The strength of central X-ray emission should depend on the accretion process and black-hole mass, coupled with the effects of geometry and absorption. X-ray imaging has insufficient angular resolution to separate such emission from beamed radiation associated with an inner radio jet. However, in sources where an absorbed power-law component is present, it is easily recognized if it is dominant, and a heavy excess absorption helps to make spectral separation possible even if much of the emission is from surrounding hot gas. The best case is the hard X-ray detection of emission from the core of Cygnus A (Section 2.1), and absorbed power-law components are claimed in other radio galaxies with ROSAT, BeppoSAX and ASCA (e.g. Allen & Fabian 1992, Trussoni et al. 1998, Sambruna et al. 1999).
Where the excess absorption is only modest, and this covers many of the cases where the absorbed X-ray component is dominant, an association with the central engine is questionable, and the absorbed X-ray emission is most likely beamed emission associated with the radio jet (Section 3.2). This is illustrated by NGC 6251, where the X-ray absorption of ~ 1021 cm-2 agrees both with that inferred from reddening through the large-scale disk measured with HST (Ferrarese & Ford 1999) and with an HI radio absorption-line measurement (Worrall & Birkinshaw 1999b), and where the strength of X-ray relative to radio emission in comparison with other radio galaxies argues independently for a radio-related origin for the power-law X-ray emission (Worrall & Birkinshaw 1994). The `puzzling' excess absorption seen in BLRGs (Sambruna et al. 1999) might also be explained, at least in part, by cool gas on larger scales than an inner torus absorbing jet-related X-rays. This is consistent with the relatively strong X-ray emission of BLRGs and the required orientation of their jets under Unification models.
ROSAT pointed observations have shown that the central soft X-radiation of low-power radio galaxies is almost certainly dominated by nonthermal emission associated with the radio jet. Canosa et al. (1999) find that the core X-ray and radio emission are well correlated in the B2 radio-galaxy sample (Fig 5), and a similar situation holds for the low-power 3CRR radio galaxies (Hardcastle & Worrall 1999a). M 87 and Cen A are sufficiently close that jet-related X-rays are resolved, and the fact that their X-ray to radio ratio is similar to that for more distant unresolved X-ray cores is further support for a jet-related origin of the core soft X-ray emission in all such sources (Worrall 1997). Although in principle such X-ray emission could be either synchrotron or inverse Compton in origin, the relative proportions of radio, optical (HST) and X-ray core emission, as compared with radio-selected BL Lac objects, argue in favor of inverse Compton emission and predict a relatively flat spectral index (Hardcastle & Worrall 1999b). Flat-spectrum components superimposed on thermal X-rays from hot gas are reported in the ASCA spectra of several low-power radio galaxies, but are variously interpreted as thermal emission associated with an advection dominated accretion flow (Allen et al. 1999) and as higher than previously suggested (e.g. (Fabbiano et al. 1989) X-ray emission from stellar and post-stellar X-ray sources (Matsumoto et al. 1997). Jet-related X-ray emission is also likely to be a major contributor to the compact soft X-ray emission of powerful radio galaxies (Hardcastle & Worrall 1999a) see Fig 3 and, although in general their greater distance with respect to low-power sources leads to the expectation of a larger contribution from extended gaseous emission within an unresolved X-ray core.
 |
Figure 5. Core X-ray vs. core radio luminosity for the sample of B2 low-power radio galaxies observed with ROSAT in pointed observations, after separation of any extended X-ray emission. The best-fit correlation taking into account non-detections (solid line) excludes (based on astrophysical arguments) a starburst galaxy (cross) and the broad-line radio galaxy 3C 382 (diamond) which fall in the sample. Radio galaxies in previously known optical clusters are shown as squashed squares, relic radio sources are open circles, and other sample members are shown as stars. Figure from Canosa et al. (1999). |
3.3. Non-thermal emission components away from the core
X-ray emission from compact radio hotspots has been detected in a handful of sources, as summarized by Hardcastle et al. (1998a) and Harris (1998). Such measurements are potentially of great physical interest since the radio-emitting regions are normally well localized and, if it can be shown that the X-rays are of inverse Compton origin (e.g. Harris et al. 1994a), the radio and X-ray emission together probe the magnetic field strength and the balance between particle and magnetic energy density. A similar probe on larger scales is provided through inverse Compton scattering of cosmic microwave background (CMB) photons on particles in the radio lobes, with detections reported in a few sources (Feigelson et al. 1995, Tsakiris et al. 1996, Tashiro et al. 1998). Brunetti et al. (1999), in discussing extended emission in 3C 219, have emphasized the role of AGN photons in Compton up-scattering, but at present sources with extended X-ray emission of gaseous origin vastly outnumber those for which extended inverse Compton X-ray emission is likely to have been detected.