We study the properties of massive, galactic-scale outflows of molecular gas and investigate their impact on galaxy evolution. We present new IRAM PdBI CO(1-0) observations of local ULIRGs and QSO hosts: clear signature of massive and energetic molecular outflows, extending on kpc scales, is found in the CO(1-0) kinematics of four out of seven sources, with measured outflow rates of several 100 M yr −1 . We combine these new observations with data from the literature, and explore the nature and origin of massive molecular outflows within an extended sample of 19 local galaxies. We find that starburst-dominated galaxies have an outflow rate comparable to their SFR, or even higher by a factor of ∼2-4, implying that starbursts can indeed be effective in removing cold gas from galaxies. Nevertheless, our results suggest that the presence of an AGN can boost the outflow rate by a large factor, which is found to increase with the L AGN /L bol ratio. The gas depletion time-scales due to molecular outflows are anti-correlated with the presence and luminosity of an AGN in these galaxies, and range from a few hundred million years in starburst galaxies, down to just a few million years in galaxies hosting powerful AGNs. In quasar hosts the depletion time-scales due to the outflow are much shorter than the depletion time-scales due to star formation. We estimate the outflow kinetic power and find that, for galaxies hosting powerful AGNs, it corresponds to about 5% of the AGN luminosity, as expected by models of AGN feedback. Moreover, we find that momentum rates of about 20 L AGN /c are common among the AGN-dominated sources in our sample. For "pure" starburst galaxies our data tentatively support models in which outflows are mostly momentum-driven by the radiation pressure from young stars onto dusty clouds. Overall, our results indicate that, although starbursts are effective in powering massive molecular outflows, the presence of an AGN may strongly enhance such outflows and, therefore, have a profound feedback effect on the evolution of galaxies, by efficiently removing fuel for star formation, hence quenching star formation.

Context. Feedback from accreting supermassive black holes is often identified as the main mechanism responsible for regulating star-formation in AGN host galaxies. However, the relationships between AGN activity, radiation, winds, and star-formation are complex and still far from being understood. Aims. We study scaling relations between AGN properties, host galaxy properties and AGN winds. We then evaluate the wind mean impact on the global star-formation history, taking into account the short AGN duty cycle with respect to that of star-formation. Methods. We first collect AGN wind observations for 94 AGN with detected massive winds at sub-pc to kpc spatial scales. We then fold AGN wind scaling relations with AGN luminosity functions, to evaluate the average AGN wind mass-loading factor as a function of cosmic time. Results. We find strong correlations between the AGN molecular and ionised wind mass outflow rates and the AGN bolometric luminosity. The power law scaling is steeper for ionised winds (slope 1.29±0.38) than for molecular winds (0.76±0.06), meaning that the two rates converge at high bolometric luminosities. The molecular gas depletion timescale and the molecular gas fraction of galaxies hosting powerful AGN driven winds are 3-10 times shorter and smaller than those of main-sequence galaxies with similar star-formation rate, stellar mass and redshift. These findings suggest that, at high AGN bolometric luminosity, the reduced molecular gas fraction may be due to the destruction of molecules by the wind, leading to a larger fraction of gas in the atomic ionised phase. The AGN wind mass-loading factor η =ṀOF /SFR is systematically higher than that of starburst driven winds. Conclusions. Our analysis shows that AGN winds are, on average, powerful enough to clean galaxies from their molecular gas only in massive systems at z < ∼ 2, i.e. a strong form of co-evolution between SMBHs and galaxies appears to break down for the least massive galaxies.

In the standard scenario for galaxy evolution young star-forming galaxies transform into red bulge-dominated spheroids, where star formation has been quenched. To explain this transformation, a strong negative feedback generated by accretion onto a central supermassive black hole is often invoked. The depletion of gas resulting from quasar-driven outflows should eventually stop star-formation across the host galaxy and lead the black hole to "suicide" by starvation. Direct observational evidence for a major quasar feedback onto the host galaxy is still missing, because outflows previously observed in quasars are generally associated with the ionized component of the gas, which only accounts for a minor fraction of the total gas content, and typically occurrs in the central regions. We used the IRAM PdB Interferometer to observe the CO(1−0) transition in Mrk 231, the closest quasar known. Thanks to the wide band we detected broad wings of the CO line, with velocities of up to 750 km s −1 and spatially resolved on the kpc scale. These broad CO wings trace a giant molecular outflow of about 700 M /year, far larger than the ongoing star-formation rate (∼200 M /year) observed in the host galaxy. This wind will totally expel the cold gas reservoir in Mrk 231 in about 10 7 yrs, therefore halting the starformation activity on the same timescale. The inferred kinetic energy in the molecular outflow is ∼1.2 × 10 44 erg/s, corresponding to a few percent of the AGN bolometric luminosity, which is very close to the fraction expected by models ascribing quasar feedback to highly supersonic shocks generated by radiatively accelerated nuclear winds. Instead, the contribution by the SNe associated with the starburst fall short by several orders of magnitude to account for the kinetic energy observed in the outflow. The direct observational evidence for quasar feedback reported here provides solid support to the scenarios ascribing the observed properties of local massive galaxies to quasar-induced large-scale winds.

Abstract. We present results of a systematic analysis of the XMM-Newton spectra of 40 quasars (QSOs) (z ≤ 1.72) from the Palomar-Green (PG) Bright Quasar Survey sample (M B < −23). The sample includes 35 radio-quiet quasars (RQQs) and 5 radio-loud quasars (RLQs). The analysis of the spectra above 2 keV reveals that the hard X-ray continuum emission can be modeled with a power law component with Γ 2−12 keV = 1.89 ± 0.11 and Γ 2−12 keV = 1.63−0.01 for the RQQs and RLQs, respectively. Below 2 keV, a strong, broad excess is present in most QSO spectra. This feature has been fitted with four different models assuming several physical scenarios. All tested models (blackbody, multicolor blackbody, bremsstrahlung and power law) satisfactorily fitted the majority of the spectra. However, none of them is able to provide an adequate parameterization for the soft excess emission in all QSOs, indicating the absence of a universal shape for this spectral feature. An additional cold absorption component was required only in three sources. On the other hand, as recently pointed out by Porquet et al. (2004) for a smaller sample of PG QSOs, warm absorber features are present in 50% of the QSO spectra in contrast with their rare occurrence (∼5−10%) in previous studies. The XMM-Newton view of optically selected bright QSOs therefore suggests that there are no significant differences in the X-ray spectral properties compared with those of the low-luminosity Seyfert 1 galaxies. Properties of the Fe Kα emission lines are presented in a companion paper.