[Paper Review] Three-dimensional radiative transfer modeling of AGN dusty tori as a clumpy two-phase medium

Marko Stalevski, J. Fritz+3 more|arXiv (Cornell University)|Sep 6, 2011

Astrophysics and Star Formation StudiesPhysics and Astronomy70 references134 citations

TL;DR

This paper presents a 3D Monte Carlo radiative transfer model of AGN dusty tori as a clumpy two-phase medium, combining high-density clumps and low-density interclump material. It shows that the two-phase structure naturally reproduces near-infrared emission deficits seen in observations, resolving a key shortcoming of clumps-only models, while the silicate feature strength remains consistent across smooth, clumpy, and two-phase configurations under varying parameters.

ABSTRACT

We investigate the emission of active galactic nuclei (AGN) dusty tori in the infrared domain. Following theoretical predictions coming from hydrodynamical simulations, we model the dusty torus as a 3D two-phase medium with high-density clumps and low-density medium filling the space between the clumps. Spectral energy distributions (SED) and images of the torus at different wavelengths are obtained using 3D Monte Carlo radiative transfer code SKIRT. Our approach of generating clumpy structure allows us to model tori with single clumps, complex structures of merged clumps or interconnected sponge-like structure. A corresponding set of clumps-only models and models with smooth dust distribution is calculated for comparison. We found that dust distribution, optical depth, clump size and their actual arrangement in the innermost region, all have an impact on the shape of near- and mid-infrared SED. The 10 micron silicate feature can be suppressed for some parameters, but models with smooth dust distribution are also able to produce a wide range of the silicate feature strength. Finally, we find that having the dust distributed in a two-phase medium, might offer a natural solution to the lack of emission in the near-infrared, compared to observed data, which affects clumpy models currently available in the literature.

Motivation & Objective

• To address the persistent lack of near-infrared emission in current clumpy AGN torus models, which contradicts observed SEDs.
• To investigate how dust distribution—specifically a two-phase medium of clumps and low-density interclump material—affects infrared SEDs and silicate features.
• To compare the infrared emission properties of two-phase models with clumps-only and smooth dust distributions under identical global parameters.
• To determine whether the two-phase model can naturally explain the observed silicate feature strength and near- and mid-infrared SED shapes in AGN.
• To explore the impact of clump size, optical depth, and spatial arrangement on the emergent SED and silicate feature morphology.

Proposed method

• The study employs a 3D Monte Carlo radiative transfer code (SKIRT) to simulate radiative transfer through a geometrically thick, axisymmetric dusty torus.
• The torus is modeled as a two-phase medium: dense, optically thick clumps embedded in a lower-density interclump medium, based on hydrodynamical simulation predictions.
• A grid of models is generated with varying clump size (σ = 12.5 or 100 pc), optical depth, and radial/polar density profiles (p, q parameters), with random clump distributions.
• For each two-phase model, corresponding clumps-only and smooth-dust models are created with identical global mass, luminosity, and radial density profiles.
• SEDs and images are computed at multiple wavelengths, and the silicate feature strength and near-infrared emission are analyzed across model types.
• The models are compared across inclination angles and parameter sets to isolate the effects of dust distribution and clump morphology.

Experimental results

Research questions

• RQ1 Does modeling the dusty torus as a two-phase medium—clumps plus interclump material—resolve the near-infrared emission deficit observed in clumps-only models?
• RQ2 How do clump size, optical depth, and spatial arrangement affect the shape of the near- and mid-infrared SED and the silicate feature?
• RQ3 Can two-phase models reproduce the full range of observed silicate feature strengths seen in AGN, including both emission and absorption?
• RQ4 Is the silicate feature suppression in clumpy models a robust feature, or does it depend on specific dust distribution parameters?
• RQ5 How does the inclusion of low-density interclump material influence the near-infrared flux compared to clumps-only models?

Key findings

• The two-phase model significantly enhances near-infrared emission compared to clumps-only models, resolving a key discrepancy with observed SEDs.
• Clumpy models with small clumps (σ = 100 pc) produce SEDs and silicate features very similar to smooth-dust models, with shallow silicate absorption and reduced near-IR emission.
• Larger clumps (σ = 12.5 pc) lead to more pronounced differences: reduced silicate emission depth and further suppression of near-IR flux compared to smooth models.
• The silicate feature strength in clumpy models is not systematically suppressed; it depends strongly on clump size, optical depth, and radial/polar density distribution (p, q).
• Models with non-constant radial density (p=1) and constant polar density (q=0) show the strongest silicate suppression, but this effect weakens or vanishes as q increases.
• Smooth-dust models can reproduce the same range of silicate feature strengths as two-phase and clumps-only models, indicating that silicate feature strength alone cannot distinguish between dust distribution types.