5.2.2. One source per individual cluster of galaxies
Several studies have used the presence of one central radio source per cluster, to establish statistically over many clusters that the rotation measure is intrinsically high near the centers of clusters of galaxies. Thus Taylor and Barton (1994) have found that high RM is produced by magnetic fields associated with the relatively dense hot x-ray spherical cores in cooling flow clusters of galaxies.
Taylor & Perley (1993) studied one radio source in the Hydra A cluster, and interpreted its RM distribution with the help of a cluster disk of diameter 150 kpc (larger than the radio source) with an uniform magnetic field of 6 µGauss, inclined 48° to the line of sight. There is no optical or X-ray evidence that such a large-scale disk might exist in this cluster.
Carvalho (1994) used the variance in RM towards different clusters of galaxies, and found a cluster magnetic field strength of 0.3 to 1 µGauss. Crusius-Wätzel et al. (1990) found a typical cluster field of 5 µGauss, based on a statistical study of 1 source in each of 5 clusters of galaxies.
On the theoretical side, it is argued but not proven that cluster magnetic fields are maintained by dynamos. Still, assuming that a dynamo operates in clusters, then Poezd & Sokoloff (1993) used the non-linear turbulent dynamo to estimate the intracluster magnetic field. Lacking a large scale body of matter, they found that no cluster-scale regular component of a magnetic field can exist. Thus their "cellular dynamo" predicted a chaotic magnetic field with a strength of a few µGauss in a cluster of galaxies, distributed in a medium of cells. Norman and Meiksin (1996) have proposed that cool magnetic flux loops can reconnect with hot magnetic flux loops, to effect a rapid recycling of mass between the hot and cool phases of an intracluster medium, and to significantly reduce the mass inflow rates in clusters of galaxies. Many details remain to be worked out. A recent short review is given in Böhringer (1995).