Astron. Astrophys. 329, 809-820 (1998)

5. Magnetic fields at z 2

Regardless of the specifics of a Faraday screen model, the intrinsic RMs of several thousand rad m^-2 require magnetic fields of the order of a G correlated on scales of many kpc (see Eqn. 2) in the extended environments of galaxies at z 2. The universe was just a sixth of its present age at z = 2.25 and only a tenth at z = 3.8. It is surprising that fields at such early times are similar to those found in galaxies and clusters today. These observations pose a considerable challenge to the models of magnetic field generation because of the small time available for amplifications at those redshifts.

The currently popular theory is that magnetic fields have arisen from amplification of a very weak seed field (10^-21 -10^-19 gauss) by a dynamo mechanism (e.g. Zeldovich et al. 1983). Dynamos typically lead to an e-fold amplification over the dynamical time scale of the velocity field responsible for the dynamo action. The dynamo due to the turbulent motion of galaxies in the ICM would have dynamical timescales of the order of 10⁸ yr or higher (galaxy sized 25 kpc eddies with speeds 1000 km s^-1). This would only lead to an amplification of 10⁴ -10⁵, within the available 10⁹ yr at z = 2.5, of the initial seed field of 10^-19 gauss. This suggests that the Faraday screens cannot be more than 5 kpc (which reduces the dynamical timescale, providing sufficient time for the required number of cycles of amplification). However, this would necessitate larger magnetic fields to generate the same RM.

Lesch & Beck (1995) have suggested that the seed field amplification would proceed much faster in collapsed disk galaxies than in the ICM (see also Beck et al. 1994). The initial collapse would jump-start the process by amplifying the seed field by factors of about 10⁴. Their calculations suggest that a dynamo in the rotating disk would further amplify the field by a factor of 10¹⁰ or higher by z 2 if z 5. This correlated G field is confined to the inner few kpc of the evolving disk galaxy. However, the amplification is a strong function of and would be inadequate if is much higher than 0.1. If this is indeed correct, it raises the possibility that the high RM radio sources are in the midst of (rich) clusters which already contain many collapsed systems and as such may be good candidates for targetted searches of high redshift clusters.

Another possibility is that the ambient thermal plasma is magnetised by the active radio source itself. The synchrotron plasma in the radio lobes have strong magnetic fields of hundreds of G or even a magnitude higher in the active hotspot, particularly in powerful radio sources. The magnetic field is believed to be transported by the radio jet from the active nucleus. MHD instabilities (Bicknell et al. 1990) and turbulent diffusion may lead to the transport of the magnetic fields into the surrounding thermal plasma. However observations indicate that there is very little mixing of the thermal and the synchrotron plasma especially near the hotspot where the two are well separated by the bow-shock (Carilli et al. 1994b; Bohringer et al. 1993). It is believed that the radio jet evacuates the cluster medium along its path with very little mixing of the two media. A similar effect is seen in the galaxy 0902+343 (z = 3.4) with its line emitting gas anticorrelating with the radio structure (Carilli 1995). Even theoretically, it is not clear if the hotspot field can be diffused into the region around it on a timescale fast enough to produce a deep Faraday screen surrounding the hotspot before it moves ahead substantially (ie. within 10⁶ yr).

Tribble (1993) has suggested that synchrotron plasma from an extinct radio source may be responsible for magnetising the ICM. The absence of signs of a relic radio source in the vicinity of these HRRGs and the large times required for diffusion of the magnetic field to large distances, pushes the time of formation of the presumed earlier generation sources to even higher redshifts. The present understanding of what triggers a radio source is rather sketchy and whether the very short time available is sufficient for more than one radio source generation is not clear.

It has also been suggested that the galactic magnetic fields of today have arisen from a primordial magnetic field with a present day strength of 10^-10 gauss and correlated over megaparsecs (Kulsrud & Anderson 1992). The field strength would go as (1/S)², where S is the cosmological scale factor. Such a primordial field would be of the order of 10^-8 gauss at z_m = 9 (the redshift of turn around in section 4.2.1- corresponding to z = 5) and correlated over hundreds of kpc. Galactic and sub-galactic clumps condensing out of such material would have G fields correlated over several kpc. However the nature of the processes which can generate such fields in the early universe is not clear.

© European Southern Observatory (ESO) 1998

Online publication: December 16, 1997
helpdesk.link@springer.de