Next Contents Previous

6.1.4. The X-ray Background

An isotropic distribution of X-rays was originally discovered in 1962 by Ricarrdo Giacconi and colleagues. As this predates the discovery of the CMB, this discovery of an X-ray background (XRB) was initially somewhat puzzling. Over the spectral range 3-40 Kev, the XRB spectrum is very well described by a thermal bremsstrahlung model with a characteristic temperature of 40 keV. This led to a series of models in the 80's in which the spectrum was produced by a rather hot, diffuse IGM (see Subrahmanyan and Cowski 1989; Anderson and Margon 1987; Giacconi 1987). The heating sources of this hot (approx 108) gas, however, are rather unclear. Care must be taken not to violate any Big Bang constraints and several investigations (see Taylor and Wright 1989) have shown that a hot IGM requires that the gas be re-heated in the redshift range z = 3-5 which means the rest-frame temperature of the IGM is of order 150-250 keV. To obtain the observed flux in the XRB also requires a baryon density of Omegab geq 0.2. Thus, the confirmation of the XRB as being due to a hot diffuse IGM has major significance on the question of how the baryons are distributed. This confirmation would also seriously call in the question the limits on Omegab obtained from primordial nucleosynthesis arguments.

A clear prediction of the existence of a hot IGM would be the production of Compton distortions in the CMB, of the kind described in Chapter 2 regarding the Sunyaev-Zeldovich effect. These distortions would have been detected in just the first few minutes of observation with the COBE satellite and they were one of the first things that could be ruled out (see Mather et al. 1990). With more COBE data, it is now possible to constrain the contribution of a smoothly distributed hot, diffuse IGM to less than 1 part in 104 to the observed XRB (see Wright et al. 1994). While a cooler more strongly clumped IGM can circumvent some of this constraint, in order to be consistent with the observed COBE anisotropy requires a very large number of clumps that are essentially galaxy sized but nevertheless are sources of X-ray emission. While the gas that is bound in galaxy halos by dark matter can be a source of X-ray emission, its characteristic temperature, which reflects the depth of the potential, is much too low to account for the observed XRB. These considerations make it plain that the only feasible contribution to the XRB comes from discrete sources.

Galaxies, galaxy clusters and QSOs/AGNS are all potential discrete source contributors to the XRB. Sorting out the potential contribution of all these sources became a serious endeavor in the mid 80's and early 90's. Now with the ROSAT all-sky survey in soft X-rays, as well the GINGA,EXOSAT and ASCA missions there is a great deal of observational data which can be used to constrain the contributions from each of these sources. Of these sources, the overall spectral shape of the XRB now strongly favors QSOs/AGNs as being the dominant contributor (see Setti and Comastri 1996) and there is essentially no room for additional classes of objects and certainly no room for a hot, diffuse IGM. The key feature of the model which allows for this conclusion is the incorporation of absorbed sources as a function of neutral hydrogen column density. Briefly, it is now realized that most AGN are surrounded by a torus of absorbing material (e.g., gas and dust) and that the source-torus-observer geometry determines the observational attributes of a particular AGN. Thus, some fraction of X-ray AGN have their emission strongly self-absorbed and/or scattered by this torus. Incorporating this feature into the model produces a very good fit to the XRB data, particularly that determined by ROSAT (see Hasinger et al. 1993).

Next Contents Previous