Menu

Research topics

AGN Hosts

Compton-thick AGN & XRB

Normal Galaxies

LSS & Cosmology

 

 

Normal Galaxies

The group has been leading efforts to explore the X-ray evolution of normal galaxies, i.e. systems with X-ray emission dominated by stellar processes (e.g. X-ray binaries) and hot diffuse gas, rather than accretion on a central Super-Massive Black Hole (i.e. AGN). Selecting normal galaxies in X-ray surveys is not easy, as the dominant population at these wavelengths are AGN. The group has been exploring the efficiency and limitations of different methods for finding normal galaxies (e.g. Georgakakis et al. 2004; Tzanavaris et al. 2006; Georgakakis 2008). We have recently developed a methodology which uses infrared data to provide nearly unbiased samples of X-ray selected normal galaxies with minimal AGN contamination ( Georgakakis et al. 2007). Highlights from our work include the first estimate of the X-ray Luminisity Function (XLF) of normal galaxies at low redshift z ~0.1 (see Figure; Georgantopoulos et al. 2005; Georgakakis et al. 2006) and the determination of the evolution of (i) the determination of the X-ray evolution of both passive ans star-forming galaxies out to z ~1 both by modelling the observed X-ray number counts (see Figure; Georgakakis et al. 2007).and by estimating the XLF at different redshift intervals ( Tzanavaris et al. 2007).     

Left The X-ray luminosity function of late-type galaxies at z~0.1. Right: X-ray source counts for star-forming galaxies Comparison with models that assume different rates of evolution for the X-ray luminosity function on the left are also plotted. These suggest X-ray evolution for the star-forming galaxy population of the form ~(1+z)2.4.

Properties of AGN Host Galaxies

The group studies the properties of galaxies that host AGN by combining information from different wavelengths, X-rays, optical, infrared and radio. The star-formation history, the morphology and environment of AGN hosts are all powerful diagnostics of the physical processes that are responsible for triggering the accretion on the Supermassive Black Hole (SBH) at the centres of galaxies. We are using observations for large multi-wavelength programs at z~1, such as the AEGIS and GOODS.

Highlights from our work include (i) evidence that the X-ray AGN population is associated with galaxies with intermediate rest-frame colours, in between the blue and the red clouds of the Colour-Magnitude Diagram (see Figure; ; Georgakakis et al. 2008), while AGN/starburst composites are rare at these redshifts ( Georgakakis et al. 2004; ), (ii) the first estimate of the fraction of AGN in groups at ~1, which suggests that these systems are often found in moderate density environments where massive bulges (hosting massive black holes) are more often reside ( Georgakakis et al. 2008), (iii) evidence that heavily obscured, possibly Compton Thick, AGN at z~2 are in ULIRGs, possibly undergoing powerful starbursts (Georgantopoulos et al. 2008.

 

Left: Colour Magnitude Diagram (CDM) of X-ray AGN (red) and optically selected galaxies (blue) at z~1 in the AEGIS survey. The line separates the blue (star-forming) from the red (quiescent) clouds. X-ray AGN are found in between the blue and the red clouds. This suggests that the star-formation of AGN hosts has been recently terminated and are on the move from blue star-forming systems to red and dead galaxies.

Right: Optical spectrum of one of the X-ray AGN in AEGIS. The deep Balmer absorption lines suggest recently terminated star-formation (post-starburst spectrum). At z ~1 we find evidence that X-ray AGN are preferentially associated with post-starbursts ( Georgakakis et al. 2008).

