In recent years, statistical studies of complete samples of extragalactic radio sources have suggested a unified scheme for radio-loud AGN. According to this picture, radio galaxies, radio-loud quasars, and blazars are the same physical objects seen at decreasing angles with respect to the jet axis (Antonucci 1993; Urry & Padovani, 1995). A similar scenario has been proposed for radio-quiet AGN (i.e., Seyfert 1 and Seyfert 2 galaxies; Antonucci & Miller 1985; Miller & Goodrich 1990). The common motivation for such unified schemes is that the radiation field is almost certainly anisotropic , which automatically implies orientation-dependent observational properties. Some degree of anisotropy could be caused by an opaque circumnuclear torus, which for some lines of sight may prevent direct view of the active nucleus and of the broad emission line region. In radio-loud AGN, additional anisotropy is likely due to relativistic beaming of the continuum produced in the jet.
Broadly speaking, a radio-loud source should show a featureless (or almost featureless) spectrum when observed face-on, a Seyfert 1-like spectrum at intermediate angles and a Seyfert 2-like spectrum when seen edge-on. Broad line Radio Galaxies (BLRG) and Narrow Line Radio Galaxies (NLRG) are therefore considered the radio-loud counterpart of Seyfert 1 and Seyfert 2 galaxies, respectively. However, it is still unclear whether the accretion processes are the same in radio-loud AGN and in Seyfert galaxies. Indeed, Rees et al. (1982) suggested that in radio galaxies the accreting gas flow is not in a cold, geometrically thin, disk configuration, but it rather forms a hot, geometrically thick, ion-supported torus, characterized by low radiative efficiency. The Advection Dominated Accretion Flow (ADAF) models, more recently proposed by other authors (see Narayan, Mahadevan & Quataert 1998, hereinafter NMQ, for a recent review), follow similar lines of thought. Shapiro, Lightman & Eardley (SLE, 1976) found a solution to the accretion problem that, in many respects, resembles the later ion-supported torus, and suggested its relevance for the black hole candidate Cygnus X-1. In the SLE solution, however, the energy produced in the flow by viscosity is locally radiated and advection is implicitly assumed to be negligible. A hot ion torus, surrounded by a geometrically thin cold accretion disk irradiated by the hard radiation produced by the torus, was proposed by Chen and Halpern (1989) in the context of BLRG showing double peaked emission lines, notably Arp 102B and 3C 390.3.
From the observational point of view the situation is still confused. Ginga data of radio-loud objects showed uncertain detection of the iron line and/or the reflection component (Nandra & Pounds 1994). ASCA observations at better energy resolution showed the presence of the iron line in several radio-loud AGNs, but did not constrain the reflection hump (Eracleous et al. 1997, Allen & Fabian 1992, Grandi et al. 1997a, Yamashita & Inoue 1996). It is then unclear whether a Seyfert-like nucleus is present in radio galaxies. The wide energy band (-100 keV) covered by the instruments on board BeppoSAX is particularly appropriate to address the problem. For this reason, a BeppoSAX Core Program has been dedicated to the spectral study of bright radio galaxies (-10 keV10-11 erg cm-2 sec-1). Here we present the observation of 3C 390.3, the first source observed.
3C 390.3 is a well known Broad-Line Radio Galaxy (z=0.057) with an FRII morphology and a core showing superluminal motion (Alef et al. 1996). Its spectrum is characterized by double-peaked emission lines in the optical and UV bands (Eracleous and Halpern 1994, Zheng 1996, Wamsteker et al. 1997). The UV bump, typically observed in most Seyfert galaxies, is weak or even absent (Wamstecker et al. 1997). The Einstein-IPC data revealed the presence of a strong intrinsic absorption (Kruper et al. 1990). Ghosh and Soundararajaperumal (1991) claimed the presence of a soft excess in the EXOSAT data, but their results have not been confirmed by the later ROSAT and ASCA observations (Walter et al. 1994, Leighly et al. 1997). At higher energies, analysis of Ginga data produced ambiguous results. Inda et al. (1994) resolved the iron line at 6.4 keV but not the reflection component. Nandra & Pounds (1994) could give only an upper limit on the equivalent width of the emission line, but revealed a weak reflection component of small covering factor (). ASCA confirmed the presence of the iron K emission line (Eracleous et al. 1996, Leighly et al. 1997), but could not constrain the reflection hump, most probably because of the limited energy range. 3C 390.3 has also been detected by OSSE in the soft ray domain, above 50 keV (Dermer & Gehrels 1995).
© European Southern Observatory (ESO) 1999
Online publication: March 1, 1999