Interactions with gas-rich barred galaxies

Abstract

In this work we study the effects interactions have on the evolution of gas-rich barred galaxies using N-body/smoothed-particle-hydrodynamics simulations.In the first part we investigate the dynamical effects of an interaction between an {\em initially} barred galaxy and a small spherical companion. In these models the small companion passes through the disc of the larger galaxy nearly perpendicular to it"s plane. The impact position and time are varied with respect to the phase of the bar and the dynamical evolution of the disc. We find that the interactions produce expanding ring structures, offset bars, spokes, and other asymmetries in the stars and gas. We describe how the evolution of the bar strength, pattern speed, and gas inflow rate are affected by the interaction. The results are compared with pure stellar simulations to assess the role played by the dissipative component on the evolution of the disc and bar during the interaction.In the second part, we study the regeneration of stellar bars triggered by a tidal interaction, using numerical simulations of either purely stellar or stellar+gas disc galaxies. We find that interactions which are sufficiently strong to regenerate the bar in the purely stellar models do not lead to a regeneration in the dissipative models, owing to the induced gas inflow. In models in which the bar can be regenerated, we find a tight correlation between the strength and the pattern speed of the induced bar. This relation can be explained by a significant radial redistribution of angular momentum in the disc due to the interaction, similar to the processes and correlations found for isolated barred spirals. We furthermore show that the regenerated bars show the same dynamical properties as their isolated counterparts.In the final part, we present a systematic study of the influence of numerical effects on the evolution of the pattern speed of bars in fully self-consistent simulations. We show that the evolution of the pattern speed is very sensitive to both the intrinsic numerical noise of the model, as well as to the numerical accuracy in the force calculation. Owing to the superposition of these effects the pattern speed shows an uncertainty of roughly 13 per cent. We conclude that large particle numbers and high force accuracy are required for a robust determination of the pattern speed.