Absorption- And Emission Lines Of Outflows In Active Galactic Nuclei

Among the various feedback mechanisms of active galactic nuclei (AGN) to the host galaxies, AGN outflows are the most common one. The ouflows produce broad absorption lines (BALs) of large blueshifted velocity offsets present in the ultraviolet spectra of 15-30% radio-quiet quasars, as well as blue excess components in the the asymmetric profiles of emision lines (particularly the high-ionization ones). Moreover, according to the numerical simulations, AGN outflows are regarded as the key agent to establish the famous MBE-σ+ relationship between central supermassive black holes and the bulges of their host galaxies. Recently direct imaging observations have re-vealed that AGN outflows can be driven as far as onto the spatial scales of 1-10 kpc. Yet, unfortunately, due to the limit of current facilties, only a few nearest AGNs have been able to be observed by such direct, spatially resolved observation. Thus it is urgent to find an approach to carry out direct, quantitative studies on the physical conditions of the outflows for a large number of AGNs.To this end, during these years of pursueing my PhD degree we have accomplished two projects:(1) systematically finding out He I* BAL QSOs and exploiting the merits of He I* BALs to measure properties of the outflowing gas, and (2) constraining the physical conditions of the outflowing gas by taking advantage of both the emission lines and absorption lines produced in the outflows.(1) Previous numerical simulations showed that AGN outflows impact signifi-cantly the host galaxies only when the loss rate of the kinetic energy of the outflows Ek≥0.05LEdd; after a rough calculation, it requires an outflowing gas of NH> 1022 cm-2. In other words, it is very important to study the low ionization absorption line (LoBAL) quasars whose outflows are of large column density. It is however difficult to study LoBALs in this respect because the common lines such as C Ⅳ and Mg Ⅱ are so easy to get saturated in high-NH outflows. In order to address this problem, we take a new way to exploit He i* absorption lines that are very useful, but hard to identify and thus seldom used in the literature. As the He I* atoms have much small abundance, He I* absorption multiplet lines, which spans in a wide wavelength from the UV to NIR, are not easy to get saturated; this enables the measurement of high column density and the diagnositcs of the partial covering situation of the outflow along the line of sight. Besides, as pho-toionization calculations demonstrated, He i* absorption lines are mainly produced in the layer tightly around the Hydrogen ionization front inside the gas slab, so even in the case of non-detection they can be used to estimate the upper limit to the clumn density. We have developed a set of procedure to find weak He I* BALs (currently in samples of Mg Ⅱ LoBAL quasars), systematically for the first time. In 285 MgⅡ BAL quasars, we find 101 (35.4%) having He I*λ3889 BALs. The fraction depends strongly on the the spectral S/N, reaching> 90%when S/N> 30. This suggests that (almost) all MgⅡ BAL quasars should have He I* BALs. By jointly using He I* and other absorption lines, we can determine the gas properties such as NH, ionization parameter U and density n, and further the distance to the AGN central engine, and finally the rate of kinetic energy injection and the mass loss rate of the outflow. To date this can be applied to only a few AGNs. Being of the the principal merit, this sample of Hei* BAL quasars enables the first large-scale quantitative analysis of the physical condition of AGN outflows. In addition, Hei*λλ3889,10830 lines being in the optical and NIR, we compiled from the SDSS/BOSS spectral data set the first homogenous sample of 19 low-z BAL quasars at z<0.3.(2) For long the outflow-produced emission lines and absorption lines are treated separately. Although the similarity-in an ensemble sense-between the BALs and the blueshifted components of C Ⅳ emission lines was occasionally noticed in the literature, no specific study has been performed on both the blueshifted absorption and emission lines of individual AGNs simultaneously. From the theoretical perspective, it is yet a natural picture for the outflows to produce both emission lines and absorption lines, which is well illustrated by photoionization modelling. In practice, both emission and absorption lines have their own advantages to determine the physcial parameters of the outflowing gas. The absorption lines trace the the gas along the sightline, giving accu-rate measurement of the LOS velocity and column density of the outflow, yet is not able to constrain the global properties of the outflow (e.g., the covering factor of the outflow to the AGN central engine). Instead, the emission lines reflect the global properties such as the geometry and mass of the outflow, yet have the weakness of involving too many parameters integrated together. Thus it is neccesary to use both complementarily, which is illustrated in our pilot studies of two cases, J1634+2049 and J1110+1930, as follows. J1634+2049 is a low-redshift IR-bright, ringed galaxy (zem= 0.1293), compiled in our sample of low-redshift He I BAL quasars. There are prominent He I*λλ3889,10830, Na I D BALs in its spectrum,besides apparent blueshifted components of its various emission lines. Using photoionization program CLOUDY to model their respect ab-sorption lines and emission lines, we find that the line-emitting gas and absorbing gas share a well constrained parameter space in terms of density (n) and ionization param-eter (U), which suggests that both emission and absorption lines are produced by the same outflowing gas. We thus combine the covering factor as derived from the emis-sion lines and the column density and velocity dervied from the absorption lines, and estimate the rate of the kinetic energy injection and the mass loss rate of the outflow in this AGN. J1110+1930 is a partially obscured quasar at a moderate redshift (z=2.512), with a special merit that the AGN torus obscures the FUV continuum and broad emis-sion lines seriously so that the emission lines origniated from the outflow exterior to the scale of the torus are present evidently in the observed spectrum (such as the rest-frame O Ⅵ, N v and C Ⅳ emission lines). Besides, the [OⅢ]A5007 emission line has a blueshifted component similar to the outflowing component in those UV lines. Using the aforementioned technique of comparing the line ratios with repect to the CLOUDY modeling, we determine the physcial properties of the outflowing gas. It is worthy of noting that the restframe equivalent widths of the outflow-produced UV lines are rather small, and thus without the partial obscuration of the torus the outflow emission whould be completed overshone by the normal BLR lines. Hence, in this second case we find another powerful approach, namely taking advantage of the partial obscuration of the torus, to study the properties of the outflows in a large population of AGNs.