Annu. Rev. Astron. Astrophys. 1996. 34: 749-792
Copyright © 1996 by Annual Reviews Inc. All rights reserved

Next Contents Previous


Recent theoretical models have been quite successful in explaining several important features that appear to be common to most gas-rich mergers. For example, theorists reacted rapidly to the observational discovery of high concentrations of cold gas in the central regions of some mergers and developed new dynamic models with gas+stars that have provided much needed insight into how such gas concentrations are produced (e.g. Barnes & Hernquist 1992, Barnes 1995). During collisions, the gas readily losses angular momentum due to dynamical friction, decouples from the stars, and inflows rapidly toward the merger nuclei. The typical half-mass radius of the gas is only 2.5% that of the stars, which scaled to Milky Way units implies ~ 5 x 109 Msmsun of neutral gas within a radius of ~ 140 pc and a gas density of ~ 103 Msmsun pc-3 (Barnes 1995).

Although all gas-rich disk-disk mergers appear to end up with enormous gas densities at their centers, independent of the specific orbital parameters, the rate of gas inflow may depend on the structure of the progenitor galaxies, specifically, on the size of a dense central bulge (Mihos & Hernquist 1994). Galaxies without bulges develop bars that produce a more steady inflow that may last ~ 1.5 x 108 years. By contrast, in objects with large bulges the disk tends to be stabilized against strong inflow until just before the galaxies finally coalesce, when strong dissipation finally drives the contained gas to the center in ltapprox 5 x 107 years. If the level of infrared activity depends on the rate of gas infall into what is to be the maximum central gas density configuration, or if some other mechanism such as a binary black hole is required to create an extreme nuclear starburst (e.g. Taniguchi & Wada 1996), then it may be the case that not all mergers of gas-rich spirals will produce ULIGs.

Noninteracting spirals typically have large quantities of H I beyond their optical disks. For instance, in the Milky Way most of the H I mass is beyond the solar circle, whereas most of the molecular gas is at galactocentric radii < 0.7 Rsmsun. Computer models predict that a large fraction of the H I gas in the outer regions of the pre-encounter disks will be pulled out to large radii in the form of tidal tails, out of which dwarf irregular galaxies can be formed (Barnes & Hernquist 1992, Elmegreen 1993). Elmegreen et al. (1993) have proposed that the heating of the interstellar medium due to the encounter increases the Jeans' mass and that the H I gas that leads the matter launched into intergalactic space will form self-gravitationally bound cloud complexes, which may collapse and appear as detached irregular dwarf galaxies. This model forms large H I clouds with masses as large as 109 Msmsun at the end of the tidal tails opposite the companion (Mirabel et al. 1992, Duc & Mirabel 1994). In the Barnes & Hernquist (1992) model, however, the clumps form from collapse of the stellar population, with mass fluctuations on all scales, producing also ``failed dwarfs'' of only old stars due to the fact that the potential wells are too shallow to capture enough H I to form new stars. This result is consistent with the wide range of B - V colors of the condensations along the tidal tails of some ULIGs (Mirabel et al. 1991).

Next Contents Previous