Unveiling Intrinsic Properties of Dusty Red AGNs

Abstract

Theoretical models suggest that dust-obscured active galactic nuclei (AGNs) appear for a certain period when merger-driven star forming galaxies evolve to galaxies harboring unobscured type 1 AGNs. These dust-obscured AGNs would look red due to the dust extinction, but observational properties of such an AGN population have not been studied extensively so far, leaving a hole in the understanding of the AGN evolution scheme. The most important expected property of the intermediate-stage, dusty AGNs is that they have higher accretion rates than unobscured type 1 AGNs, and this needs to be tested observationally. Red AGNs have been sampled in differ ent ways in the hope of identifying the intermediate-stage dusty AGNs, but it is not yet clear if they really correspond to the dusty AGNs as suggested in the models. In this thesis, we study the near-infrared (NIR) and mid-infrared (MIR) spectra of unobscured type 1 AGNs and red AGNs that are selected in various ways. There are three main themes: (i) derivation of black hole (BH) mass estimators and line diagnostics that can be used for red AGN study

(ii) investigation of red AGN selection methods to test its usefulness to identify dusty AGNs

and (iii) investigation of the accretion rates of red, dusty AGNs to see if they have properties consistent with the intermediate-stage, dusty AGNs as predicted in the models. First, we derive methods to measure the BH masses of red AGNs using NIR and MIR hydrogen lines that are not strongly affected by dust extinction. It is necessary to find such a method, since the dust-obscuration in red AGNs will make it challenging to use the BH estimators that are based on the optical/UV lines. This is done by investigating the AKARI MIR spectra of unobscured type 1 AGNs. We derive Brβ- and Brα-based BH estimators and find that BH masses can be estimated with an accuracy of 0.20–0.36 dex using these estimators. We also investigate the Balmer/Paschen/Brackett line luminosity ratios of unobscured type 1 AGNs and find that these line ratios are consistent with the theoretically expected line ratios from CLOUDY code. Together with Paschen line-based MBH estimators and these line diagnostics provide tools to understand properties of red AGNs. Moreover, we examine how the hot and warm dust components of the type 1 AGNs behave by adding AKARI and WISE MIR data points to the analysis. The measured temperatures of the hot and warm dust components are ∼1100K and ∼220K, respectively, and the hot dust temperatures are somewhat cooler than the value quoted in previous studies (∼1500K). Second, we test how effective the NIR or optical-NIR color selections of red AGNs are in identifying dusty AGNs. In order to test it, their rest-frame optical to NIR spectroscopic properties are examined to see if a reliable estimation of the dust extinction is possible. More specifically, we tested two red AGN selection methods, one by using NIR color (J −K > 2mag) and another using optical-NIR color (r0−K > 4mag and J −K > 1.3mag) with FIRST radio detection criteria. For the red AGNs selected from the NIR color, we measure the E(B −V ) values in two ways by using line luminosity ratios and continuum slopes, finding that all of the g0 −K . 4 NIR-selected red AGNs (∼40% of the NIR-selected red AGNs) have no significant dust extinction (E(B −V ) ∼ 0). In contrast, for the red AGNs selected from the optical-NIR colors, their E(B−V ) values are ∼0.804 (from 0.275 to 3.050), and the line luminosity ratios, from Hβ to Pα line, are difficult to explain without dust obscuration. Third, armed with the knowledges gained through the first and second themes, we apply several hydrogen NIR-based BH mass estimators to derive the BH masses and the bolometric luminosities of red AGNs. This is done by using red AGNs z ∼ 0.3 and z ∼ 0.7 selected from the optical-NIR color selection. We find that the measured Eddington ratios of red AGNs (∼0.69) are higher than those of unobscured type 1 AGNs by a factor of ∼4, consistent with the expectation from some merger driven galaxy evolution models. We derived the NIR-line based MBH estimators and the line ratio diagnostics to study dust-obscuration and BH masses in red AGNs. With these tools, we find that radio and optical-NIR colors can effectively select dust-obscured AGNs. Finally, we show that the red AGNs selected this way have high accretion rates, and we suggest that they are the intermediate-stage, dusty AGNs in the merger-driven galaxy evolution models.