Annu. Rev. Astron. Astrophys. 1992. 30: 429-456 Copyright © 1992 by Annual Reviews. All rights reserved

5. MODELS FOR THE RESIDUAL BACKGROUND

Diffuse models for the XRB were among the first to be put forward. Felten & Morrison (1966) suggested that the XRB might be due to inverse Compton scattering of MWB photons by intergalactic electrons. Little has since been done with this idea, except to note that the change in slope in the spectrum at 30-40 keV (Figure 4) is difficult to produce (Cowsik & Kobetich 1972). Hoyle (1963) proposed that the XRB was thermal bremsstrahlung from the hot intergalactic medium (IGM) created by decaying neutrons in the Steady-State Cosmology. This was soon seen to overproduce the XRB (Gould & Burbidge 1963; see also Friedman 1990).

The basic idea of thermal bremsstrahlung from a hot IGM in a Big Bang Cosmology was taken up again by Cowsik & Kobetich (1972) and Field & Perrenod (1977). The temperature required corresponds to kT_IGM ~ 40 (1 + z_heat) keV, where z_heat is the redshift at which the gas is heated. Later work by Guilbert & Fabian (1986) and by Barcons (1987) including relativistic corrections to the bremsstrahlung formulae at these high temperatures showed that the baryonic density, _baryon, needed to be at least 23% of the critical density. This exceeds that allowed by standard cosmological nucleosynthesis (_baryon = 0.06±0.02; Peebles et al 1991). Plausibility problems were also raised about the high energy density needed in hot gas (comparable to the energy density in the MWB). On the other hand, a significant contribution of sources with steep X-ray spectrum at, say more than 30% at 3 keV leaves a background which is too flat to be accounted for by thermal bremsstrahlung (Giacconi & Zamorani 1987, Boldt 1987). This difficulty also applies to discrete source models, since the more contribution you want to get from QSOs discovered in imaging telescopes, the flatter you make the residual background and so the more difficult to explain it with these objects. In any case, the uniform IGM model is now ruled out since it predicts a Compton distortion in the spectrum of the MWB, which is not observed in the spectrum obtained from the COBE satellite (Mather et al 1990). We then have to conclude that the perfect bremsstrahlung shape of the XRB is just a cosmic conspiracy.

Clumping of the IGM can reduce the required value of _baryon, but this is also ruled out since at high redshifts the clumps produce observable fluctuations in the MWB by the Sunyaev-Zel'dovitch effect (Barcons & Fabian 1988, Barcons et al 1991) and at low redshifts the clumps would be detectable as individual X-ray sources (the source cannot then be called the IGM). Besides, the baryon fraction contained in such clumps must correspond to _IGM 0.01, which leads to unacceptably high fluctuations in the temperature of the MWB, unless they are galaxy-sized objects (i.e. not a diffuse medium). In particular, models that have been put forward along these lines (Daly 1987, Subhramanyan & Cowsik 1989) produce too many fluctuations in both the MWB and the XRB.

The IGM may still contribute at some level to the spectrum of the XRB, possibly in the soft X-ray band. The formation of galaxies, groups, and clusters in which some or all of the baryons are heated to the virial temperature can also lead to an observable soft XRB. Conversely, the observed soft XRB can limit _baryon involved in dissipational structure formation (Thomas & Fabian 1990), although soft X-ray absorption in the young objects may seriously affect this.