X-Ray Properties of the Northern Galactic Cap Agns in the 58-Month
Total Page:16
File Type:pdf, Size:1020Kb
Accepted for publication in ApJ Preprint typeset using LATEX style emulateapj v. 08/22/09 X-RAY PROPERTIES OF THE NORTHERN GALACTIC CAP SOURCES IN THE 58-MONTH SWIFT-BAT CATALOG Ranjan V. Vasudevan1,*, William N. Brandt2,3, Richard F. Mushotzky1, Lisa M. Winter4, Wayne H. Baumgartner5, Thomas T. Shimizu1, Donald. P. Schneider2,3, John Nousek2 Accepted for publication in ApJ ABSTRACT We present a detailed X-ray spectral analysis of a complete sample of hard X-ray selected AGN in the Northern Galactic Cap of the 58-month Swift Burst Alert Telescope (Swift/BAT) catalog, consisting of 100 AGN with b > 50◦. This sky area has excellent potential for further dedicated study due to a wide range of multi-wavelength data that are already available, and we propose it as a low-redshift analog to the ‘deep field’ observations of AGN at higher redshifts (e.g. CDFN/S, COSMOS, Lockman Hole). We present distributions of luminosity, absorbing column density, and other key quantities for the catalog. We use a consistent approach to fit new and archival X-ray data gathered from XMM-Newton, Swift/XRT, ASCA and Swift/BAT. We probe to deeper redshifts than the 9-month BAT catalog (hzi =0.043 compared to hzi =0.03 for the 9-month catalog), and uncover a broader absorbing column density distribution. The fraction of obscured (logNH ≥ 22) objects in the sample is ∼ 60%, and 43–56% of the sample exhibits ‘complex’ 0.4–10 keV spectra. We present the properties of iron lines, soft excesses and ionized absorbers for the subset of ob- jects with sufficient signal-to-noise ratio. We reinforce previous determinations of the X-ray Baldwin (Iwasawa-Taniguchi) effect for iron K-α lines. We also identify two distinct populations of sources; one in which a soft excess is well-detected and another where the soft excess is undetected, suggesting that the process responsible for producing the soft excess is not at work in all AGN. The fraction of Compton-thick sources (log NH > 24.15) in our sample is ∼ 9%. We find that ‘hidden/buried AGN’, (which may have a geometrically thick torus or emaciated scattering regions) constitute ∼ 14% of our sample, including seven objects previously not identified as hidden. Compton reflection is found to be important in a large fraction of our sample using joint XMM-Newton+BAT fits (hRi = 2.7 ± 0.75), indicating light bending or extremely complex absorption. High energy cut-offs generally lie outside the BAT band (E > 200keV) but are seen in some sources. We present the average 1–10 keV spectrum for the sample, which reproduces the 1–10 keV X-ray background slope as found for the brighter 9- month BAT AGN sample. The 2–10 keV log(N)-log(S) plot implies completeness down to fluxes ∼ 4 times fainter than seen in the 9-month catalog. We emphasize the utility of this Northern Galactic Cap sample for a wide variety of future studies on AGN. Subject headings: 1. INTRODUCTION in AGN studies because absorbed AGN are thought to Active galactic nuclei (AGN) are among the most pow- constitute a significant proportion of the overall AGN erful energy sources in the Universe, and their luminous population; it is therefore essential to have as complete output is due to accretion onto supermassive black holes a survey of AGN as possible across a range of absorbing column densities before one draws conclusions about the arXiv:1212.2957v1 [astro-ph.HE] 12 Dec 2012 (‘SMBHs’, e.g. Rees 1984). Strong emission from AGN has been observed across the entire spectrum, including AGN population as whole. at radio, sub-mm, infrared, optical and ultraviolet wave- While X-ray surveys of AGN are more penetrating lengths, but an invaluable key to understanding them is than optical ones (e.g., Mushotzky 2004), those AGN with heavy obscuration (with neutral Hydrogen column provided by X-ray observations. X-rays are not subject 23−24 to the heavy host-galaxy dilution present in other bands, density NH > 10 ) still can fall out of the purview and can penetrate through greater amounts of absorbing of typical X-ray imaging satellites. Progressively larger material in the line of sight than is possible with observa- amounts of absorption depress the fluxes at successively tions in other wavebands. This last feature is important higher energies, and eventually the 0.4–10 keV, band ex- plored by observatories such as XMM-Newton and Chan- 1 Department of Astronomy, University of Maryland, College dra, becomes insufficient to identify and constrain ab- Park, MD sorption levels in highly-absorbed AGN. Sensitivity at 2 Department of Astronomy & Astrophysics, The Pennsylvania > 10 keV is required to obtain a more