AGN Feedback Efficiency of NAL Quasars

Abstract: We consider if outflowing winds that are detected via narrow absorption lines (NALs) with FWHM of $<$ 500 km/s (i.e., NAL outflows) in quasar spectra contribute to feedback. As our sample, we choose 11 NAL systems in eight optically luminous quasars from the NAL survey of Misawa et al. (2007a), based on the following selection criteria: i) they exhibit ``partial coverage'' suggesting quasar origin (i.e., intrinsic NALs), ii) they have at least one low-ionization absorption line (C II and/or Si II), and iii) the Ly$\alpha$ absorption line is covered by available spectra. The results depend critically on this selection method, which has caveats and uncertainties associated with it, as we discuss in a dedicated section of the paper. Using the column density ratio of the excited and ground states of C II and Si II, we place upper limits on the electron density as $n_{\rm e}$ $<$ 0.2 - 18 cm$^{-3}$ and lower limits on their radial distance from the flux source $R$ as greater than several hundreds of kpc. We also calculate lower limits on the mass outflow rate and kinetic luminosity of $\log(\dot{M}/{\rm M_{\odot}~s}^{-1}) > 79$ - (3.1$\times 10^{5})$ and $\log(\dot{E_{\rm k}}/{\rm erg~s}^{-1}) > 42.9$ - 49.8, respectively. Taking the NAL selection and these results at face value, the inferred feedback efficiency can be comparable to or even larger than those of broad absorption line and other outflow classes, and large enough to generate significant AGN feedback. However, the question of the connection of quasar-driven outflows to NAL absorbers at large distances from the central engine remains open and should be addressed by future theoretical work.