The ICM has a metallicity 3 in solar units. That means the ICM
cannot be purely of primordial origin, but it must have been processed at least partially in the cluster galaxies, and later it must have been expelled from the galaxy potential into the cluster potential and hence into the ICM. As mentioned already in § 4.2, the star formation activity, i.e. the metal production rate, and its connection to mergers, is still controversial. The gas ejection processes and their time scales are also still under discussion. Several gas ejection processes have been suggested in the seventies: ram-pressure stripping (Gunn & Gott 1972), supernova-driven galactic winds (De Young 1978), galaxy-galaxy interaction and jets emerging from active galaxies.
In order to decide which process is dominating at what time, observations and simulations were performed and compared, in particular with respect to the metallicity distribution within clusters and to the metal-licity evolution with redshift. So far the best measured metal abundance distributions based on a large sample of clusters is the one derived using BeppoSAX data (De Grandi & Molendi 2001; Irwin & Bregman 2001). The metallicity is found to be relatively homogeneous, except for cooling flow clusters, where a decrease of metallicity with radius is observed. New XMM measurements confirm this result (Arnaud 2001; Mushotzky 2001). The new measurements also show that the ratio of elements is not solar everywhere in the clusters, but an underabundance on Oxygen is found e.g. in the centre of the cluster Sersic 159/03 (Kaastra et al. 2001). XMM and CHANDRA observations will provide measurements of the evolution of the metallicity with redshift out to redshifts of unity and beyond, which were hardly possible with previous instruments (Schindler 1999).
As many different scales are involved in the metal enrichment process, the existing numerical models must span a large range of scales. Simulations on cosmological scales taking into account the effects of galactic winds were performed by Cen & Ostriker (1999). They found that the average metallicity increases from 0.01 solar at z=3 to 0.2 solar at present. The metallicity distribution is not constant but denser regions have generally higher metal abundances. Gnedin (1998) took into account not only galactic winds but also galaxy-galaxy interactions and concluded that most metals are ejected by galaxy mergers. In contrast to this result Aguirre et al. (2001) found that galaxy-galaxy interactions and ram-pressure stripping are of minor importance while galactic winds dominate the metal enrichment of the ICM. A problem with this kind of simulations is that they must cover large scales as well as galaxy scales. Therefore the resolution is not very good at small scales and hence the results have large uncertainties, which is probably the reason for the discordant results.
Also the effects of supernova-driven winds have been investigated with models on cluster scales. David et al. (1991) calculated the first models and found that the results depend sensitively on the input parameters: the stellar initial mass function, the adopted supernova rate and the primordial fraction of intra-cluster gas. In the first 3D models which took into account the full gas dynamics and the effects of galactic winds on cluster scales, Metzler & Evrard (1994, 1997) showed that winds can account for the observed metal abundances. Very strong metallicity gradients were found (almost a factor of ten between cluster centre and virial radius) which are not in agreement with the observations. The authors found that these metallicity gradients are hardly affected by cluster mergers. From simulations on galaxy scale Murakami & Babul (1999) concluded that galactic winds are not very efficient for the metal enrichment process.
Another process which is probably important for metal enrichment is ram-pressure stripping. As a galaxy approaches the cluster centre it experiences an increasing pressure and at some point the galaxy potential is not strong enough to retain the galaxy gas. The gas is stripped off starting from the outer regions of the galaxies and the metals are released into the intra-cluster medium. Two spectacular examples where the stripping process can be observed are two galaxies in the Virgo cluster, NGC4501 and NGC4548 (Cayatte et al. 1990).
Simulations of ram-pressure stripping are relatively difficult to carry out because not only the conditions of the gas inside the galaxy and the potential of the galaxy must be taken into account, but also the conditions of the surrounding medium. In early models the effect was calculated with relatively simple means (Takeda et al. 1984; Gaetz et al. 1987; Portnoy et al. 1993; Balsara et al. 1994).
Recently, high resolution simulations were carried out to study the stripping process in different types of galaxies. Abadi et al. (1999) and Quilis et al. (2000) performed simulations of spiral galaxies. They found that the interstellar medium can be removed if it is not homogeneous. For dwarf galaxies Mori & Burkert (2000) found in their simulations that the gas can be easily stripped off when these galaxies move through the intra-cluster medium. Simulations of elliptical galaxies (Fig. 8.7; Toniazzo & Schindler 2001) showed that the gas cannot only be stripped off as the galaxy approaches the cluster centre, but the galaxy can again accumulate some gas when it is in the apocentre of its orbit. Also the X-ray morphologies of simulated and observed galaxies can be compared (see Fig. 8.8).
All these simulations show that ram-pressure stripping can be an important metal enrichment process for the ICM. Merging activity increases the effect even more because the ram pressure is proportional to the square of the relative velocity of intra-cluster gas and galaxies. During mergers, not only is the gas density increased but also the rel-
Was this article helpful?