X–ray Emission in Gamma-ray Blazars

As shown by Maraschi (this volume) the overall spectral energy distribution (SED) of the blazars detected by EGRET shows two peaks, in a ν–νF (ν) plot. With few exceptions, the minimum between these two peaks occurs in the X–ray range. We can exploit this observational fact to constrain the models proposed to account for the SED of blazars, and to reach an important conclusion: the conversion of the primary power carried by the jet into radiation occurs primarily at some distance from the central powerhouse. Assume in fact that part of the γ–ray radiation is absorbed by γ–γ collisions (Blandford & Levinson 1995). This implies that, in the comoving frame of the jet/blob emitting the high energy radiation, there is a sufficient amount of target X–rays. The pairs created in this way are relativistic, and can emit at lower frequencies, or escape, if their cooling time is sufficiently long. In the first case, all the absorbed power in the γ–ray band reappears at lower energies, namely the X–ray band. Therefore it is inevitable to predict that the X–ray luminosity should be of the same order of the γ–ray luminosity. A way out of this is to assume that the cooling time of the pairs is longer than their escape time. In this case only a fraction of the power absorbed in the γ–ray band is reprocessed into radiation of softer energy, mainly X–rays. This model then requires that there are sufficient X–rays to absorb the γ–rays, but not enough photons for Compton cooling. In principle, this is possible, because the scattering between pairs and X–rays occurs in the inefficient Klein Nishina regime, but it is highly unlikely, because the X–ray emission should always be accompanied by at least a comparable amount of optical UV radiation. Requiring efficient absorption of γ–rays, therefore, implies short cooling times of the pairs, resulting in an excess of X–ray emission (Ghisellini & Madau 1996). Since this is not observed, we conclude that: • The γ–ray emitting region is transparent. • In order to be transparent, it must be at some distance from the (X–ray emitting) accretion disk. On the other hand, the short variability timescales observed at high energies limit the dimensions of the source, and hence its location in the jet. By combining both limits, we can derive a typical distance at which dissipation occurs, of few hundreds Schwarzschild radii. • In the inner part of the jet, energy must be transported efficiently, without dissipation. Possibilities are: i) cold (in the comoving frame) protons with bulk Lorentz factor Γ; ii) Poynting flux. The kinetic luminosity carried by the protons is