A two-component ionized reflection model of MCG-6-30-15

Ionized reflection has often been considered as the explanation for the unusual Fe Kα variability observed in MCG-6-30-15. In this paper, we test this model using a 325-ks observation of MCG-6-30-15 by XMM-Newton and BeppoSAX. The data are fitted between 2.5 and 80 keV with the constant-density models of Ross & Fabian. The best-fitting ionized reflection model requires the Fe Kα line to be split into two reprocessing events: one from the inner disc to build up the red wing, and the other from the outer accretion disc to fit the blue horn. The implied geometry is a disc that becomes strongly warped or flared at large radii. A good fit was obtained with a solar abundance of iron and a reflection fraction (R) of unity for the inner reflector. The combination of the two reflection spectra can appear to have R > 2 as required by the BeppoSAX data. The inner reflector has an ionization parameter of log ξ= 3.8, but the outer one is neutral with an inner radius of ~70 gravitational radii (r_g), corresponding to a light crossing time of approximately an hour for a 10⁷ M_solar black hole. Applying this model to time-resolved spectra shows that the inner reflector becomes more ionized as the source brightens. This reduces the strength of the red wing at high flux states. The X-ray source is constrained to arise from a narrow annulus at ~5r_g, with only 6 per cent of the 2-10 keV flux being due to the outer reprocessor. This amount of localized energy generation is extremely difficult to produce without resorting to other energy sources such as the black hole spin. In fact, all the Fe Kα models fitted to XMM-Newton spectra of MCG-6-30-15 require a large increase in energy production at the inner edge of the accretion disc.