Suzaku View of the Swift/BAT Active Galactic Nuclei (I): Spectral Analysis of Six AGNs and Evidence for Two Types of Obscured Population

We present a systematic spectral analysis with Suzaku of six AGNs detected in the Swift/BAT hard X-ray (15–200 keV) survey, Swift J0138.6-4001, J0255.2-0011, J0350.1-5019, J0505.7-2348, J0601.9-8636, and J1628.1-5145. This is considered to be a representative sample of new AGNs without X-ray spectral information before the BAT survey. We find that the 0.5–200 keV spectra of these sources can be uniformly fit with a base model consisting of heavily absorbed (log $N_{\rm{H}} > 23.5 \rm{cm}^{-2}$) transmitted components, scattered lights, a reflection component, and an iron-K emission line. There are two distinct groups, three “new type” AGNs (including the two sources reported by \citealt{Ueda2007}) with an extremely small scattered fraction ($f_{\rm{scat}} < 0.5%$) and strong reflection component ($R = \Omega / 2 \pi \gtrsim 0.8$ where $\Omega$ is the solid angle of the reflector), and three “classical type” ones with $f_{\rm{scat}} > 0.5%$ and $R \lesssim 0.8$. The spectral parameters suggest that the new type has an optically thick torus for Thomson scattering ($N_{\rm{H}} \sim 10^{25} \rm{cm}^{-2}$) with a small opening angle $\theta \sim 20^{\circ}$ viewed in a rather face-on geometry, while the classical type has a thin torus ($N_{\rm{H}} \sim 10^{23-24} \ \rm{cm}^{-2}$) with $\theta \gtrsim 30^{\circ}$. We infer that a significant number of new type AGNs with an edge-on view is missing in the current all-sky hard X-ray surveys.

💡 Research Summary

This paper presents a comprehensive broadband X‑ray spectral analysis of six active galactic nuclei (AGNs) that were newly identified in the Swift/BAT all‑sky hard X‑ray survey. The authors combined Suzaku observations (XIS covering 0.5–10 keV and HXD/PIN covering 15–70 keV) with the long‑term Swift/BAT spectra (15–200 keV) to obtain a continuous 0.5–200 keV view of each source. The sample—Swift J0138.6‑4001, J0255.2‑0011, J0350.1‑5019, J0505.7‑2348, J0601.9‑8636, and J1628.1‑5145—represents a set of AGNs that previously lacked any detailed X‑ray spectral information.

The authors adopted a physically motivated baseline model consisting of four components: (1) a heavily absorbed primary power‑law transmitted through a line‑of‑sight column density log NH > 23.5 (cm⁻²), (2) a scattered fraction of the primary continuum (parameterized as f_scatt), (3) a cold‑reflection component modeled with pexrav, expressed as R = Ω/2π, and (4) a narrow Fe‑Kα emission line at 6.4 keV. All six spectra are well described by this model, but the fitted parameters separate the objects into two distinct populations.

Three sources belong to a “new type” class characterized by an extremely low scattered fraction (f_scatt < 0.5 %) and a strong reflection component (R ≳ 0.8). Their line‑of‑sight column densities are of order 10^25 cm⁻², indicating a Compton‑thick, optically thick torus that almost completely blocks Thomson scattering. The authors infer a torus geometry with a very small half‑opening angle (θ ≈ 20°) and a relatively face‑on viewing angle, which explains why the direct transmitted component is still detectable despite the extreme obscuration.

The remaining three AGNs form a “classical type” group with f_scatt > 0.5 % and modest reflection (R ≲ 0.8). Their column densities lie in the 10^23–10^24 cm⁻² range, implying a thinner torus. The inferred half‑opening angles are larger (θ ≥ 30°), allowing a broader range of viewing angles to expose both the transmitted and scattered components.

These findings have important implications for population studies of obscured AGNs. Hard X‑ray surveys such as Swift/BAT are highly efficient at detecting the “new type” objects only when the line of sight is close to the torus axis; edge‑on orientations suffer from severe attenuation and are likely under‑represented. Consequently, the true space density of heavily buried AGNs with small opening angles may be substantially larger than current surveys suggest. By modeling the torus geometry (NH, opening angle, and inclination) the authors quantify this selection bias and propose that forthcoming missions with higher sensitivity and broader energy coverage (eROSITA, Athena, XRISM) will be essential to uncover the missing edge‑on population.

In summary, the Suzaku–BAT joint analysis demonstrates that a simple, uniform spectral model can capture the diversity of heavily obscured AGNs, while revealing two physically distinct obscuration regimes. The work underscores the need to correct for orientation‑dependent detection biases in hard X‑ray surveys and provides a framework for interpreting future all‑sky hard X‑ray catalogs in the context of torus geometry and AGN evolution.