The vast majority of cooling flow clusters contain powerful radio sources associated with central cD galaxies. Initial evidence of radio sources displacing, and evacuating cavities in, the X-ray-emitting intracluster medium (ICM) was found with ROSAT observations of a few sources including Perseus (Böhringer et al. 1993), Abell 4059 (Huang & Sarazin 1998), and Abell 2052 (Rizza et al. 2000). Models predicted that the radio sources would shock the ICM, and that the X-ray emission surrounding the lobes would appear hot (Heinz, Reynolds, & Begelman 1998). High-resolution images from Chandra have revealed many more cases of radio sources profoundly effecting the ICM by displacing it and creating X-ray deficient "holes" or "bubbles." The Chandra data allow us to study the physics of the interaction in much more detail (i.e. Hydra A, McNamara et al. 2000; Perseus, Fabian et al. 2000; Abell 2052, Blanton et al. 2001; Abell 2597, McNamara et al. 2001; Abell 496, Dupke & White, 2001; MKW 3s, Mazzotta et al. 2002; RBS797, Schindler et al. 2001; Abell 2199, Johnstone et al. 2002; Abell 4059, Heinz et al. 2002; Virgo, Young et al. 2002; Centaurus, Sanders & Fabian 2002; Cygnus A, Smith et al. 2002; Abell 478, Sun et al. 2003).
A long-standing problem with cooling flow models has been that the mass of gas measured to be cooling from X-ray temperatures, based on surface-brightness and spectral studies with Einstein and ROSAT has not been detected in sufficient quantities at cooler temperatures. High-resolution spectroscopy with XMM-Newton provided direct evidence that gas was cooling in these clusters, but very large masses of gas (hundreds of solar masses) were seemingly cooling over only a limited range of temperatures. Emission lines such as Fe XVII expected from gas cooling below approximately 2 keV were not detected and in general, most of the ICM seems to cool to about one-half to one-third of the cluster ambient temperature (Peterson et al. 2003). Several possible solutions have been proposed for the lack of cool X-ray gas seen in the new observations (Fabian et al. 2001; Peterson et al. 2001). These include mixing, heating by central active-galactic nuclei (AGN), inhomogeneous abundances, and differential absorption. Heating of the gas by a central radio source has also been discussed recently by Böhringer et al. (2002), Churazov et al. (2002), Ruszkowski & Begelman (2002), Kaiser & Binney (2003), and Brüggen (2003), among others.
A common finding with the Chandra data is X-ray deficient holes that correspond with radio emission from the lobes of a central AGN. These holes are typically surrounded by bright shells of dense, X-ray-emitting gas. One of the surprises of the Chandra observations is that the X-ray-bright rims surrounding the radio sources observed in cooling flow clusters were found to be cooler, rather than hotter, than the neighboring cluster gas (i.e. Perseus, Schmidt et al. 2002; Hydra A, Nulsen et al. 2002; Abell 2052, Blanton et al. 2003). The bright shells show no evidence of current strong shocks. Soker, Blanton, & Sarazin (2002) found that for the case of Abell 2052, the morphology was well-explained by weak shocks occurring in the past, and strong shocks were generally ruled out. To date, no case of current strong-shock heating of the ICM in a cooling flow cluster by an AGN has been observed. Although strong-shock heating from the radio source as proposed by Heinz, Reynolds, & Begelman (1998) and Rizza et al. (2000) may not be the explanation for the fate of the missing cool gas, energy input from a radio source in the form of weak shocks (e.g. Reynolds, Heinz, & Begelman 2002) and buoyantly rising bubbles of relativistic plasma (e.g. Churazov et al. 2002) can still contribute to heating.