A NuSTAR Observation of the black hole candidate XTE J1859+226 at a low Eddington fraction

ATel #14512; Paul A. Draghis (University of Michigan), Jon M. Miller (University of Michigan), Mayura Balakrishnan (University of Michigan), Mark Reynolds (University of Michigan), Abigail Stevens (University of Michigan), Bailey Tetarenko (University of Michigan), Nicolas Trueba (University of Michigan), Abderahmen Zoghbi (University of Michigan)
on 5 Apr 2021; 22:01 UT
Distributed as an Instant Email Notice Transients
Credential Certification: Paul Draghis (pdraghis@umich.edu)

Tweet

The black hole LMXB system XTE J1859+226 was observed to brighten in optical in February 2021 using ZTF (ATel #14372) and in X-rays using Swift (ATel #14375). We observed the source using NuSTAR on 2021 February 13 under ObsID 90701305002 for a net exposure of 41 ks, with an average count rate of 0.1 counts/second when combining the two FPM NuSTAR sensors.

The light curve shows no variability on the time scale of the observation. Spectra were extracted from circular regions with 60 arcsecond radii centered at the position of the source with similar sized background regions. The spectra are dominated by the background flux above 20 keV, so all spectral analysis was performed in the 3-20 keV band. The spectra were binned to have a minimum of 20 counts per bin. All uncertainties reported represent the 90% confidence ranges for each parameter.

Fitting the spectra with a power law produces a good fit, with chi2/dof = 79.30/85 (0.93), returning a value of the photon index of the power law Gamma = 1.84 +/- 0.12. The fit shows evidence of residual emission in the 6-7 keV range, which can be linked to reflection from the accretion disk. Reflection can be used to determine the geometry of the flow and the inner radius of the accretion disk, especially at low Eddington fractions. We explored whether the data supports a Gaussian emission line. Adding a Gaussian component to the power law model and fitting for its position, width, and normalization returns as best fit parameters a line centered at 6.7 +0.2/-0.3 keV with a FWHM of 0.2 +0.8/-0.2 keV and an equivalent width of 270 +/- 230 eV, without changing the power law index. The addition of the Gaussian component improves the fit, producing a chi2/dof = 73.21/82 (0.89). The change in chi2 of 6.09 at the cost of three extra parameters gives an improvement at the level of 1.6 sigma. This line is consistent with both the Fe K line and with He-like Fe XXV. We note additional residual emission in the 8-9 keV range which, when fitted with a second Gaussian component, produces a chi2/dof = 65.32/79 (0.83). The line center of this second Gaussian component is at 8.4 +/- 0.2 keV and the FWHM is 0.5 +0.7/-0.5 keV, with an equivalent width of 370 +/-310 eV.

The 1-10 keV flux of XTE J1859+226 at the time of this observation was 1.23 E-12 erg/cm^2/s and the 0.3-8 keV flux of 1.56 E-12 erg/cm^2/s was three orders of magnitude larger than the quiescent flux of this source (1.5 E-15 erg/cm^2/s) predicted by Tomsick et al. 2003. For the parameter estimates from literature of the distance to XTE J1859+226 and the mass of the central black hole, the Eddington fraction of the source during the observation was in the range of 0.03 - 0.30 %. Tomsick et al. 2009 found that at a similarly low Eddington fraction (0.14%), the black hole GX 339-4 shows evidence of an accretion disk truncated at more than 35 R_g, suggesting that at low luminosity the inner part of the disk is not present, unlike during high luminosity phases. While additional fits with reflection models suggest the possibility of an accretion disk truncated at large distances, for XTE J1859+226, the data are unable to significantly distinguish between a disk truncated at the ISCO or at larger radii.

We thank Fiona Harrison and the NuSTAR planning team for executing this observation.

References:

Tomsick, J. A., Corbel, S., Fender, R., et al. 2003, ApJL, 597, L133;

Tomsick, J. A., Yamaoka, K., Corbel, S., et al. 2009, ApJL, 707, L87.