Volume 448, Number 3, March IV 2006
|Page(s)||1125 - 1137|
|Section||Stellar structure and evolution|
|Published online||03 March 2006|
Bimodal spectral variability of Cygnus X-1 in an intermediate state
Centre d'Étude Spatiale des Rayonnements (CNRS/UPS/OMP), 31028 Toulouse, France e-mail: Julien.Malzac@cesr.fr
2 Laboratoire d'Astrophysique Observatoire de Grenoble, BP 53, 38041 Grenoble Cedex 9, France
3 Service d'Astrophysique, CEA-Saclay, Bat. 709, L'Orme des Merisiers, 91191 Gif-sur-Yvette, Cedex, France
4 APC-UMR 7164, 11 place M. Berthelot, 75231 Paris, France
5 Cavendish Laboratory, University of Cambridge, Madingley Road, Cambridge CB3 0HE, UK
6 AIM - Astrophysique Interactions Multi-échelles (Unité Mixte de Recherche 7158 CEA/CNRS/Université Paris 7 Denis Diderot), CEA-Saclay, Bât. 709, L'Orme des Merisiers, 91191 Gif-sur-Yvette Cedex, France
7 INAF - Osservatorio Astronomico di Brera, via E. Bianchi 46, 23807 Merate, Italy
8 INTEGRAL Science Data Center, Chemin d'Écogia 16, 1290 Versoix, Switzerland
Accepted: 16 November 2005
We report the results of an observation of Cygnus X-1 performed on June 7–11, 2003 with INTEGRAL that we combine with simultaneous radio observations with the Ryle telescope. Both spectral and variability properties of the source indicate that Cygnus X-1 was in an Intermediate State. The INTEGRAL spectrum shows a high-energy cut-off or break around 100 keV. The shape of this cut-off differs from pure thermal Comptonisation, suggesting the presence of a non-thermal component at higher energies. The average broad band spectrum is well represented by hybrid thermal/non-thermal Comptonisation models. However, models with mono-energetic injection, or models with an additional soft component are favoured over standard power-law acceleration models. During the 4 day long observation the broad band (3–200 keV) luminosity varied by up to a factor of 2.6 and the source showed an important spectral variability. A principal component analysis demonstrates that most of this variability occurs through 2 independent modes. The first mode consists in changes in the overall luminosity on time scale of hours with almost constant spectra (responsible for 68% of the variance) that are strikingly uncorrelated with the variable radio flux. We interpret this variability mode as variations of the dissipation rate in the corona, possibly associated with magnetic flares. The second variability mode consists in a pivoting of the spectrum around ~10 keV (27% of the variance). It acts on a longer time-scale: initially soft, the spectrum hardens in the first part of the observation and then softens again. This pivoting pattern is strongly correlated with the radio (15 GHz) emission: radio fluxes are stronger when the INTEGRAL spectrum is harder We propose that the pivoting mode represents a “mini” state transition from a nearly High Soft State to a nearly Low Hard State, and back. This mini-transition would be caused by changes in the soft cooling photons flux in the hot Comptonising plasma associated with an increase of the temperature of the accretion disc. The jet power then appears to be anti-correlated with the disc luminosity and unrelated to the coronal power. This is in sharp contrast with previous results obtained for the Low Hard State, suggesting a different mode of coupling between the jet, the cold disc, and the corona in Intermediate States. From this interpretation we also infer that the bolometric luminosity jumps by a factor of about 2 during the transition hard to soft, suggesting a radiatively inefficient accretion flow in the Low Hard State.
Key words: gamma-rays: observations / black hole physics / radiation mechanisms: non-thermal / X-rays: binaries / radio continuum: stars / X-rays: individuals: Cygnus X-1
© ESO, 2006
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.