High-ionization Fe K emission from luminous infrared galaxies

While far-infrared (FIR) luminosity 6 is often assumed to be a good indicator of the star formation rate (SFR) in extragalactic objects (e.g. Young & Scoville 1991, Kennicutt 1998), it has also been argued that, in the absence of active galactic nuclei (AGN), X-ray emission could trace the SFR as well; this latter assumption is based primarily on the observed correlation between X-ray luminosity (usually in the 2-10 keV energy range) and FIR luminosity initially found for nearby actively star-forming galaxies with SFRs ∼3-30 M ⊙ yr -1 (e.g., Ranalli et al. 2003, Grimm et al. 2003;Gilfanov et al. 2004). This relation appears to extend to objects with higher SFR (up to 100 M ⊙ yr -1 ) and higher redshift (e.g., Hornschemeier et al. 2005;Persic & Raphaeli 2007, Lehmer et al. 2008). The hard X-ray emission is generally assumed to have a power-law form and to be predominantly due to the collective emission of high-mass X-ray binaries (HMXBs), which are accreting compact objects (neutron stars or black holes) formed following the death of short-lived massive stars, and therefore expected to have a close relationship to the starburst.

There have been, however, indications that objects with very high computed SFRs, such as ultraluminous infrared galaxies (ULIRGs), might depart from the correlation because they are underluminous in X-rays for their computed SFRs (Persic & Raphaeli 2007, Barger, Cowie & Wang 2007). Arp 220, the nearest ULIRG, is a prime example, with a 2-10 keV luminosity ∼ 1 dex below the general L X -L FIR correlation (see Iwasawa et al. 2005).

A further puzzle comes from its X-ray spectrum, which shows a strong high-ionization Fe K line (mainly Fe xxv, Iwasawa et al. 2005). This latter finding means that Arp 220 is not only X-ray under-luminous, but that the hard X-ray emission is not primarily due to X-ray binaries because of the presence of a high-ionization Fe K line. In nearby star-forming galaxies, Fe xxv is much weaker (e.g., Cappi et al 1999) or undetected, which is consistent that the hard X-ray emission is dominated by HMXBs. These peculiar properties of the nearest ULIRGs clearly warrant further investigation with a larger sample of (U)LIRGs.

The Great Observatory All-sky LIRGs Survey (GOALS) 7 is a multi-wavelength study of the most luminous infrared-selected galaxies in the local Universe, selected from the flux-limited IRAS Revised Bright Galaxy Sample (RBGS: Sanders et al. 2003). An overview of GOALS is given in Armus et al. (2009). C-GOALS (PI: D. B. Sanders) is the X-ray component of the project utilizing data from the Chandra X-ray Observaotry (Chandra, hereafter). Details of the X-ray observations are described in Iwasawa et al. in prep. Here we report results focusing primarily on the L X -L FIR relation and the spectral properties of the hard X-ray emission, with special attention to the Fe K band.

The cosmology used to calculate luminosities in this paper is H 0 = 70 km s -1 Mpc -1 , Ω M = 0.3, Ω Λ = 0.7, based on the latest WMAP results (Hinshaw et al. 2009).

The current C-GOALS sample is complete down to log (L IR /L ⊙ ) = 11.73, and consists of 44 galaxies from the RBGS with redshifts z = 0.010 -0.088 (see Table 1). We first removed obvious AGN as follows. The primary criterion was a flat X-ray spectrum, assessed by the X-ray color or hardness ratio (HR). The X-ray color is defined as HR = (H -S)/(H + S), where H is the 2-8 keV counts and S is the 0.5-2 keV counts. Objects with HR > -0.3 are classified as an AGN. This threshold is chosen because ULIRGs known to host AGN (Mrk 231, Mrk 273, UGC 5101) cluster just above this value. All of the optically identified AGN are selected by this criterion. However, Compton-thick AGN are generally missed by this criterion because of their weakness in the hard band. Therefore, objects that show a strong Fe K line at 6.4 keV, a characteristic signature of a Comptonthick AGN, are also classified as AGN (NGC 6240, NGC 3690 West, VV 340a). These criteria classify 15 objects as AGN, and they are excluded from further discussion leaving an “hard X-ray quiet” (HXQ, as defined by their small HR) sample of 30 objects (including NGC 3690 East). The SFR of these 30 HXQ galaxies, calculated assuming that their FIR luminosity is due to dust heated by star formation alone, ranges from 60 to 300 M ⊙ yr -1 .

The sensitivity of Chandra declines steeply above 7 keV. The 2-10 keV luminosity is estimated by extrapolating the spectral model that describes the data up to 7 keV. When multiple hard X-ray sources are present in a single object, those which have no Spitzer-MIPS 24 µm counterpart are excluded for calculating the 2-10 keV luminosity, assuming they have no relation with the IRAS measured luminosity. In NGC 3690, the western and eastern galaxies are treated separately due to the difference in classification, and their FIR luminosity ratio is assumed to be 1:2 based on the 38 µm study (Charmandaris et al. 2002). The X-ray luminosity is as obser

…(Full text truncated)…

This content is AI-processed based on ArXiv data.