Astron. Astrophys. 338, 781-794 (1998)

1. Introduction

The spectra of active galactic nuclei (AGNs) at high energies (3 keV) provide important information on the physics of the nuclear region. Unfortunately, the limited sensitivity of current and past X-ray missions has usually restricted the investigation to bright AGNs. Little is known about the hard X-ray properties of low luminosity AGNs. This problem concerns particularly Seyfert 2s, since they are generally weaker than their type 1 counterparts. Yet, the hard X-ray properties of type 2 Seyferts are most interesting, since they provide important information on the obscuration affecting this class of objects.

Several observational data indicate that type 2 Seyfert nuclei suffer significant obscuration along our line of sight. The nature of the obscuring medium is matter of debate. The unified model (Antonucci 1993) ascribes the obscuration of Sy2 nuclei to a gaseous pc-scale circumnuclear torus. According to this model Sy1s and Sy2s would be identical physical objects, while the orientation of the line of sight with respect to the torus would be responsible for the obscuration of the BLR and of the nuclear engine (X-UV source) in type 2 Seyferts. Hard X-ray spectra are probably the best observational tool to directly measure the absorption affecting Sy2 nuclei. Indeed, Sy2 spectra in the 2-10 keV range show evidence for a power law component similar to that observed in Sy1s (photon index 1.7) and a cutoff due to photoelectric absorption. The latter indicates absorbing column densities between and cm^-2 (e.g. Awaki 1991, Ueno et al. 1996, Smith & Done 1996, Turner et al. 1997a), often ascribed to the obscuring torus.

In some sources the absorbing column density is so high (N_H 10²⁴cm^-2) that it is optically thick to Compton scattering. In this case the direct component is completely absorbed in the 2-10 keV range. However, the nuclear radiation can be Compton reflected by cold material surrounding the X-ray source (possibly the same torus responsible for the obscuration) or scattered by free electrons in a highly ionized warm gas. If these media extend outside the absorbing torus, then signatures of the X-ray nuclear activity are observable via the reflected/scattered continuum. Also, the cold reflecting medium produces a fluorescence iron line at 6.4 keV, while He- and H-like iron in the warm scattering medium should emit lines at 6.7 and 6.96 keV. Since the direct continuum is completely suppressed in Compton thick sources, these iron lines are characterized by equivalent widths ( keV) larger than in Sy1s or in Compton thin Sy2s. Until a few years ago NGC1068 was the only Compton thick source known. A few more sources of this class have been discovered recently (see Matt 1997 for a review).

As anticipated above, Sy2s in early spectroscopic studies were mostly selected from former all-sky X-ray surveys, hence these studies were generally biased for X-ray bright objects. Very likely, this selection criterion resulted in a bias in favor of low N_H Sy2s. Later on, hard X-ray spectroscopic studies (mostly by means of ASCA) probed fainter samples of AGNs (e.g. Turner et al. 1997a). However, many of the Sy2s observed by ASCA were selected amongst sources known to show broad lines in polarized light (Awaki et al. 1997). This selection criterion might introduce a bias for low N_H as well. Heisler et al. (1997) showed that the detectability of polarized broad lines is related to the obscuration of the nuclear region.

Summarizing, former X-ray spectroscopic surveys were seriously biased against heavily obscured Sy2s and, therefore, they are not suitable to study the real distribution of the absorbing column densities N_H.

The knowledge of the distribution of N_H in Sy2s is important to understand the nature of their obscuring medium, which has implications for the unified model. Also, the distribution of N_H is relevant to the synthesis of the X-ray background. Indeed, obscured AGNs are thought to contribute to most of the high energy ( 2 keV) extragalactic background (Comastri et al. 1995, Madau et al. 1993).

© European Southern Observatory (ESO) 1998

Online publication: September 17, 1998
helpdesk.link@springer.de