Support ATel At Patreon

[ Previous | Next | ADS ]

INTEGRAL hard X-ray spectroscopy of AT2018cow: preliminary detection of a cutoff at ~40 keV.

ATel #11788; C. Ferrigno (ISDC, CH), E. Kuulkers (ESA/ESTEC, NL), D. Goetz (CEA, Saclay), V. Savchenko, E. Bozzo (ISDC, CH), L. Ducci (IAAT, Germany / ISDC, CH), V. Beckmann (CNRS/IN2P3, France), W. Hermsen (SRON, NL), C. Sanchez-Fernandez, G. Belanger (ESA/ESAC, Spain), P. Ubertini (IAPS/INAF, Italy), R. Diehl (MPE, Germany), A. Lutovinov (IKI, Russia) A. Malizia (OAS-Bologna/INAF, Italy)
on 27 Jun 2018; 16:09 UT
Credential Certification: Carlo Ferrigno (Carlo.Ferrigno@unige.ch)

Subjects: X-ray, Gamma Ray, Supernovae, Transient

Referred to by ATel #: 11813, 11843

INTEGRAL is performing a public target of opportunity observation of the transient AT2018cow (ATel #11727) since 2018-06-22 18:38 UTC. We report here on the first 307 ks of on-source time taken until 2018-06-27 04:58 UTC.

The source is detected by IBIS/ISGRI at a signal to noise ratio of 7.2 in the 30-100 keV range. The IBIS/ISGRI spectrum can be described with a power law with photon index Γ=1.7±0.6 and 20-100 keV flux 6 ±2 × 10-11 erg/s/cm2 (uncertainties at 90% confidence level in the whole telegram).

We note that the NuSTAR spectrum above 15 keV give a hard power law with photon index Γ=0.5 (ATel #11775), which is not consistent with the INTEGRAL spectrum, as described above. Therefore, we determined an average broad-band high-energy spectrum by combining the data of AT2018cow as measured by Swift/XRT (22-25 June, see also ATel #11739), NuSTAR (23 June, ATel #11775), and INTEGRAL/IBIS/ISGRI (22-27 June). To account for source variability, we leave the intercalibration constant between the instruments as a free parameter. The broad-band spectrum can be well fit with a broken power law with exponential cutoff at high energy.

The best-fit model parameters are the following:
Parameter Value
NH (cm-2) 2.0±1.8 ×1020
Ecut (keV) 39+29-12
Γ1 1.41±0.08
Ebreak (keV) 14.4±0.7
Γ2 -0.3±0.3
Flux (0.3-15 keV)
erg/s/cm2
1.08±0.06 ×10-11
Flux (15-100 keV)
erg/s/cm2
5.9±0.7 ×10-11
χ2red./d.o.f. 1.32/284
A plot of the unfolded spectrum with residuals can be found at this link. Residuals at 7-10 keV in the NuSTAR spectrum are present, but their discussion is beyond the scope of this communication.

To estimate the significance of the high-energy cutoff, we jointly model only the NuSTAR and INTEGRAL spectra above the power-law break, i.e., we limited NuSTAR data above E> 15 keV. If we use a simple power law, we obtain χ2/d.o.f. = 86.9/64; if we add a cutoff, we obtain χ2/d.o.f. = 64.2/63. An F-test yields a significance of 4.3σ for the presence of a spectral cutoff (i.e., the probability that the χ2 improvement is due to chance is 1.3×10-5). More observations are planned and they might improve the significance of this detection.

Swift/XRT data are processed using the service by the UK Swift Science Data Centre at the University of Leicester (Evans et al. 2009, MNRAS, 397, 1177)