A compact, metal-rich, kpc-scale outflow in FBQS J0209-0438: Detailed diagnostics from HST/COS extreme UV observations

Abstract

We present HST/COS observations of highly ionized absorption lines associated with a radio-loud QSO at $z=1.1319$. The absorption system has multiple velocity components, tracing gas that is largely outflowing from the QSO at velocities of a few 100 km s$^{-1}$. There is an unprecedented range in ionization, with detections of HI, NIII, NIV, NV, OIV, OIV*, OV, OVI, NeVIII, MgX, SV and ArVIII. We estimate the total hydrogen number density from the column density ratio N(OIV*)/N(OIV) to be $\log(n_{\textrm{H}}/\textrm{cm}^3)\sim 3$. Assuming photoionization equilibrium, we derive a distance to the absorbing complex of $2.3<R<6.0$ kpc from the centre of the QSO. A range in ionization parameter, covering $\sim 2$ orders of magnitude, suggest absorption path lengths in the range $10^{-4.5}<l_{\textrm{abs}}<1$ pc. In addition, the absorbing gas only partially covers the background emission from the QSO continuum, which suggests clouds with transverse sizes $l_{\textrm{trans}}<10^{-2.5}$ pc. Widely differing absorption path lengths, combined with covering fractions less than unity across all ions pose a challenge to models involving simple cloud geometries. These issues may be mitigated by the presence of non-equilibrium effects, together with the possibility of multiple gas temperatures. The dynamics and expected lifetimes of the gas clouds suggest that they do not originate from close to the AGN, but are instead formed close to their observed location. Their inferred distance, outflow velocities and gas densities are broadly consistent with scenarios involving gas entrainment or condensations in winds driven by either supernovae, or the supermassive black hole accretion disc. In the case of the latter, the present data most likely does not trace the bulk of the outflow by mass, which could instead manifest itself as an accompanying warm absorber, detectable in X-rays.