Particle hydrodynamics (SPH) simulations are presented of direct collisions between two model galaxies, most consisting of a rigid halo and a gas disk. Local self-gravity is also computed in the gas. The companion galaxy in these simulations is about one-third of the mass of the primary, and its disk is half the size. An adiabatic equation of state is combined with simple approximations for the effects of radiative cooling and local heating due to young star activity, which allows a continuous range of thermal phases to develop. These terms and multiple phases have not generally been included in galaxy collision simulations to date. Their effects are assessed in part by repeating runs with an isothermal equation of state and comparing the results. One model with a star plus gas disk is also included for comparison. These models are most relevant to interactions involving low surface brightness, or other late-type galaxies with extensive gas disks, including the precursors to well-known ring galaxies like the Cartwheel and VII Zw 466. In the simulations, the companion impact is slightly off center in the target disk, as is probably the case in these systems. In all cases, clear ring waves develop in the primary despite the disruption of parts of the disk by impact shocks. The gas density in the disk of the primary is initialized to values slightly below the gravitational instability threshold throughout, and the ring waves induce star formation in all the heating and cooling models. The structure of the waves and other interaction morphologies are found to be quite similar on large scales in both isothermal and heating/cooling cases, despite the fact that at certain stages large quantities of gas are heated above the initial temperature in the latter. On a finer scale, there are clear differences, including the fact that star formation heating in ring waves increases the vertical scale height of the primary gas disk and delays spoke development. The companion disk is largely disrupted in most of these simulations, and a substantial mass of gas is splashed out into a bridge connecting the two potential centers. The companion disk reforms by accreting gas out of the bridge, though generally in a different plane than its initial one. There is also a good deal of infall back onto the primary disk. Although heated by impact, the gas in the bridge cools rapidly. However, kinematic expansion prevents it from reaching threshold density, and there is no star formation heating there. A comparison run with a diskless companion produced no significant bridge, so in this type of collision the bridge is primarily a hydrodynamic phenomenon. The amount of material pushed out into the splash bridge and how much of it comes from each galaxy depends on the relative orientation of the disks at impact. This orientation also affects how much bridge material accretes onto each galaxy. The onset of accretion is initially delayed but then accelerates to a peak and declines thereafter in both galaxies. The infall is spatially asymmetric and is primarily located in well-defined streams. Most of the accreted gas ends up in the central regions of the model galaxies, but only after spiraling around the center and passing through one or more shocks. Accretion heating is substantial, and is shown to inhibit or delay global star formation enhancements. The thermal effects of the impact between galaxies are short-lived, but the models predict that accretion and young star heating effect the global thermal phase balance for a much longer period. The magnitude and duration of these effects also depend on the relative orientation of the disks at impact. Thus, the postcollision Hubble type of the companion is a sensitive function of initial orientation.
Read full abstract