[ Previous | Next | ADS ]

Chandra Spectroscopy of the Black Hole GRS 1716-249 (X-ray Nova Ophiuchi 1993)

ATel #10296; J. M. Miller (Univ. of Michigan), A. Fabian (Cambridge), J. Kaastra (SRON), T. Kallman (NASA/GSFC), A. L. King (Stanford), M. Nowak (MIT), D. Proga (UNLV), J. Raymond (SAO), C. Reynolds (Univ. of Maryland), M. Reynolds (Univ. of Michigan), A. Zoghbi (Univ. of Michigan)
on 20 Apr 2017; 17:46 UT
Credential Certification: Jon Miller (jonmm@umich.edu)

Subjects: X-ray, Binary, Black Hole, Transient

Referred to by ATel #: 10371

We report on a Chandra/HETGS observation of the black hole X-ray binary GRS 1716-249 (Nova Ophiuchi 1993) made after the hard X-ray peak of its ongoing outburst (see ATEL #9876, #9895, #10036, #10196, #10236).

The source was observed for 30 ks starting on 2017 February 06, at 12:00:56 UT (obsID 20008). The ACIS-S array was operated in continuous clocking mode to prevent photon pile-up. Combined first-order MEG and HEG spectra and responses were created using CIAO 4.7 and corresponding calibration files. The MEG and HEG spectra were fit jointly, allowing a constant to float between them. The MEG was considered over the 0.7-6.0 keV band, and the HEG over the 1.2-8.5 keV band, determined by the location of chip gaps and sensitivity.

A simple power-law continuum, modified by ISM absorption via "tbabs" (with recommended cross sections and abundances; Wilms et al. 2000) gives N_H = 7.3(1) E+21 cm^-2, Gamma = 1.78(1), K_pow = 1.31(1), and chi^2/nu = 7641.78/6731 = 1.135 (1-sigma errors; numbers in parentheses indicate the error in the last digit). This model predicts an unabsorbed flux of 7.64(4) E-9 erg cm^-2 s^-1 (0.5-10 keV).

The addition of a disk blackbody component yields a significant improvement (chi^2/nu = 7441.57/6729 = 1.106). With this model, N_H = 6.5(1) E+21 cm^-2, kT = 0.59(1) keV, K_dbb = 83(9), Gamma = 1.53(3), and K_pow = 0.80(4). This model gives an unabsorbed flux of 7.1(3) E-9 erg cm^-2 s^-1 (0.5-10.0 keV).

It is notable that the column density is a few times higher than predicted by the HEASARC "nH" tool, based on Dickey & Lockman (N_H = 2.5 E+21 cm^-2, Dickey & Lockman 1990). No plausible continuum model is able to fit the spectra well when the predicted column density is enforced.

For a distance of 2.4 kpc (Della Valle et al. 1994), the fluxes translate to luminosities in the 4.9-5.3 E+36 erg s^-1 range. If this distance estimate is robust, the apparent disk temperature is anomalously high, and would ordinarily be observed at a luminosity about 100 times higher. The anomalous temperature may be physical, and it could also instead be an artifact of fitting a more complex Comptonization spectrum with simple components.

In our view, the disk component should likely be viewed with caution. Opposing orders in the MEG and HEG are found to have local flux variations that differ at the 10 percent level. Such effects are known to be more pronounced for very hard spectra, and smaller for thermal spectra. In this observation, the effect is particularly strong near to 2 keV, which makes the disk component suspicious since blackbody fluxes peak at 2.8*kT. At higher energy, weak features consistent with narrow ionized Fe emission lines in the HEG+1 spectrum are inconsistent with the HEG-1 spectrum, possibly also owing to CTI.

We thank Chandra director Belinda Wilkes for executing this observation, and Norbert Schulz and David Huenemoerder for insights regarding the observation particulars.


Delle Valle, M., et al., 1994, A&A, 290, 803

Dickey, J., & Lockman, F., 1990, ARAA, 28, 215

Wilms, J., et al., 2000, ApJ, 542, 914