the absorption one for comparing the physical parameters in other phases. Therefore, we have possibly detected an emission line at the cyclotron resonance energy in the dim phase.
Recent X-ray observations by Fermi/GBM discovered a new torque reversal of 4U 1626-67 after 18 years of steady spinning down. Using Swift/BAT observations we were able to center this new torque reversal on Feb 4 2008, lasting approximately 150 days. From 2004 up to the end of 2007, the spin-down rate averaged at a mean rate of ~dnu/dt=-4.8e-13 Hz s-1 until the torque reversal reported here. Since then it has been following a steady spin-up at a mean rate of ~dnu/dt= 4e-13 Hz s-1. The properties of this torque reversal, as well as the lack of correlation between the X-ray flux and the torque applied to the neutron star before this transition, challenges our understanding of the physical mechanisms operating in this system.
After about 18 years of steadily spinning down, the accretion-powered pulsar 4U 1626-67, experienced a torque reversal at the beginning of 2008. For the present study we have used all available Fermi/GBM data since its launch in 2008 June 11 and over 5 yr of hard X-ray Swift/BAT observations (starting from 2004 October up to the present time). This second detected torque reversal is centered near MJD 54500 (2008 Feb 4) and it lasts approximately 150 days. From 2004 up to the end of 2007 4U 1626-67 the spin-down rate decreased at a mean rate of ~ -5.5E-13 Hz s-1 until the source reversed torque again. Since then it has been following a steady spin-up at a mean rate of ~ 5E-13 Hz s-1. In addition, 4U 1626-67 increased its flux simultaneously (a ~2.5 factor). We present detailed long-term timing analysis of this source and a long term spectral hardness ratio study in order to see whether there are spectral changes around this new observed torque reversal.