X-ray transient sources

As an example of an X-ray nova, we shall describe the system GS 1124$-$683, a.k.a. Nova Muscae. Its behavior was characteristic of many X-ray transients observed prior to the launch of the Rossi X-ray Timing Explorer (RXTE). In the span of less than 10 days, the 1-10keV source flux rose by more than a factor of 100 [69]. At its maximum, the X-ray spectrum exhibited a strong, soft quasi-thermal component and a power law tail with $\Gamma \approx 2.5$ [56,70]. The flux then exponentially decayed with a time constant of $\approx 30$days for the soft X-ray flux and $\approx 13$days for the hard flux. The X-ray variability was also seen to decrease as the power-law component flux decreased [56,70]. Approximately 5 months after maximum flux, the source switched into a `hard state' with $\Gamma \approx 1.7$ and then continued to decay.

After the X-ray source decayed away, optical observations revealed modulations on an orbital period of 10.4 hours. Combining this with velocity measurements of the companion star, the lower limit to the compact object mass was determined to be $3.1\hbox{$\rm\thinspace M_{\odot}$}$ and its most likely value was determined to be $6\hbox{$\rm\thinspace M_{\odot}$}$ [71]. Prior to those observations, several other quiescent X-ray novae had revealed dynamical evidence for massive compact objects. GS 2023+33, a.k.a. V404 Cyg was found to have a compact object mass of $10\hbox{$\rm\thinspace M_{\odot}$}$, and A0620$-$00 was found to have a compact object mass of $7.3\hbox{$\rm\thinspace M_{\odot}$}$ (lower limit $3.2\hbox{$\rm\thinspace M_{\odot}$}$) [72,73]. There have been approximately two dozen such transients, roughly half of which have measured masses. To date, our greatest source of information about the mass distribution of stellar mass black holes has come from optical observations of transient systems in quiescence [74]. The X-ray nova outburst acts like a beacon calling attention to the system, while the subsequent decay into quiescence allows the orbital parameters of the binary system to be measured via observations of the typically very faint companion star.

Figure 3: Sampling of lightcurves (1s bins) from various RXTE observations of the Galactic ``micro-quasar'' GRS 1915+105 [75]. Bottom panels show 2-60keV count rates in units of thousand of counts per second. (Counting noise is negligible for these observations.) Top panels show ``hardness ratios'' over the course of these observations. HR$_1$ is defined as the ratio of the 5-13keV count rate to the 2-5keV count rate, while HR$_2$ is defined as the ratio of the 13-60keV count rate to the 2-5keV count rate.

X-ray transients that deviate strongly from the `fast-rise/exponential decay' behavior of Novae Muscae also have been discovered. A wide variety of lightcurve morphologies have been observed by RXTE [75]. Among the most interesting examples are GRO J1655$-$40 and GRS 1915+105, both of which were found to exhibit relativistic jets of plasma that were observed in the radio band with inferred velocities in excess of 0.9$c$ [76,77]. Both of these sources have shown dramatic variability (see Fig. 3), and the latter source remains in an active X-ray state to the present day. It has since been realized that the formation of radio jets or outflows is often associated with accretion in stellar mass black hole systems [78]. Typically, the radio flux is intermittent and `optically thin' (i.e., $F_\nu \propto \nu^{-\alpha}$, with $\alpha = \Gamma - 1 > 0$, where $F_\nu$ is the radio energy flux per unit frequency, ${\rm ergs~cm^{-2}~s^{-1}~Hz^{-1}}$) at the highest fluxes within the soft state, becomes quenched at lower luminosities in the soft state, and then re-occurs in the hard state. The radio spectrum in the X-ray hard state is usually persistent and `flat' or `inverted' (i.e. $F_\nu \propto \nu^{-\alpha}$ with $\alpha \le 0$), and the radio flux is positively correlated with the X-ray flux. [59,78,79,80,81,82].

There is currently a great deal of controversy regarding the underlying mechanisms behind these state transitions, the formation of the radio outflow, as well as the basic geometry of the accretion flow itself. Observations of relativistically distorted X-ray emission lines, the prime focus of this review, are being used as probes of the accretion flow properties near the event horizons of stellar mass black holes in binary systems precisely for the purpose of gaining more information about these issues.