3.3 XD Clusters and Mass Deposition

As described above, clusters with massive dominant, central galaxies (XD clusters) are common and are easily seen in X-rays because of their high central surface brightness. Considerable evidence has accumulated that the hot X-ray emitting gas in the cores of these clusters is cooling around these central galaxies (see "Cooling Flows in Clusters and Galaxies" ed. A. Fabian for an extensive collection of reviews and papers on all aspects of cooling gas in galaxies and clusters). The evidence for cooling gas is threefold:

• observed X-ray surface brightness profiles imply high central gas densities and therefore short cooling times,

• X-ray spectroscopic observations show the presence of gas in the cores of these clusters with temperatures up to 10 times lower than the cluster mean,

• optical observations show the presence of emission line filaments in clusters where there is evidence for cooling gas.

The surface brightness profiles of clusters have been analyzed in two ways. First, comparison of cluster models to the profiles (Jones and Forman 1984) show that some clusters exhibit an excess of X-ray emission in their cores over a simple function which adequately describes the profiles at large radii. The excesses originate within a radius where the gas cooling time is comparable to a Hubble time. These excesses are not found in symmetrical clusters where the central gas density is low (and hence the cooling time of the gas is long). Also, the excesses occur in clusters which have a central dominant galaxy. A second method of analysis uses a deprojection method to determine the gas temperature and density profile, by assuming a gravitational potential (Fabian et al. 1981; Stewart et al. 1984; Arnaud 1988).

The most complete study of mass deposition in clusters is that of Arnaud (1988) who analyzed 103 single-peaked clusters from the Einstein survey. He found that 40% of these single-peaked clusters have central cooling times less than 2 x 10¹⁰ years and that of these cooling clusters, 20% have inferred mass deposition rates exceeding 100 M yr^-1. The mass deposition rates derived from this method correlate well with the excesses measured from the first technique (Stewart et al. 1984). These surveys show that this phenomenon is widespread and can potentially deposit enormous amounts of material.

The strongest evidence for cooling gas, although from a smaller number of clusters, comes from the X-ray spectroscopic observations. Different instruments have been used to sample emission from different temperature regimes. Detailed analyses yield mass deposition rates which are in agreement over a wide range of temperatures. The best studied example is Perseus (NGC1275) where emission from gas at T ~ 5 x 10⁶K implies mass deposition at a rate of 200 M yr^-1. The analysis of emission from gas at ten times higher temperatures yields comparable mass deposition rates (120 M yr^-1; Mushotzky and Szymkowiak 1988 and Canizares, Markert, and Donahue 1988). Both estimates agree with that derived by Arnaud (1988) from the image deprojection technique.

In addition to the evidence from the X-ray observations for cooling gas, extensive optical emission line observations have been made which show gaseous filaments in and near the central dominant galaxy at temperatures around 10,000K (e.g., Cowie et al. 1983; Hu, Cowie, and Wang 1985; Heckman et al. 1989). In general, extensive emission lines are found in clusters with large inferred mass deposition rates (see e.g., Heckman et al. 1989).

Objections to the existence and magnitude of the rates of mass deposition around central galaxies in clusters arise from three major considerations:

• the inferred mass deposition rates are large and can exceed 100 M yr^-1 (PKS0745, A1795, A2597, Hydra A),

• the final repository of the cooling mass, assumed to be low mass stars, is unobserved,

• the X-ray and optical emission-line estimates of the amount of cooling gas are inconsistent since the X-ray observations require the matter to be deposited over a wide range of radii (roughly r) while the emission lines are concentrated to the cluster center and the emission lines require 100-1000 recombinations per cooling atom.

  Various heating mechanisms, such as thermal conduction and supernova explosions, could reduce the mass deposition rate. However, detailed calculations have not been successful in modeling the observed temperature distributions and reducing the mass deposition rate. For example, Bregman and David (1988) showed that the mass deposition rate could be reduced by factors of about 3 if the conduction efficiency factor is fine-tuned from cluster to cluster which they found to be implausible. Supernovae also could reduce the mass deposition rate but would require 3-5 supernovae yr^-1, which are not observed, to produce reductions by a factor of 10.

  The evidence for large mass deposition rates is considerable. As Fabian (1987) has emphasized, either extensive mass deposition occurs around central cluster galaxies and produces low mass stars (considerably less than 1 M) or there is a heating mechanism operating in these clusters which is poorly understood and can provide 10⁶¹ - 10⁶² ergs over the lifetimes of these systems. While gaps remain in our understanding of large mass deposition rates in clusters, the alternatives remain poorly developed and unattractive.