We study how feedback influences baryon infall onto galaxies using cosmological, zoom-in simulations of haloes with present mass $M_{vir}=6.9times10^{11} M_{odot}$ to $1.7times10^{12} M_{odot}$. Starting at z=4 from identical initial conditions, impl
ementations of weak and strong stellar feedback produce bulge- and disc-dominated galaxies, respectively. Strong feedback favours disc formation: (1) because conversion of gas into stars is suppressed at early times, as required by abundance matching arguments, resulting in flat star formation histories and higher gas fractions; (2) because 50% of the stars form in situ from recycled disc gas with angular momentum only weakly related to that of the z=0 dark halo; (3) because late-time gas accretion is typically an order of magnitude stronger and has higher specific angular momentum, with recycled gas dominating over primordial infall; (4) because 25-30% of the total accreted gas is ejected entirely before z~1, removing primarily low angular momentum material which enriches the nearby inter-galactic medium. Most recycled gas roughly conserves its angular momentum, but material ejected for long times and to large radii can gain significant angular momentum before re-accretion. These processes lower galaxy formation efficiency in addition to promoting disc formation.
One quarter of all nearby early-type galaxies (ETGs) outside Virgo host a disc/ring of HI with size from a few to tens of kpc and mass up to ~1e+9 solar masses. Here we investigate whether this HI is related to the presence of a stellar disc within t
he host making use of the classification of ETGs in fast and slow rotators (FR/SR). We find a large diversity of HI masses and morphologies within both families. Surprisingly, SRs are detected as often, host as much HI and have a similar rate of HI discs/rings as FRs. Accretion of HI is therefore not always linked to the growth of an inner stellar disc. The weak relation between HI and stellar disc is confirmed by their frequent kinematical misalignment in FRs, including cases of polar and counterrotating gas. In SRs the HI is usually polar. This complex picture highlights a diversity of ETG formation histories which may be lost in the relative simplicity of their inner structure and emerges when studying their outer regions. We find that LCDM hydrodynamical simulations have difficulties reproducing the HI properties of ETGs. The gas discs formed in simulations are either too massive or too small depending on the star formation feedback implementation. Kinematical misalignments match the observations only qualitatively. The main point of conflict is that nearly all simulated FRs and a large fraction of all simulated SRs host corotating HI. This establishes the HI properties of ETGs as a novel challenge to simulations.
We analyze 40 cosmological re-simulations of individual massive galaxies with present-day stellar masses of $M_{*} > 6.3 times 10^{10} M_{odot}$ in order to investigate the physical origin of the observed strong increase in galaxy sizes and the decre
ase of the stellar velocity dispersions since redshift $z approx 2$. At present 25 out of 40 galaxies are quiescent with structural parameters (sizes and velocity dispersions) in agreement with local early type galaxies. At z=2 all simulated galaxies with $M_* gtrsim 10^{11}M_{odot}$ (11 out of 40) at z=2 are compact with projected half-mass radii of $approx$ 0.77 ($pm$0.24) kpc and line-of-sight velocity dispersions within the projected half-mass radius of $approx$ 262 ($pm$28) kms$^{-1}$ (3 out of 11 are already quiescent). Similar to observed compact early-type galaxies at high redshift the simulated galaxies are clearly offset from the local mass-size and mass-velocity dispersion relations. Towards redshift zero the sizes increase by a factor of $sim 5-6$, following $R_{1/2} propto (1+z)^{alpha}$ with $alpha = -1.44$ for quiescent galaxies ($alpha = -1.12$ for all galaxies). The velocity dispersions drop by about one-third since $z approx 2$, following $sigma_{1/2} propto (1+z)^{beta}$ with $beta = 0.44$ for the quiescent galaxies ($beta = 0.37$ for all galaxies). The simulated size and dispersion evolution is in good agreement with observations and results from the subsequent accretion and merging of stellar systems at $zlesssim 2$ which is a natural consequence of the hierarchical structure formation. A significant number of the simulated massive galaxies (7 out of 40) experience no merger more massive than 1:4 (usually considered as major mergers). On average, the dominant accretion mode is stellar minor mergers with a mass-weighted mass-ratio of 1:5. (abridged)
