The BeppoSAX view of bright Compton-thin Seyfert 2 galaxies

INAF – Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, 50125 Firenze, Italy Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138, USA

Corresponding author: risaliti@arcetri.astro.it

Received: 28 September 2001
Accepted: 30 January 2002

Abstract

We present the analysis of 31 observations (17 of which are published here for the first time) of 20 bright Compton thin Seyfert 2 s, in the 0.1–200 keV band, performed with the BeppoSAX satellite. The sample consists of all Seyfert 2 s in the BeppoSAX public archive, with a 2–10 keV flux higher than  erg cm^-2 s^-1. The good statistics available and the broad energy band permit a detailed study of the main continuum components of these sources, i.e. the primary power-law, the reflected component, the soft emission and the high-energy cut-off. The main results of our analysis are: (1) the 3–200 keV intrinsic power-law has a mean photon index , with a dispersion of . (2) The high-energy exponential cut-off at  keV is not an ubiquitous property of Seyfert galaxies: in ~30% of the objects the continuum power-law does not drop up to energies of 300 keV or more. (3) A reflected component is present in almost all the sources (17 out of 21). The small variations of this component with respect to the intrinsic continuum, in objects with multiple observations, suggests that the reflector is not the accretion disk, but must be located much farther from the nucleus. (4) The range of ratios between the reflected and intrinsic components suggests that the circumnuclear medium is not homogeneous, and a significant fraction of the solid angle is covered by a gas thicker than that along the line of sight. (5) The iron Kα line is present is all but one the sources. The equivalent width is in the typical range of Seyfert 1 s ( eV) in sources with low absorption ( cm^-2), and increases in more absorbed objects, as expected according to unified models. (6) The energy resolution of BeppoSAX is in general too low to measure the iron line width. However, in 6 cases we measured a significant line broadening.