The kinematics of 3:1 merger remnants and the formation of low-luminosity elliptical galaxies

We test the formation of low-luminosity elliptical galaxies through collisionless mergers of unequal-mass disk galaxies. The kinematic properties of a small survey of simulated merger end products with initial disk mass ratios of 3:1 are compared to a sample of seven low-luminosity galaxies observed by H.-W. Rix and coworkers that were chosen photometrically to be "ellipticals." In this paper, we go beyond a comparison in terms of global properties (using one number to characterize a model or a galaxy, e.g., <a(4)>, ellipticity at some fixed radius, or central velocity dispersion) and examine the detailed kinematics as a function of galactocentric distance. The merger remnants are "observed" through a slit along the major and minor axes, using a pixel binning and slit width similar to the ones used during the spectroscopic observations. Inside each bin, we determine the line-of-sight velocity distributions and parameterize them with Gauss-Hermite functions. We compare the rotational support of the merger remnants, i.e., the ratio of the mean line-of-sight velocity upsilon to the velocity dispersion sigma along the major axis, the normalized rotation on the minor axis, and the major-axis Gauss-Hermite moments h(3), to those of the observed galaxies. The N-body remnants are very flattened when viewed edge-on (< epsilon > similar to0.6) and should be inclined before making a fair comparison with the data set of H.-W. Rix and coworkers (which has < epsilon > similar to 0.3). When the merger remnants are appropriately inclined, their upsilon/sigma profiles rise slower than the observed ones: upsilon/sigma (merger) is in the range [0.1,0.8] at 1R(eff) and [0.2, 1.6] at 2R(eff), whereas upsilon/sigma (observed) spans the intervals [0.8,2.0] at 1R(eff) and [1.4,3.5] at 2R(eff). Note that even when the remnants are viewed edge-on, the via profiles do not match the observations. The detailed comparison of the observations with our set of purely collisionless 3:1 merger remnants shows that these objects and low-luminosity ellipticals do not have similar kinematic profiles. This suggests that this kind of dissipationless merger (or mergers with more even masses, e.g., mass ratios of 2:1 or 1:1) is not likely to be the dominant formation channel for low-luminosity elliptical galaxies.