SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 363, 1055-1064 (2000)

Previous Section Next Section Title Page Table of Contents

2. Compton scattering in AGN

It is widely believed that the production of soft X-ray excess in AGN (Active Galactic Nuclei), and hard X-ray tail in both AGN and GBHs (Galactic Black Holes) continuum spectra are caused by Comptonization.

The observed hard X-ray spectra of those objects consist of primary power-law X-ray radiation emitted from the hot plasma, and Compton reflection bump (for a review see Mushotzky et al. 1993 for AGN, and Tanaka & Lewin 1995 for GBHs). The former component results from Compton cooling of hot, probably thermal plasma by soft photons incoming from the accretion disk. Compton reflection bump exhibits [FORMULA] line with energy about 6.4 keV, which implies that the reflecting matter is cold, [FORMULA] K. High energy cut-off is observed at [FORMULA] keV. The latter feature results from two physical reasons. First of all, for high energy photons the cross-section decreases with increasing photon energy, therefore probability of interaction with electrons decreases. However, if scattering occurred, then photon loses all its energy at the time of Compton scattering on cold electrons (Rybicki & Lightman 1979).

Approximately 10 per cent of hard X-ray photons is reflected, while the remaining 90 per cent is absorbed by cold matter, and energy of these photons is reemitted in soft X-ray band. This is the most probable explanation of soft X-ray excess, which appears as significant low energy ([FORMULA] keV) spectral steepening. Such a feature is widely observed in radio-quiet AGN (Turner & Pounds 1989; Walter & Fink 1993).

Therefore, it is very important to calculate spectra from illuminated disk atmospheres with account of Compton scattering. Many attempts were done on this subject, however, the problem is extremely complex. It is very difficult to compute the structure and spectra of optically thick media self-consistently, taking into account both process of Comptonization and absorption of hard X-ray radiation.

In the existing codes calculating spectrum of the illuminated disk atmosphere, Compton scattering was treated in three different ways. The scattering could be roughly approximated by the local heating and cooling rates, as was done in the photoionization codes CLOUDY, XSTAR and MAPPING. The above approach could be easily implemented also in few radiative transfer calculations (Ko & Kallman 1994; Collin-Souffrin et al. 1996; Sincell & Krolik 1997).

In other approach one solved the Kompaneets equation as part of the radiative transport equation, but it was usually done only for fully ionized hydrogen and helium disk, i.e. neglecting absorption (Shimura et al. 1995). In other papers their authors added some absorption, but only in uniform slab of matter (Ross & Fabian 1993; Matt et al. 1993). Most recently Nayakshin et al. (2000) used modified Kompaneets equation with approximate Compton scattering redistribution function for the computations of disk structure in hydrostatic equilibrium. Nevertheless, Kompaneets equation can approximate only Compton scattering on electrons in non-relativistic thermal motion.

Up to now, the most advanced approach to the Comptonization in illuminated accretion disk was done by Zycki et al. (1994), who presented Monte Carlo simulations. However, their computations treated slab of matter with constant density.

Our approach to Compton scattering is much more general, since it is suitable for photons and electrons of any energies and allows to calculate the structure of an atmosphere both in hydrostatic and radiative equilibrium. I.e. in models presented below we apply the full art of model stellar atmosphere computations. Our code can be easily used for the accretion disk atmospheres, assuming that all the energy has been generated below the thermalization depth via viscosity. We are able to compute self consistently the structure and outgoing spectra, to precisely reproduce reflection and redistribution of external X-rays towards longer wavelengths. We include here hydrogen, helium and iron, all these elements considered in every possible ionization state.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: December 5, 2000
helpdesk.link@springer.de