Counterrotating stars in disk galaxies are a puzzling dynamical feature whose origin has been ascribed to either satellite accretion events or to disk instabilities triggered by deviations from axisymmetry. We use a cosmological simulation of the formation of a disk galaxy to show that counterrotating stellar disk components may arise naturally in hierarchically-clustering scenarios even in the absence of merging. The simulated disk galaxy consists of two coplanar, overlapping stellar components with opposite spins: an inner counterrotating bar-like structure made up mostly of old stars surrounded by an extended, rotationally-supported disk of younger stars. The opposite-spin components originate from material accreted from two distinct filamentary structures which at turn around, when their net spin is acquired, intersect delineating a "V"-like structure. Each filament torques the other in opposite directions; the filament that first drains into the galaxy forms the inner counterrotating bar, while material accreted from the other filament forms the outer disk. Mergers do not play a substantial role and most stars in the galaxy are formed in situ; only 9% of all stars are contributed by accretion events. The formation scenario we describe here implies a significant age difference between the co- and counterrotating components, which may be used to discriminate between competing scenarios for the origin of counterrotating stars in disk galaxies.

We present a systematic study of stellar feedback processes in simulations of disk galaxy formation. Using a dark matter halo with properties similar to the ones for the Milky Way's stellar halo, we perform a comparison of different methods of distributing energy related to feedback processes to the surrounding gas. A most promising standard model is applied to halos spanning a range of masses in order to compare the results to disk galaxy scaling relations. With few exceptions we find little or no angular momentum deficiency for our galaxies and a good agreement with the angular momentum-size relation. Our galaxies are in good agreement with the baryonic Tully-Fisher relation and the slope of the photometric Tully-Fisher relation is reproduced. We find a zero-point offset of 0.7 to 1 magnitudes, depending on the employed IMF. We also study our standard feedback model in combination with additional physical processes like a UV background, kinetic feedback, a delayed energy deposition as expected for type Ia supernovae, mass return and metal-dependent cooling. Only a combination of effects yields a real improvement of the resulting galaxy by reducing the bulge, while including metal-dependent cooling increases the bulge again. We find that in general the stellar mass fraction of our galaxies is too high. In an ad-hoc experiment we show that an removal of the bulge could reconcile this. However, the fit of the Tully-Fisher relation can only be improved by delaying the star formation, but not suppressing it completely. Our models do not seem to be efficient enough to achieve either effect. We conclude that disk formation is a complex, highly interconnected problem and we expect a solution to come from a combination of small effects.

We conduct a systematic study of the angular momentum problem in numerical simulations of disk galaxy formation. We investigate the role of numerical resolution using a semi-cosmological setup which combines an efficient use of the number of particles in an isolated halo while preserving the hierarchical build-up of the disk through the merging of clumps. We perform the same simulation varying the resolution over 4 orders of magnitude. Independent on the level of resolution, the loss of angular momentum stays the same and can be tied to dynamical friction during the build-up phase. This is confirmed in a cosmological simulation. We also perform simulations including star formation and star formation and supernova feedback. While the former has no influence on the angular momentum problem, the latter reduces the loss to a level potentially in agreement with observations. This is achieved through a suppression of early star formation and therefore the formation of a large, slowly rotating bulge. We conclude that feedback is a critical component to achieve realistic disk galaxies in cosmological simulations. Numerical resolution is important, but by itself not capable of solving the angular momentum problem.