Fe Kαand Fe Kβline detection in theNuSTARspectrum of the ultra-bright Z source Scorpius X–1

Abstract

Context. Low-mass X-ray binaries hosting a low-magnetised neutron star, which accretes matter via Roche-lobe overflow, are generally grouped into two classes called Atoll and Z sources after the path described in their X-ray colour-colour diagrams. Scorpius X–1 is the brightest persistent low-mass X-ray binary known so far, and it is the prototype of the Z sources. Aims. We analysed the first NuSTAR observation of this source to study its spectral emission, exploiting the high-statistics data collected by this satellite. The colour-colour diagram shows that the source was probably observed during the lower normal and flaring branches of its Z track. We separated the data from the two branches in order to investigate the evolution of the source along the track. Methods. We fitted the 3−60 keV NuSTAR spectra using the same models for the two branches. We adopted two descriptions for the continuum: in the first case, we used a blackbody and a thermal Comptonisation with seed photons originating in the accretion disc, and in the second case, we adopted a disc-blackbody and a Comptonisation with a blackbody-shaped spectrum of the incoming seed photons. A power-law fitting of the high-energy emission above 20 keV was also required in both cases. Results. The two models provide the same physical scenario for the source in the two branches: a blackbody temperature between 0.8 and 1.5 keV, a disc-blackbody with a temperature between 0.4 and 0.6 keV, and an optically thick Comptonising corona with an optical depth between 6 and 10 and a temperature about 3 keV. Furthermore, two lines related to the Kα and Kβ transitions of the He-like Fe XXV ions were detected at 6.6 keV and 7.8 keV, respectively. A hard tail modelled by a power law with a photon index between 2 and 3 was also required for the two models.

Insight-HXMT Detections of Hard X-Ray Tails in Scorpius X-1

“…Panels (a) and (b) of Figure 6 show the two representative fittings of model 2 and model 3, respectively. As listed in Table 3, the values of kT s and kT e obtained from model 2 or model 3 are ∼0.4 and ∼3 keV, respectively, basically being consistent with those obtained from fitting the NuSTAR spectrum of this source (Mazzola et al 2021). The kT e of Sco X-1 is only about a few keV, which is much lower than that of .…”

“…The previous spectral studies of Sco X-1 authenticate a line component in its low-energy range of ∼5.5-7.5 keV (White et al 1985;D'Aí et al 2007;Church et al 2012;Ding et al 2021), most likely being the Fe Kα line. In particular, two line components were detected in the NuSTAR spectrum of Sco X-1 (Mazzola et al 2021). The two lines are centered at 6.6 and 7.8 keV, respectively, being confirmed to be the Fe Kα and Fe Kβ lines, respectively.…”

“…It is noteworthy that in the ME energy range, in the residual distributions, a structured feature is visible in ∼20-25 keV, as demonstrated in Figures 5 and 6. Such a profile, seeming to inform a certain emission line in this energy interval, is not present in the RXTE or NuSTAR spectral fittings of Sco X-1 (D' Aí et al 2007;Ding et al 2021;Mazzola et al 2021), so this phenomenon cannot result from the inherent properties of this source. To calibrate the energy scale and resolution of the ME instrument of Insight-HXMT, two 241 Am sources are contained in each ME detector box, then the four nuclear lines of 241 Am in ∼14-27 keV, i.e., the ME energy range, are used to determine the Energy-Channel (E-C) relation of ME, and finally the E-C relation is verified by a Ag emission line with the peak at 22.5 keV (Li et al 2020).…”

“…Therefore, the Eastern and Western models are written in the forms of BB+LINE+DISKBB and of BB+LINE+CPL, respectively. Taking into account the interstellar absorption by fixing N H at 0.3 × 10 22 atoms cm −2 (Christian & Swank 1997;Church et al 2012;Ding et al 2021;Mazzola et al 2021), we apply the two models to fit the LE+ME spectrum in 2-35 keV. When performing fitting, taking into account the calibration differences between the LE and ME instruments, the models are multiplied by a constant.…”

“…The line centroid spans a range of ∼6-6.7 keV, and thus the line could be the Fe Kα one. Unlike the situation of the fitting of the NuSTAR spectrum (Mazzola et al 2021), in the fitting of the low-energy spectrum of Insight-HXMT only one line component is needed, not two line components. The EW of the line changes in a wide range of ∼50-300 eV, implying that the line profile evolves remarkably, as shown in Figure 7.…”