Insight-HXMT detection of the hard-to-soft state transition in the black hole candidate EXO 1846-031

ATel #13036; Yi-Jung Yang (IHEP), Roberto Soria (UCAS/NAOC/Uni. of Sydney), Dave Russell (NYU Abu Dhabi), Guangcheng Xiao, Jinlu Qu, Shuang-Nan Zhang (IHEP), on behalf of the HXMT team
on 17 Aug 2019; 04:32 UT
Credential Certification: Roberto Soria (rsoria@physics.usyd.edu.au)

Tweet

A new outburst of the black hole candidate EXO 1846-031 was first reported by MAXI on July 23 (ATel #12968), followed by several reports in different wavebands (ATels #12969, #12976, #12977, #12991, #12992, #13012).

We triggered a ToO monitoring program on Insight-HXMT starting from 2019 August 2. EXO 1846-031 is being observed regularly, and can be detected by all three instruments onboard HXMT (LE: 1-15 keV; ME: 5-30 keV; HE: 20-250 keV). Here we report results from the observations taken on August 2 (MJD 58697.3416 at mid exposure), August 3 (MJD 58698.1730) and August 8 (MJD 58703.1950).

We fitted the spectra from all three instruments (1-250 keV) simultaneously. We find that both the August 2 and 3 spectra are well fitted with an absorbed broken power-law model (significantly better fit than a single power-law model). We also tried adding a disk blackbody component, but it does not improve the fit. In the August 2 spectrum, at high energies, a broken power law is a better fit than an exponentially cutoff power law; the two models are equivalent for the August 3 spectrum. The power-law break is at an energy E = (40 +/- 1) keV and E = (29 +/- 2) keV in the two spectra, respectively. The photon index below the break is Gamma = 1.84 +/- 0.01 and Gamma = 1.96 +/- 0.02, respectively. The photon index above the break remained constant (2.76 +/- 0.05 and 2.7 +/- 0.1, respectively). The steepening of the low-energy photon index and the cooling of the break energy as the flux increased are typical of transient black hole X-ray binaries approaching their peak luminosity in the hard state, before a state transition towards the thermal high/soft state.

The observed flux in the 1-10 keV band increased from 3.7E-9 erg/cm^2/s (August 2) to 4.2E-9 erg/cm^2/s (August 3). The observed flux in the full 1-250 keV band was 13.9E-9 erg/cm^2/s on August 2 and 12.3E-9 erg/cm^2/s the next day. The extrapolated unabsorbed model flux in the 0.5-10 keV band was 8.6E-9 erg/cm^2/s and 10.3E-9 erg/cm^2/s, respectively. At an assumed distance of 8 kpc, this corresponds to a 0.5-10 keV luminosity of 6.6E37 erg/s (August 2) and 7.9E37 erg/s (August 3). The 2-10 keV luminosity was 3.9E37 erg/s and 4.4E37 erg/s, respectively. The extrapolated luminosity in the 0.5-250 keV band was 1.4E38 erg/s both on August 2 and 3. Such luminosities are consistent with the peak of the hard state (around 10 per cent of the Eddington luminosity) just before a soft transition.

On August 8, the spectrum had softened substantially, and was well fitted with an absorbed disk-blackbody plus power-law model. The observed 1-10 keV flux was 7.3E-9 erg/cm^2/s, and the extrapolated 0.5-10 keV isotropic luminosity was 2.2E38 erg/s (8.4E37 erg/s in the 2-10 keV band). The power-law photon index was 2.45 +/- 0.03. The peak colour temperature of the disk was T_{in} = (0.93 +/- 0.09) keV, at an apparent inner radius r_{in} sqrt{cos theta} = (11.7 +/- 2.9) km. If the compact object is a black hole, this small radius is consistent with a high-inclination disk and/or near-maximal spin, in agreement with the conclusions of Miller et al. (ATel #13012) based on relativistic reflection models fitted to the NuSTAR data. In addition, an Fe Kalpha emission line was also detected by HXMT on August 8, centred at (6.8 +/- 0.1) keV, with a full-width-half-maximum ~ 0.1 keV and equivalent width ~ 45 eV.

In summary, the X-ray spectral properties are typical of canonical outbursts in low-mass black hole X-ray binaries, and suggest that EXO 1846-031 transitioned from the hard to the soft-intermediate state between August 2 and 8. New radio and sub-mm observations are strongly encouraged, to determine whether the jet (reported in ATels #12977 and #12992) is now quenched.

A sample of spectral models is shown here:
http://www.hxmt.org/images/atel/201908/Atel_EXO1846-031.png