We explore a detailed model in which the active galactic nucleus (AGN) obscuration results from the extinction of AGN radiation in a global flow driven by the pressure of infrared radiation on dust grains. We assume that external illumination by UV and soft X-rays of the dusty gas located at approximately 1pc away from the supermassive black hole is followed by a conversion of such radiation into IR. Using 2.5D, time-dependent radiation hydrodynamics simulations in a flux-limited diffusion approximation we find that the external illumination can support a geometrically thick obscuration via outflows driven by infrared radiation pressure in AGN with luminosities greater than $0.05\, L_{\rm edd}$ and Compton optical depth, $\tau_{\rm T}\gtrsim 1$.
A fundamental assumption of active galactic nuclei (AGN) unification schemes is that type 1 and type 2 AGNs have similar intrinsic properties. The basic premise of this paradigm is that obscuration and orientation effects are the major contributors to the observational dichotomy of AGNs. The goal of this paper is to suggest an approach which explains the AGN dichotomy as resulting from the extinction of the AGN radiation in a hydrodynamical outflow powered by the pressure of the infrared radiation on the dusty plasma of AGN outskirts.
The suggestion that Seyfert 2 galaxies suffer from enhanced extinction compared to Seyfert 1 galaxies was made by Rowan-Robinson (1977) based on the infrared observations. However it was not until the seminal work of Antonucci (1984), and Antonucci & Miller (1985) when key evidence was collected from studies based on optical spectropolarimetry. The detection of broad permitted lines in the polarized UV and optical spectrum of the nearby, luminous Seyfert 2 galaxy NGC 1068, confirmed that a bright, Seyfert 1 core is hidden behind optically thick, obscuring material. Notice that a prediction of polarization of the X-ray flux in the 0.1 -10 keV range was made by Dorodnitsyn & Kallman (2010) based on theoretical modeling of AGN outflows.
Direct evidence of the existence of toroidal obscuration comes from the mid-infrared observations of Seyfert 2 galaxies, such as the prototypical Seyfert 2 galaxy NGC 1068, and the closest AGN, the Circinus galaxy. Observations of NGC 1068 using VLTI reveal a multi-component, multi-temperature dusty conglomerate: an inner, relatively small (∼ 1 pc) and hot (∼ 800 K) component embedded into an outer (∼ 3.5 pc) component which is much colder (T ∼ 320 K) (Jaffe et al. 2004;Raban et al. 2009). In the Circinus galaxy the observed elongated, 0.4 pc in diameter component is interpreted as a disk-like structure seen almost edge-on. This disk-like structure is co-incident with that inferred from the VLBI maps of H 2 O maser emission (Greenhill et al. 2003), being embedded into a much larger rounded component. This is interpreted as a geometrically thick torus with temperature T 300K (Tristram et al. 2007).
The premise and the principal puzzle of AGN unification is the physical mechanism responsible for the geometrical thickness of the torus. Ample observational evidence for dust rules out support of the torus by gas pressure, as in such a case the temperature of the gas should be approximately of the order of the virial temperature, T vir,g = 2.6 × 10 6 M 7 /r pc K, where M 7 is the black hole (BH) mass in 10 7 M , and r pc is the distance in parsecs.
Various mechanisms have been proposed to settle this issue: for example, in one of the first models a torus was considered being made of clumps having highly supersonic velocities (Krolik & Begelman 1988;Beckert & Duschl 2004). Magnetic fields are implicitly necessary in this model to provide enough elasticity to the clouds in order to avoid large dissipation though cloud-cloud collisions. Another model suggested by Phinney (1989), and Sanders et al. (1989) considered a locally geometrically thin, but globally warped disk. Global magnetic fields were suggested to be a key ingredient either though hydromagnetic winds (Konigl & Kartje 1994), or as directly supporting the vertical balance of a quasi-static torus (Lovelace et al. 1998). The main difficulty with hydromagnetic models comes from the large poloidal magnetic flux needed to support such a wind. It is unclear whether such a strong global poloidal magnetic field exists at large distances from a BH. The clumpy nature of an outflow was addressed by Elitzur & Shlosman (2006). These authors consider the dusty hydromagnetic obscuring wind as an alternative to quasi-static torus models.
It has been pointed out by Pier & Krolik (1992) that the infrared radiation pressure on dust may suffice to balance the vertical gravitational force. Based on these ideas Krolik (2007), Shi & Krolik (2008) constructed a semi-analytic model of a static, infrared-supported torus. In all models in which infrared radiation is responsible for the torus thickness it is tacitly assumed that external radiation ranging from UV to soft X-rays is absorbed and converted into IR at the inner face of the torus.
To sum up, previous models of AGN torus obscuration divide with respect to whether they are i) static, i.e. such as in a model of self-gravitating clouds or in a model of a static IR supported torus; ii) dynamic, such as in hydromagnetic wind scenario.
In this paper we present simulations which demonstrate that obscuration at parses scale can be produced by global outflows driven by infrared radiation pressure on dust. Arguments for this model are both observational and theoretical. Simple estimates show that the observed dust temperatures (see above) translate into high infrared radiation energy densities. The latter when coupled with the high opacity of dust to IR radiation w
This content is AI-processed based on open access ArXiv data.
