The modelling of feedback processes in cosmological simulations of disc galaxy formation

We present a systematic study of stellar feedback processes in simulations of disc 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 haloes spanning a range of masses in order to compare the results to disc 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 0.7-1 mag, depending on the employed IMF. Applying the standard model to Milky Way-type haloes with different assembly histories, we show that a quiet assembly history does not guarantee the formation of a disc. 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. A mechanism to create bulge-less disc galaxies in simulations therefore remains elusive. We conclude that disc formation is a complex, highly interconnected problem that is difficult to handle in a controlled manner. We expect a solution to be coming from a combination of small effects rather than one large breakthrough.