Paper detail

Constraining the dynamical importance of hot gas and radiation pressure in quasar outflows using emission line ratios

Quasar feedback models often predict an expanding hot gas bubble which drives a galaxy-scale outflow. In many circumstances this hot gas radiates inefficiently and is therefore difficult to observe directly. We present an indirect method to detect the presence of a hot bubble using hydrostatic photoionization calculations of the cold (~10^4 K) line-emitting gas. We compare our calculations with observations of the broad line region, the inner face of the torus, the narrow line region (NLR), and the extended NLR, and thus constrain the hot gas pressure at distances 0.1 pc -- 10 kpc from the center. We find that emission line ratios observed in the average quasar spectrum are consistent with radiation-pressure-dominated models on all scales. On scales <40 pc a dynamically significant hot gas pressure is ruled out, while on larger scales the hot gas pressure cannot exceed six times the local radiation pressure. In individual quasars, ~25% of quasars exhibit NLR ratios that are inconsistent with radiation-pressure-dominated models, though in these objects the hot gas pressure is also unlikely to exceed the radiation pressure by an order of magnitude or more. The derived upper limits on the hot gas pressure imply that the instantaneous gas pressure force acting on galaxy-scale outflows falls short of the time-averaged force needed to explain the large momentum fluxes \dot{p} >> L_AGN/c inferred for galaxy-scale outflows. This apparent discrepancy can be reconciled if optical quasars previously experienced a buried, fully-obscured phase during which the hot gas bubble was more effectively confined and during which galactic wind acceleration occurred.