complete AGN cen- State University, 525 Davey Lab, University Park PA 16802 3 Institute for Gravitation and the Cosmos, The Pennsylvania sus. The Burst Alert Telescope (BAT, Barthelmy et al. State University, University Park, PA 16802, USA 2005) on the Swift satellite has proven extremely useful 4 Atmospheric and Environmental Research, 131 Hartwell Av- for this purpose, and is producing an all-sky survey of enue, Lexington, MA AGN in the 14–195 keV band, the Swift/BAT catalog of 5 NASA/Goddard Space Flight Center, Astrophysics Science Di- vision, Greenbelt, MD 20771 AGN. The survey is augmented by detections at increas- * [email protected] ing depth as the BAT instrument continues to survey the 2 sky. Source lists and sample properties have been pre- the importance of each of these goals below. sented for the catalog after the first 9 months of survey- Understanding the true distribution of column den- ing (Tueller et al. 2008), 22 months (Tueller et al. 2010), sities in the AGN population has been a ques- 36 months (Ajello et al. 2009), 58 months7, 60 months tion of particular interest in X-ray studies of AGN. (Ajello et al. 2012), and the 70 month catalog source list Studies of the X-ray background (e.g., Gilli et al. is now in preparation. 2007, Brandt & Hasinger 2005, Treister & Urry 2005, Much work has been done on the 9-month BAT cat- Worsley et al. 2005, Gandhi & Fabian 2003) suggest that alog (consisting of 153 sources), such as studies of their the majority of accretion in the Universe must be ob- X-ray properties (Winter et al. 2009, W09 hereafter), op- scured. If we wish to determine the true, intrinsic AGN tical properties (Winter et al. 2010), host-galaxy prop- power output, a knowledge of the amount of absorbing erties (Koss et al. 2011a) and construction the nuclear material is needed. We aim to determine the true dis- AGN spectral energy distributions (Vasudevan et al. tributions of X-ray absorption and emission properties 2009, 2010). The X-ray absorption properties of the using a complete and representative subset of the least- AGN in the 36-month catalog have been presented in biased sample of local AGNs. We use the best-quality Burlon et al. (2011) (199 sources). This work draws X-ray spectral data available: therefore XMM-Newton is from the published 58-month catalog which contains 1092 employed preferentially if archival data are present (sup- sources, of which ∼ 720 are AGN candidates (i.e., have a plemented by 13 new XMM-Newton observations taken counterpart identified as a galaxy, AGN, Seyfert, blazar, specifically for this study); other sources of X-ray data or BL Lac, but confirmed to not have a counterpart are detailed in §2. We determine the column density that is a Galactic black hole binary/neutron star/white and nature of the X-ray absorption. In W09, complex dwarf/pulsar). As the catalog becomes increasingly sen- X-ray absorption was found to be common (≈ 55%) in sitive, the data present a great opportunity to improve the 9-month catalog BAT AGNs; we produce new esti- and refine the conclusions drawn from the previous ver- mates of the covering fraction of such complex absorption sions of the catalog, in particular the detailed X-ray anal- where it is present. Quality measurements of the level ysis of W09. and nature of X-ray absorption are important for de- The numbers of AGN in the BAT catalog make it pro- riving reliable absorption-corrected luminosities over the hibitive to perform pointed observations and analysis for broadest possible X-ray bandpass; the luminosities pre- the X-ray properties for all ∼ 720 AGN. The fraction sented here will therefore be useful in constructing low- of AGN with good (XMM-Newton quality, at & 4000 redshift luminosity functions used to assess the amount counts) 0.4–10 keV data over the whole sky is much of accretion and black-hole growth in the local universe. smaller for the 58-month sources than it is for the 9- Our absorption measurements can also serve as a use- < month sources, requiring a more targeted approach. The ful z ∼ 0.1 “anchor” for assessments of the dependence various studies on properties of BAT AGN mentioned of X-ray absorption upon luminosity, Eddington ratio above have concentrated on subsamples from the BAT (L/LEdd), and redshift. The 0.4–10 keV spectra are catalog, including a recent, very thorough X-ray spec- fit jointly with Swift/BAT spectra from 14–195 keV to tral analysis of 48 Seyfert 1–1.5 AGN in (Winter et al. constrain absorption reliably in heavily obscured AGNs > 23 −2 2012). However, the sources in that paper were selected (NH ∼ 10 cm ), including those with Compton-thick 24 −2 based on optical type and specifically to understand the X-ray absorption (NH > 1.4 × 10 cm ).