Comprehensive Spectral Analysis of Cyg X-1 using RXTE Data
We analyse a large number ($> 500$) pointed RXTE observations of Cyg X-1 and model the spectrum of each one. A subset of the observations for which there is simultaneous reliable measure of the hardness ratio by the All Sky Monitor, shows that the sample covers nearly all the spectral shapes of Cyg X-1. The relative strength, width of the Iron line and the reflection parameter are in general correlated with the high energy photon spectral index $\Gamma$. This is broadly consistent with a geometry where for the hard state (low $\Gamma \sim 1.7$) there is a hot inner Comptonizing region surrounded by a truncated cold disk. The inner edge of the disk moves inwards as the source becomes softer till finally in the soft state (high $\Gamma > 2.2$) the disk fills the inner region and active regions above the disk produce the Comptonized component. However, the reflection parameter shows non-monotonic behaviour near the transition region ($\Gamma \sim 2$), suggestive of a more complex geometry or physical state of the reflector. Additionally, the inner disk temperature, during the hard state, is on the average higher than in the soft one, albeit with large scatter. These inconsistencies could be due to limitations in the data and the empirical model used to fit them. The flux of each spectral component is well correlated with $\Gamma$ which shows that unlike some other black hole systems, Cyg X-1 does not show any hysteresis behaviour. In the soft state, the flux of the Comptonized component is always similar to the disk one, which confirms that the ultra-soft state (seen in other brighter black hole systems) is not exhibited by Cyg X-1. The rapid variation of the Compton Amplification factor with $\Gamma$, naturally explains the absence of spectra with $\Gamma < 1.6$, despite a large number having $\Gamma \sim 1.65$.
💡 Research Summary
This paper presents a systematic spectral study of the persistent black‑hole binary Cygnus X‑1 using more than five hundred pointed observations obtained with the Rossi X‑ray Timing Explorer (RXTE). For each observation the authors fit a phenomenological model consisting of a multicolour disc blackbody, a thermal Comptonisation component (represented by a power‑law with a high‑energy cutoff), a reflection continuum, and an iron Kα Gaussian line. The key spectral index, the high‑energy photon index Γ, is used to trace the source’s state from the canonical hard regime (Γ≈1.7) to the soft regime (Γ>2.2).
A subset of the data with simultaneous All‑Sky Monitor (ASM) hardness ratios demonstrates that the sample spans virtually the entire range of spectral shapes exhibited by Cyg X‑1. The analysis reveals several robust correlations: (i) the equivalent width and width of the Fe Kα line increase with Γ, (ii) the reflection scaling factor R grows as the spectrum softens, and (iii) the fluxes of the disc, Comptonised, and reflected components all rise monotonically with Γ. These trends are consistent with the widely‑adopted picture in which a truncated, cold accretion disc surrounds a hot inner Comptonising region in the hard state; as the source softens the disc inner radius moves inward, eventually filling the innermost region in the soft state, while magnetic or coronal active regions above the disc produce the Comptonised emission.
However, the study also uncovers deviations from this simple geometry. Near the transition region (Γ≈2) the reflection factor R shows a non‑monotonic dip, suggesting either a rapid change in the ionisation state of the reflector, a more complex disc‑corona configuration, or transient warping of the inner disc. Moreover, the fitted inner‑disc temperature (kT_in) is, on average, higher in the hard state than in the soft state, contrary to naïve expectations. The authors attribute this to limitations of the RXTE bandpass (≈3–25 keV) and the use of an empirical model with several parameters held fixed, which can bias temperature estimates.
A particularly striking result concerns the Compton amplification factor A. The authors demonstrate that A increases sharply as Γ approaches 1.6, effectively preventing the appearance of spectra with Γ<1.6 despite a large number of observations clustering around Γ≈1.65. This rapid variation of A with Γ provides a natural explanation for the observed lower bound on the photon index.
The paper also addresses the issue of hysteresis, a phenomenon observed in many transient black‑hole systems where the spectral state depends on the direction of luminosity change. In Cyg X‑1, the flux of each spectral component correlates tightly with Γ, and no hysteresis loop is evident, indicating that the source follows a unique, monotonic track in the hardness–intensity diagram. In the soft state the disc and Comptonised fluxes are comparable, confirming that Cyg X‑1 never reaches the ultra‑soft state seen in brighter transients.
Overall, the work provides a comprehensive, statistically robust picture of Cyg X‑1’s spectral evolution, confirming many aspects of the truncated‑disc/inner‑hot‑flow paradigm while highlighting subtleties—such as the non‑monotonic reflection behaviour and the unexpected disc temperature trend—that point to a more intricate physical environment. The authors acknowledge that the empirical fitting approach and the limited energy coverage of RXTE impose systematic uncertainties, and they suggest that future observations with higher spectral resolution and broader bandpasses (e.g., NICER, NuSTAR, XRISM) combined with physically motivated reflection and Comptonisation models will be essential to resolve these outstanding issues and to refine our understanding of state transitions in accreting black‑hole binaries.