GRO J1655-40: from ASCA and XMM-Newton Observations

TL;DR

This paper analyzes multiple X-ray observations of GRO J1655-40 in high/soft state, modeling the spectra to estimate black hole spin, spectral hardening factor, and distance, revealing variability across observations.

Contribution

It applies relativistic disk modeling to multiple datasets to estimate black hole parameters and explores the variability of spectral hardening factor and distance constraints.

Findings

Black hole spin estimated from spectral modeling.

Spectral hardening factor varies between observations.

Distance constraints depend on the spectral hardening factor.

Abstract

We have analysed four ASCA observations (1994--1995, 1996--1997) and three XMM-Newton observations (2005) of this source, in all of which the source is in high/soft state. We modeled the continuum spectra with relativistic disk model kerrbb, estimated the spin of the central black hole, and constrained the spectral hardening factor f_col and the distance. If kerrbb model applies, for normally used value of f_col, the distance cannot be very small, and f_col changes with observations.

Figures

Full text

11footnotetext: MPE, Postfach 1312, 85741 Garching, Germany, zhangx@mpe.mpg.de22footnotetext: Tsinghua Univ, 100084, Beijing, China33footnotetext: U. of Alabama, Huntsville, AL 35899, USA44footnotetext: NSSTC, Sparkman Dr. 320, Huntsville AL 35805, USA55footnotetext: MIT Kavli Inst. for Astro. and Space Research, 70 Vassar Street, Cambridge, MA 02139

1. Background

GRO J1655-40, the second microquasar (after GRS 1915-105), had X-ray outbursts in 1994-1995, 1996-1997, 2005. Its geometric parameters are considered best determined: mass $M_{\mathrm{BH}}=7.0\pm 0.2M_{\odot}$, inclination angle \theta=69.50\hbox{{}^{\circ}}\pm 0.08 (Orosz & Bailyn 1997), distance $D=3.2\pm 0.2$ kpc (Hjellming & Rupen 1995), which makes it a very good laboratory of studying black holes and environments.

The spin of the central black hole has been estimated by various authors with various methods (see, e.g., Zhang et al. 1997; Abramowicz & Kluźniak 2001; Aschenbach 2004; Shafee et al. 2006), and the reported value range from 0.2 (Abramowicz & Kluźniak 2001) to 0.996 (Aschenbach 2004).

In estimating black hole spin from continuum spectral modeling, the color correction factor $f_{\mathrm{col}}=T_{\mathrm{col}}/T_{\mathrm{eff}}$, is one of the key factors. The normally used value of $f_{\mathrm{col}}$ is 1.7, following Shimura & Takahara (1995), while many authors believe it should not be constant (see, e.g., Merloni et al. 2000). The distance is also very important. The widely accepted value $3.2\pm 0.2$ kpc was challenged by Foellmi et al. (2006), who gave an upper limit of 1.7 kpc.

2. Observations, data reduction and model fitting

We analysed three ASCA observations during the 1994–1995 and the 1996–1997 outbursts, and three XMM-Newton observations during the 2005 outburst, in all of which the source was in high/soft state. For ASCA, only GIS2 data were used, after gain correction and deadtime correction. For XMM-Newton, only Epic-pn data were used, after correction for rate-dependent Charge-Transfer-Efficiency (Sala et al. 2006).

The classical way of estimating black hole spin from the continuum spectral fitting is to fit the spectra with disk models, and obtain the spin directly or indirectly. All models take the source distance as parameter, and most models treat the disk as multi-temperature black-body rings and the derived spin value depends on the apparent/effective temperature ratio.

The relativistic disk model kerrbb in XSPEC was used in the fitting. We let $f_{\mathrm{col}}$ vary from 1.0 to 3.0, and $D$ vary from 1.0 kpc and 3.2 kpc. For each combination of $f_{\mathrm{col}}$ and D, we fitted the data sets and obtained a spin value, if the fit was acceptable ($\chi^{2}/dof<2$). The contour of the derived spin $a$ over $D$ and $f_{\mathrm{col}}$ are shown in the Fig. 1.

3. Conclusion

From Fig. 1 we can see,

1. for the normally used $f_{\mathrm{col}}$ value 1.7, kerrbb model does not favor small distance;
2. because the black hole spin and the source distance should be constant, $f_{\mathrm{col}}$ changes dramatically between these observations.