Spherical galaxy collisions - Head-on encounters

Abstract

view Abstract Citations (4) References (18) Co-Reads Similar Papers Volume Content Graphics Metrics Export Citation NASA/ADS Spherical Galaxy Collisions: Head-on Encounters Navarro, Julio F. ; Mosconi, Mirta B. Abstract A series of N-body simulations was performed in order to investigate highly interpenetrating (head-on) encounters between equal mass spherical galaxies of similar structure. Two different types of galaxy models were used in the experiments, differing mainly in their velocity distribution and central concentration: in the first kind of model most of the particles moved predominantly along eccentric orbits (RAD and REST models), while in the second type most of the halo kinetic energy resided on the tangential velocity components of the particles and the core was rather isotropic (CIRC model). CIRC model was also much less concentrated than RAD and REST models. RAD model remnants after successive collisions were used as initial galaxy models to simulate tandem galaxy encounters (TDG models). REST model was used as a target galaxy to simulate collisions with a single perturber of comparable half-mass radius R_h_ and total mass. No systematic differences were found between two-galaxy collision simulations and single perturber experiments, except perhaps that these latter appear to give better agreements with the impulse approximation. The experiments covered a wide range of relative velocities, from nearly merging collisions (V_p_ ~ 2.7σ_p_, where V_p_ and σ_p_ stand for the relative velocity and the three-dimensional velocity dispersion of the galaxy at closest approach), to very fast collisions (V_p_ ~ 17.0σ_p_). The galaxies always gain energy in the encounters, the slower the collision the more energetic the galaxies end up after an encounter. The galaxies lose some particles as a consequence of the encounters. Interestingly, the experiments show that the fractional mass loss δ_M_ is always smaller (in absolute value) than the fractional internal energy change δ_U_ (δ_U_ is calculated using the bound final and initial configurations of the galaxy), and that the extra energy is used to expand the galaxy rather than to increase its velocity dispersion. In fact, the velocity dispersion fractional change δsigma_ is always negative. The half-mass radius of the galaxy models always become larger after the encounters. This is related to the exiguous mass loss found throughout the experiments (δ_M_ < 16%) due to the large orbital velocities needed to avoid merging. For the same reason, the mass exchange between the colliding galaxies is negligible for head-on encounters. Unexpectedly, δ_U_ and its dependence on V_p_ are almost independent of the particular galaxy model used. This can be interpreted as the compensation of two competing effects, mass loss and energy gain, due to the particular geometry of head-on encounters. For these encounters, the more radially biased the velocity distribution of the galaxy model, the more effectively the velocity change induced on each particle modifies its kinetic energy. But it is also easier for the least bound particles to escape, and so they are not taken into account to compute the galaxy final energy. It was also found that the velocity distribution of the galaxies approaches isotropy after the collision. The comparison of our results with the impulse approximation shows important disagreements. The main discrepancies are the following: for low relative velocities (V_p_ < 5.0σ_p_) the impulse approximation predicts larger {delta_M_ than those actually found in most simulations. As a consequence, impulsive δ_U_ also overestimate the corresponding results of the simulations. For larger velocities the agreement is good, within the uncertainties. For CIRC and TDG models the discrepancies are much smaller, which suggests that the central concentration of the model is an important factor regarding the accuracy of impulsive estimates. The range of validity of the impulse approximation is difficult to establish given the small number of particles used in our simulations. Our results suggest that impulsive estimates are rather accurate if {SIGMA} >= 20.0, where {SIGMA} stands for the ratio V_p_/σ_p_ weighted by the perturber "concentration" parameter 2R_hpsigma^2^/M_p_. Publication: The Astrophysical Journal Pub Date: January 1989 DOI: 10.1086/167042 Bibcode: 1989ApJ...336..669N Keywords: Astronomical Models; Galaxies; Interacting Galaxies; Many Body Problem; Collisions; Galactic Mass; Galactic Structure; Internal Energy; Kinetic Energy; Velocity Distribution; Astrophysics; GALAXIES: INTERNAL MOTIONS; STARS: STELLAR DYNAMICS full text sources ADS |