Suzaku reveals X-ray continuum piercing the nuclear absorber in Markarian 231

We report the results from a 2011 Suzaku observation of the nearby low-ionization BAL quasar/ULIRG Markarian 231. These data reveal that the X-ray spectrum has undergone a large variation from the 2001 XMM-Newton and BeppoSAX observations. We interpret this finding according to a scenario whereby the X-ray continuum source is obscured by a two-component partial-covering absorber with NH ~10^22 and ~10^24 cm^-2, respectively. The observed spectral change is mostly explained by a progressive appearance of the primary continuum at <10 keV due to the decrease of the covering fraction of the denser absorption component. The properties of the X-ray obscuration in Mrk 231 match well with those of the X-ray shielding gas predicted by the theoretical models for an efficient radiatively-driven acceleration of the BAL wind. In particular, the X-ray absorber might be located at the extreme base of the outflow. We measure a 2-10 keV luminosity of L(2-10) = 3.3 x 10^43 erg s^-1 for the 2011 data set, i.e. an increase of 30% with respect to the 2001 value.

💡 Research Summary

The authors present a comprehensive analysis of a 2011 Suzaku observation of the nearby low‑ionization broad‑absorption‑line (BAL) quasar and ultra‑luminous infrared galaxy Markarian 231 (Mrk 231). Compared with the 2001 XMM‑Newton and BeppoSAX data, the Suzaku spectrum shows a pronounced change in shape and flux, especially below 10 keV. By fitting the 0.5–30 keV data with a physically motivated model that includes a primary power‑law continuum (photon index Γ ≈ 1.8), two partial‑covering absorbers, a distant reflection component, and a modest Fe Kα emission line, the authors demonstrate that the observed variability can be explained primarily by a reduction in the covering fraction of the denser absorber.

The two absorbers have column densities of roughly 10^22 cm⁻² (the “warm” component) and 10^24 cm⁻² (the “shielding” component). In the 2001 observations the high‑column absorber covered about 60–70 % of the X‑ray source, effectively suppressing the intrinsic continuum below 10 keV. In the Suzaku data the covering fraction of this component drops by ~0.2–0.3, allowing a fraction of the intrinsic continuum to leak through. The lower‑column absorber shows little change, suggesting it resides on larger scales or is more homogeneous. The reflection fraction and the Fe Kα line (centroid ≈ 6.4 keV, EW ≈ 80 eV) remain consistent between epochs, indicating that the distant torus‑like reflector is stable over the same timescale.

The intrinsic 2–10 keV luminosity derived from the Suzaku spectrum is L_X = 3.3 × 10^43 erg s⁻¹, about 30 % higher than the value inferred from the 2001 data. This increase is not due to a genuine brightening of the central engine but rather to the partial unveiling of the continuum as the high‑column absorber becomes less opaque. The authors argue that the high‑column, partially covering gas corresponds to the “X‑ray shielding gas” required in radiatively driven BAL wind models. Such a shield must have N_H ≈ 10^23–10^24 cm⁻², be located close to the accretion disc (tens of parsecs or less), and protect the outflowing wind from over‑ionization, thereby enabling efficient line‑driven acceleration.

Temporal variability of the covering fraction on a timescale of a few years suggests dynamical motions of clumpy gas clouds across the line of sight, or structural changes in the absorber (e.g., expansion, fragmentation). The stability of the distant reflector and Fe Kα line implies that the larger‑scale torus is decoupled from these rapid inner‑region changes.

In the broader context, the study provides direct observational support for the theoretical picture in which BAL quasars possess a compact, high‑column, partially covering X‑ray absorber that acts as a shield for the UV‑line driven wind. The detection of a modest increase in the intrinsic X‑ray luminosity also alleviates previous concerns that Mrk 231 might be intrinsically X‑ray weak; instead, its apparent weakness is largely a line‑of‑sight effect.

The paper concludes that (1) partial‑covering absorbers in Mrk 231 are variable and can significantly modulate the observed X‑ray spectrum, (2) the high‑column component likely resides at the base of the BAL outflow and fulfills the role of the theoretical shielding gas, and (3) long‑term, high‑resolution X‑ray monitoring (e.g., with XRISM or Athena) will be essential to map the geometry, kinematics, and physical conditions of these absorbers, thereby refining our understanding of AGN feedback and the co‑evolution of supermassive black holes and their host galaxies.