2019ApJ...874...22S

The Fe II/Mg II line flux ratio has been used to investigate the chemical evolution of high-redshift active galactic nuclei (AGNs). No strong evolution has been found out to z ∼ 6, implying that the SN 1a activity already occurred in the early universe. However, the trend of no evolution can be caused by the sample selection bias because previous studies have utilized mostly very luminous AGNs, which may be already chemically matured at the observed redshift. As motivated by the previously reported correlation between AGN luminosity and metallicity, we investigate the Fe II/Mg II flux ratio over a large dynamic range of luminosity, by adding a new sample of 12 quasars at z ∼ 3, of which the lower-luminosity limit is more than 1 dex smaller than that of the previously studied high-z quasars. Based on the Gemini/GNIRS observations, we find that the seven low-luminosity quasars with a mean bolometric luminosity log L_bol ∼ 46.5 ± 0.2 has an average Fe II/Mg II ratio of 0.68 ± 0.11 dex. This ratio is comparable to that of high-luminosity quasars (log L_bol ∼ 47.3 ± 0.3) in our sample (i.e., Fe II/Mg II ratio of 0.59 ± 0.15 dex) and that of the previously studied high-luminosity quasars at higher redshifts. One possible scenario is that the low-luminosity quasars in our sample are still relatively luminous and already chemically matured. To search for chemically young AGNs, and to fully understand the chemical evolution based on the Fe II/Mg II/flux ratio, investigations of much lower-luminosity AGNs are required.