Discussion

In this section, we discuss some further uncertainties and implications regarding the existence of diffuse dust in the universe.

First, the effects of intracluster dust on background sources critically depends on the amount of dust present, and its spatial distribution. Regardless of the mechanism by which grains are injected into the intracluster medium from galaxies, it is possible that a significant fraction are destroyed in the injection process. Significant amounts of hot gas are also believed to exist in the ISM of cluster ellipticals (eg. Forman et al. 1985). This gas is expected to destroy grains on timescales yr (Draine & Salpeter 1979), much shorter than injection timescales. Such destruction mechanisms can thus prevent the formation of significant quantities of dust.

It is possible that the spatial distribution of intracluster dust is not `diffuse' and uniformly distributed, but inhomogeneous. For example, Fabian et al. (1991) propose that if most of the cooled gas resulting from cluster cooling flows remains cold and becomes molecular, then this may provide suitable conditions for large amounts of dust to form. A clumpy dust distribution that follows cooling flow filaments may result, reducing the effective dust covering factor and hence background source deficit. These issues need to be addressed before attributing such deficits to extinction by dust.

The existence of a smooth IGM dust component also remains a major uncertainty. Due to their deep gravitational potential, Margolis & Schramm (1977) showed that it is unlikely for supernovae-driven winds to expel significant quantities of dust from a massive galaxy to large scales. For low mass galaxies however (eg. dwarfs), Babul & Rees (1992) show that this mechanism can be effective. Such systems are believed to comprise a majority of the `faint-blue population' which show an excess times that predicted from non-evolving galaxy models for B>24 (eg. Tyson 1988). Simulations based on star-formation rates that assume yields in metallicity from local observations, predict that the amount of metals (and hence dust, assuming a fixed fraction of metals condense into grains at a constant rate) produced from such a population will be smaller than local estimates by an order of magnitude (eg. White & Frenk 1991). If the only source of IGM dust was from these `low-mass' galaxies, then the total optical depth to would be insignificant, and effects on the background universe would be minimal.

© Copyright Astronomical Society of Australia 1997