Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Narrow iron- and nickel-K absorption lines from the eclipsing low-mass X-ray binary AX~J1745.6$-$2901

Kojiro Tanaka, Yoshitomo Maeda, Ryota Tomaru, Lia Corrales, María Díaz Trigo, Chris Done, Tadayasu Dotani, Manabu Ishida, Satoru Katsuda, Yoshiaki Kanemaru, Richard Kelley, Aya Kubota, Hironori Matsumoto, Masayoshi Nobukawa, Megumi Shidatsu, Randall Smith, Hiromasa Suzuki, Hiromitsu Takahashi, Yohko Tsuboi, Hideki Uchiyama, Shigeo Yamauchi, Anje Yoshimoto, Q. Daniel Wang, Jon M. Miller, Frederick S. Porter, Shinya Yamada

Abstract

We report the presence of a highly ionized absorber in the transient, eclipsing low-mass X-ray binary AX J1745.6-2901, observed from Feb. 26 to 29, 2024 with XRISM's Resolve and Xtend instruments. During a soft/high state without dips, Resolve's high spectral resolution (E/dE ~ 1000, full width at half maximum) revealed narrow velocity widths (sigma ~ 110 km/s) for Fe XXVI and Ni XXVIII lines, even with low photon statistics. These widths are consistent with binary orbital motion. The observed modest blueshift velocity (~160 km/s) indicates that the absorber is located sufficiently far from the neutron star (> 10^9 cm), so that gravitational redshift effects are not dominant. On the other hand, broad-band spectral analysis using a photoionized plasma model applied to the Xtend data constrains the absorber to lie within a radius of < 10^9.5 cm, as inferred from the upper limits of the best-fit ionization parameter (log xi ~ 4.4) and the large column density (~ 1.6 x 10^24 cm^-2). At this distance, the observed outward velocity of the absorber is about an order of magnitude smaller than the escape velocity from the neutron star.

Narrow iron- and nickel-K absorption lines from the eclipsing low-mass X-ray binary AX~J1745.6$-$2901

Abstract

We report the presence of a highly ionized absorber in the transient, eclipsing low-mass X-ray binary AX J1745.6-2901, observed from Feb. 26 to 29, 2024 with XRISM's Resolve and Xtend instruments. During a soft/high state without dips, Resolve's high spectral resolution (E/dE ~ 1000, full width at half maximum) revealed narrow velocity widths (sigma ~ 110 km/s) for Fe XXVI and Ni XXVIII lines, even with low photon statistics. These widths are consistent with binary orbital motion. The observed modest blueshift velocity (~160 km/s) indicates that the absorber is located sufficiently far from the neutron star (> 10^9 cm), so that gravitational redshift effects are not dominant. On the other hand, broad-band spectral analysis using a photoionized plasma model applied to the Xtend data constrains the absorber to lie within a radius of < 10^9.5 cm, as inferred from the upper limits of the best-fit ionization parameter (log xi ~ 4.4) and the large column density (~ 1.6 x 10^24 cm^-2). At this distance, the observed outward velocity of the absorber is about an order of magnitude smaller than the escape velocity from the neutron star.
Paper Structure (22 sections, 7 equations, 16 figures, 7 tables)

This paper contains 22 sections, 7 equations, 16 figures, 7 tables.

Table of Contents

  1. introduction
  2. Observations and data reduction
  3. Data analysis and results
  4. Images
  5. Lightcurves
  6. Spectra
  7. Background and energy band selection
  8. Modeling dust scattering
  9. Line width and shift using Resolve
  10. Photo-ionization modeling
  11. Discussion
  12. Absorber and its location
  13. Comments on wind-driving mechanism
  14. Comments on accretion disk and its X-ray luminosity
  15. Summary
  16. ...and 7 more sections

Figures (16)

  • Figure 1: (Left) Resolve and (right) Xtend images in the 2–10 keV band are shown in detector coordinates, with north oriented upwards. The field of view for Resolve spans approximately $3' \times 3'$. The source AX J1745.6$-$2901 is located just outside the corner of the Resolve pixel array but remains within the field of view of Xtend. Regions used for temporal and spectral analyses are indicated with solid lines for the source ($0.\mkern-4mu{}^{\prime}15$ -- $1.\mkern-4mu{}^{\prime}00$ radius annual) and dashed lines for the background ($1.\mkern-4mu{}^{\prime}00$ radius circle). These regions for Resolve are also overlaid on the Xtend image for comparison.
  • Figure 2: Folded light curves of Xtend in the 2--10 keV (upper) and 6--8 keV (lower) bands, with phase 0 defined at MJD 60366.029. The folded period is $30,063$ s. The background was not subtracted.
  • Figure 3: A schematic view illustrating the absorption and scattering processes for AX J1745.6$-$2901. Panel (a) shows the non-eclipse phase, where both direct and scattered-in light paths contribute to the observed signal. Panel (b) shows the eclipse phase, where the scattered-in component can still reach us due to the time lag between the direct light path and the scattered light path.
  • Figure 4: The scattering cross-section of the non-eclipse to eclipse spectra, taken from the $1.\mkern-4mu{}^{\prime}$ outer radius region. The solid line represents the best-fit parameters of the $E^{-2}$ model. The excess observed above 6 keV may suggest the presence of an additional intrinsic and extended component during the eclipse.
  • Figure 5: Resolve spectra in the 6.8--9 keV band fitted with the ionabs model. The spectra were binned for clarity in display.
  • ...and 11 more figures