Simulations of merging disk galaxies

The conditions under which identical disk-halo galaxies merge and the structure of the resulting merger remnants are studied with N-body simulations. The progenitor galaxies possess truncated exponential disks and ''live'' 1/r 2 halos; the intrinsic stability of these models is assessed. For both galaxy spins aligned with the orbital angular momentum, a large number of encounters are investigated on a variety of orbits to derive an accurate merging cross section. Merging occurs primarily from bound and marginally unbound orbits with a degree of overlap at pericenter. A fairly sharp distinction between ''prompt'' and ''slow'' merging may be established, based on the strength of orbital braking induced by the encounter. Impulse approximation predictions fail for resonant alignment of spin and orbital angular momenta. For a single interpenetrating parabolic encounter, a variety of relative disk orientations at pericenter are studied. The transfer of orbital to internal energy depends sensitively on both the degree of alignment of galaxy spins with the orbital angular momentum, and the extent of ''overlap'' or ''intersection'' of the disks at pericenter. A small fraction of escaping mass carries away a substantial fraction of the total angular momentum. Merger remnants are typically roughly oblate systems, although some must be classed as prolate or triaxial. Many properties of the remnants agree well with those observed for bright ellipticals: projected surface density profiles have the Hubble form, with core radii appreciably smaller than the progenitor peak rotation velocity; a radial separation of disk and halo populations (metallicity gradient) is preserved; rotation curves for the remnants exhibit peak velocities much lower than those of the progenitor; and the remnants have extended, diffuse envelopes