LSS & Cosmology

The group has a continuing interest in the study of the large-scale structure of the Universe using different mass-tracers (clusters and groups of galaxies and AGN) and halo-catalogues from large numerical simulations as well as in inferring the cosmological parameters by combining the clustering approach with the Hubble relation as traced by SN Ia’s. In more detail our interests are directed mainly towards: (a) Identifying the true shape and its relation to the internal dynamics of cosmic structures as well as the possible environmental dependences of halo shape and dynamics (eg., Plionis et al. 2003; Plionis, Basilakos & Tovmassian 2004; Plionis, Basilakos & Ragone-Figueroa 2006; Basilakos et al 2007; Ragone-Figueroa & Plionis 2006), (b) Detecting the clustering pattern of cosmic structures and its cosmological implications (eg. Plionis & Basilakos 2002a, 2002b Basilakos et al. 2004; Basilakos et al. 2005; Plionis et al. 2008). Recently, studying high-z X-ray (XMM & Chandra) selected AGN, we have established a flux-limit dependent clustering (which corresponds mainly to a luminosity dependent clustering), which for the first time provides an explanation of the conflicting results of recent surveys (Plionis et al. 2008). Furthermore, we have shown the cosmological importance of AGN high-z clustering in constraining cosmological parameters and investigating the evolution of the X-ray AGN bias (eg. Basilakos & Plionis 2005; Basilakos & Plionis 2006; Basilakos, Plionis, Ragone-Figueroa 2008).

 

References: Basilakos, S., et al., 2004, ApJ, 607, L79;   Basilakos, S., et al., 2005, MNRAS, 356, 183; Basilakos, S. & Plionis, M., 2005, MNRAS, 360, L35; Basilakos, S. & Plionis, M., 2006, ApJ, 650, L1; Basilakos et al. 2007, MNRAS, 365, 539; Basilakos, S., Plionis, M., Ragone-Figueroa, C., 2008, ApJ, 678, 627; Plionis & Basilakos 2002a, MNRAS, 329, L47; Plionis & Basilakos, 2002b, MNRAS, 330, 399; Plionis et al 2003, ApJ, 594, 144; Plionis, Basilakos & Tovmassian 2004, MNRAS, 352, 1323; Plionis, M. et al. 2008, ApJ, 674, L5

 

 

 Left:  The angular correlation length of X-ray selected AGNs as a function of survey flux-limit: Red and black points correspond to the CDFS and CDFN results. The other survey results are indicated by different symbols (from Plionis et al. 2008).

Right: Present constraints on the Dark Energy Equation-of-State parameter, w, and the cosmic matter density obtained from combining the high-z X-ray AGN clustering results of Basilakos et al. (2005) and the SNIa Hubble relation based on the recent Essence compilation (black contours) (from Basilakos & Plionis 2006)

 

Compton-thick AGN and the XRB

Compton thick are named the deeply obscured AGN having column densities larger than 1024 cm-2.  This population  is difficult to be detected even with in the 2-10 keV surveys of XMM and Chandra, as even these hard X-rays get easily absorbed by the large amounts of gas surrounding the nucleus.    We attempt to identify  this elusive Compton thick AGN population using direct X-ray spectroscopy techniques in the Chandra deep fields (eg georgantopoulos et al. 1997, Georgantopoulos et al. in prep.). The spectral signatures of Compton thick AGN are a flat ‘reflection’ spectrum at rest-frame energies below 10 keV, a large equivalent width FeKα line at 6.4 keV, as well as an absorption turnover at  high energies (>10-20 keV). At large redshifts the K-correction moves the absorption turnover at soft energies within the   Chandra bandpass making their identification easy. At smaller redshifts we are identifying the Compton thick AGN by looking for sources with a distinctively flat spectrum. We find 9 candidate Compton thick sources in the CDF-N down to a flux of 10-15 erg cm-2 s-1. Four of these coincide with the sub-mm galaxies (see Alexander et al. 2005).  Fig. 1 shows the Chandra X-ray spectrum of one such candidate Compton thick source. However, according to the X-ray background (XRB) population synthesis models the bulk of the Compton thick AGN reside at fluxes fainter than the flux limits of the Chandra deep fields. The only way to identify these sources is through IR techniques. In Georgantopoulos et al. (2008) we are comparing several mid-IR  or a combination of   mid-IR and optical selection techniques in the CDFN. The most promising technique (Fiore et al. 2008) involves the selection of faint F24μm/FR > 1000 sources with red colours (R-[3.6] > 3.7). X-ray stacking analysis of these sources shows a flat spectrum with a photon index Γ~1. Τhese sources which reside at high redshifts (z~2)   could be associated with either Compton thick AGN or alternatively moderately obscured (NH~ several times 1022 cm-2) low-luminosity AGN.