Improved Fe ii Emission-line Models for AGNs Using New Atomic Data Sets

Understanding the Fe II emission from active galactic nuclei (AGNs) has been a grand challenge for many decades. The rewards from understanding the AGN spectra would be immense, involving both quasar classification schemes such as “Eigenvector 1” and tracing the chemical evolution of the cosmos. Recently, three large Fe II atomic data sets with radiative and electron collisional rates have become available. We have incorporated these into the spectral synthesis code Cloudy and examined predictions using a new generation of AGN spectral energy distribution (SED), which indicates that the ultraviolet (UV) emission can be quite different depending on the data set utilized. The Smyth et al. data set better reproduces the observed Fe II template of the I ZW 1 Seyfert galaxy in the UV and optical regions, and we adopt these data. We consider both thermal and microturbulent clouds and show that a microturbulence of ≈100 km s −1 reproduces the observed shape and strength of the so-called Fe II “UV bump.” Comparing our predictions to the observed Fe II template, we derive a typical cloud density of 10 11 cm −3 and photon flux of 10 20 cm −2 s −1 , and show that these largely reproduce the observed Fe II emission in the UV and optical. We calculate the I (Fe II )/ I (Mg II ) emission-line intensity ratio using our best-fitting model and obtain log( I (Fe II )/ I (Mg II )) ∼ 0.7, suggesting many AGNs have a roughly solar Fe/Mg abundance ratio. Finally, we vary the Eddington ratio and SED shape as a step in understanding the Eigenvector 1 correlation.