Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 329, 965-970 (1998)

Previous Section Next Section Title Page Table of Contents

3. Discussion

RXTE-ASM monitoring shows that KS 1731-260 is a source of persistent X-ray emission. This is not too surprising given the repeated detections of the source since its discovery in 1988. The RXTE-ASM 2-12 keV light curve of KS 1731-260 is shown in Fig. 5. The mean count rate (150 mCrab; 2-12 keV) translates to a 1-20 keV flux of [FORMULA] ergs s-1 cm-2  comparable to those measured by TTM and ROSAT. Assuming a distance of 8.3 kpc and the spectral parameters observed by RXTE (Smith et al. 1997), this flux corresponds to a persistent X-ray luminosity of about [FORMULA] ergs s-1. This relatively high luminosity is associated with a soft spectrum and no hard X-ray emission. For indication, the RXTE spectrum which is well fit by a 5.5 keV thermal Bremsstrahlung model (Smith et al. 1997) predicts a hard X-ray flux which is about three orders of magnitude below the 40-150 keV flux observed by SIGMA (Barret et al. 1992). In its five year monitoring of the Galactic center region, SIGMA detected a hard X-ray tail from KS 1731-260 during only one one-day observation (Barret et al. 1992; Goldwurm et al. 1994). Since in X-ray bursters hard X-ray emission seems to be associated with low luminosity states (e.g. 4U1608-522: Zhang et al. 1996), both the SIGMA and RXTE results suggest therefore that KS 1731-260 spends most of its time in a high/soft luminosity state.

[FIGURE] Fig. 5. The 2-12 keV light curve of KS 1731-260 as observed by the All Sky Monitor aboard the ROSSI X-ray Timing Explorer. The whole set of data spans a year (from 1997/1/6 to 1996/1/13) and shows that KS 1731-260 is a persistent X-ray source at the level of [FORMULA] mCrab. Each point is a 1-day average. The dot-dashed line corresponds to the flux of a 100 mCrab source for the ASM. Data have been retrieved from the Web site http://space.mit.edu/XTE/ASM-lc.html at Massachussets Institute of Technology. The first gap in the data is caused by the problems encountered by the ASM at the beginning of the mission. The second gap (from MJD 50395 to 50442) is caused by the sun traversing the Galactic Center region. During this interval, the sun was too close to KS 1731-260 for the ASM to observe it.

As said above, the main goal of our HRI observation was to accurately locate the source in order to test the two optical counterparts proposed by Cherepashchuk et al. (1994). We show in Fig. 6 the ROSAT HRI error box overlayed on ESO R plate No 520. The photographic plate was scanned at the MAMA with a pixel size of 0.675" and retrieved at CDS using the ALADIN interactive sky atlas (Bartlett et al. 1994). A set of local PPM standards allowed the absolute positioning of the HRI center with an accuracy of [FORMULA] 0.3". None of the two former proposed candidates, objects 1 and 2 listed by Cherepashchuk et al. (1994) is now compatible with the improved HRI position. In particular, object 2, identified with variable star No 2547 from the catalogue of Terzan & Gosset (1992) cannot be associated with KS 1731-260.

[FIGURE] Fig. 6. The HRI error circle (for HRI1) overlayed on ESO R plate No 520. Objects 1 and 2 quoted as possible candidates by Cherepashchuk et al. (1994) are now excluded from the improved HRI error circle. North is at top and East to the left.

Assuming that the photoelectric absorption seen in the soft X-ray band by ROSAT is entirely of interstellar origin implies AV [FORMULA] 7.2 [FORMULA] 1.1 (Predehl & Schmitt 1995). This corresponds to AR [FORMULA] 5-6 for the effective wavelength of the ESO R plate but only AJ [FORMULA] 2.0 [FORMULA] 0.3 and AH [FORMULA] 1.25 [FORMULA] 0.20 in the infrared bands. Unfortunately, no photometric calibration is yet available for the photographic plate. However, with an estimated R (630-690nm) limiting magnitude of [FORMULA] 21-22 (West 1984), the faintest objects in the error circle seen on the ESO plate have R-J [FORMULA] 6 or more, indicating a high reddening in the field, probably consistent with that assumed for KS 1731-260.

If the intrinsic J-H colors of KS 1731-260 are those of a hot OB star ((J-H)0 = -0.16, Koorneef 1983), the counterpart should have a reddened J-H = 0.59 [FORMULA] 0.11. This assumption holds independently from the low or high mass X-ray binary nature of the source since X-ray heating in a low-mass system probably still dominates near-infrared emission. Among the objects listed in Table 1, only D, H and J may have large enough errors on their J-H color index to be compatible with those of the expected blue counterpart. This would imply that the counterpart is fainter than H [FORMULA] 14.8 and has an absolute H magnitude fainter than -1.0 (d = 8.3 kpc). Ignoring any constraints on the infrared colors, the H magnitude of the brightest candidate in the error circle (object B) implies an absolute H magnitude fainter than -2.7 (d = 8.3 kpc). These upper limits on the infrared brightness of KS 1731-260 clearly argue in favour of a low-mass companion as for all other X-ray bursters. With an absolute R magnitude of MV = 0 to 6 (corresponding to MH = -1 to 5) observed for X-ray bursters (van Paradijs & Mc Clintock 1994) the optical counterpart may have H magnitude in the range of 14 to 21 at a distance of 8 kpc and could therefore be within the reach of H band spectroscopy.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: December 16, 1997