Luminous obscured quasars in the HELLAS2XMM survey: the Spitzer perspective

1
Luminous obscured quasars in the HELLAS2XMM
survey the Spitzer perspective

• Cristian Vignali
• Dipartimento di Astronomia, Universitadegli
  Studi di Bologna
• F. Pozzi, A. Comastri, C. Gruppioni,
• L. Pozzetti, M. Mignoli, C. Lari, G. Zamorani, M.
  Brusa
• The HELLAS2XMM collaboration

2
The culprit
Who inspired this work
Elvis et al. 1994
3
and some recent results
Richards et al. 2006
4
Talk outline

• The importance of obscured AGN the X-ray view
• Why Spitzer?
• Sample selection luminous, obscured quasars in
  the HELLAS2XMM survey and KS-band morphology of
  their hosts
• A quick look at the Spitzer IRAC and MIPS data
• Broad-band spectral energy distributions using
  IRACMIPS24 data first results
• Physical properties of the X-ray selected
  obscured quasar population from HELLAS2XMM
  bolometric corrections, BH masses, and Eddington
  ratios

Mission impossible all in 10 minutes!
5
The importance of obscured AGN the X-ray view

• A large fraction of the accretion-driven energy
  density in the Universe is obscured by dust and
  gas
• (e.g., Fabian 1999)
• Ingredient of the X-ray background synthesis
  models
• (e.g., Comastri et al. 1995 Gilli, Comastri
  Hasinger 2007)
• ? the high-luminosity, obscured AGN population
  (Type 2 quasars), expected on the basis of
  Unification schemes, only partially found
• Obscured phase of the AGN growth linked to the
  host
• galaxy formation?

6
Why Spitzer?

• The nuclear emission is re-processed in the IR
  band (by the dusty torus envisaged by the
  model?), less affected by extinction
• Spitzer provides coverage over the 3.6-24 µm
  wavelength range (and also at 70 and 160 µm is
  some cases)
• IR emission important to compute the resulting
  bolometric luminosity of obscured, luminous AGN
  through their SEDs

REQUIRED removal of the galaxy contribution
7
Sample selection obscured quasarsfrom the
HELLAS2XMM survey
Mignoli et al. 2004

• SAMPLE HELLAS2XMM
• F2-10 keV 10-14 erg cm-2 s-1 over 1.4 deg2
• Optically faint (R24) sources with
• no optical identification
• ?high X-ray-to-optical flux ratio
• (X/O10) sources with indications
• of X-ray obscuration

• ALL bright in the KS band
• All have R-KS5 ? Extremely Red
• Objects (EROs), some are extreme
• Most have elliptical profiles, two are
• point-like (Mignoli et al. 2004)

ISAAC KS-band follow-up
8
High X/O population unveiled also by Chandra
10-20 of the sources detected in the X-ray
surveys have X/O1
Log (X/O)1
e.g., SEXSI survey Eckart et al. 2006
9
similarly, in the COSMOS survey
acs
mips 24
Lx1045 z1.59
Point-like steep PL SED
acs
mips 24
Lx1044.7 z1.2
Extended galaxy-dominated SED
(Civano et al., in prep.)
a significant fraction of EROs is associated with
X/O10 sources EROs have larger X/O than
broad-line AGN
10
What about the source redshifts?
Estimated redshift range 0.8photometry, Mignoli et al. 04) 6/8
sources 2/8 sources secure spectroscopic
redshifts from ISAAC (Maiolino et al. 06)
J H KS
J H KS
Maiolino et al. 2006
LINER (X/O52)
QSO 1.9 (X/O78)
NH1023 cm-2 L2-10 keV1044 erg/s ? Type 2 QSOs
11
KS-band images with 24 µm contours
40 arcsec
All sources detected in IRAC MIPS24 S(24
µm)150-1000 µJy
12
Spectral Energy Distribution the modeling (I)
Bulge-dominated sources
Host galaxy Elliptical galaxy
normalized to Ks Torus emission Template
from Silva et al. 04
Lbolf(LX NH) every source model normalized
using its X-ray luminosity and NH values
Silva et al. 2004
13
Bulge-dominated sources (in KS) nuclear vs.
host-galaxy emission

• Elliptical galaxy
• nuclear component
• match the data within a factor of 2-3
• Nuclear comp.
• consistent with KS
• upper limits
• Nucleus starts dominating at 6 µm

from Pozzi et al., WORK IN PROGRESS
NUCLEAR COMPONENT not a SED fitting! ?
normalized to X-ray observables LX and NH
14
Spectral Energy Distribution the modeling (II)
Point-like sources (in the KS band)
Extincted Type 1 quasar or red quasar template
(from Polletta et al. 2006) normalized to 24 µm

• Extincted Type1 quasar NOT able to reproduce the
  data ?sharp decrease in the optical
• Good match using a red QSO template

from Pozzi et al., WORK IN PROGRESS
15
Bolometric correction and luminosity
Bolometric luminosity LBOL kX ? L2-10 keV
bolometric correction (35 in Type 1 QSOs from
Elvis 94) BUT large dispersion in the broad-line
QSO SEDs
our approach
16

• Corrections applied to the obtained face values
  to account for
• covering factor ? opening angle of the torus ?
  N(AGN2)/N(AGN1) in this LX range from Gilli 07
  XRB models ? factor 1.5
• anisotropy ? only part of the
• re-emission from the torus reaches the observer ?
  10 according to Silva 2004 templates

from Urry Padovani 1995
17
Bolometric correction preliminary results

• Median kX LBOL/L2-10 keV 27

Red QSO at z2.08
point-like
bulge-dominated
18
Bulge and black hole masses
Bulge-dominated sources

• KS-band flux mostly from the host galaxy ? LK ?
  MBH using the Marconi Hunt (2003) local
  relation and assuming (MBH/M)z1 2?(MBH/M)z0
  (Peng 06)

109 M?
Eddington ratioLBOL/LEDD 0.06
19
BH masses comparison with SDSS quasars

• Indications that
• these luminous obscured QSOs, residing in
  KS-luminous galaxies, have massive black holes

SDSS data from McLure Dunlop 04
20
Eddington ratios comparison with SDSS QSOs

• Indications that
• these very massive, X-ray luminous black holes
  at
• z1-2 have already passed their rapidly accreting
  phase
• they have reached their final masses with low
  accretion rates

SDSS data from McLure Dunlop 04
21
if you have questions
Im not sure I can answer
THE END
22
Pks 0537_54 (Ell) R-K6.2 zph1.30
Abell2690_75 (Ell) R-K6.3 zph1.30
Pks 0312_36 (Ell) R-K5.5 zph1.05
Pks 0537_91 (Ell) R-K5.1 zph0.90
smoothed KS-band images
Pks 0312_45 (P) R-K5.7 zph1.65
BPM16274_69 (Ell) R-K6.6 zsp1.35
Pks 0537_111 (Ell) R-K6.8 zph1.45
Abell2690_29 (P) R-K7.4 zsp2.08
23
Spitzer observations

Tint Flux limit
(5s) ---------------------------------------------
--------------------- IRAC (3.6/4.5/5.8/8.0 ?m
) 480 s ?2.0 ?Jy (3.6 µm) MIPS 24 ?m
1400 s ?100 ?Jy
---------------------------------------------
24
24 micron data cleaning
25
X-ray vs. 24 µm flux
26
X-ray vs. 24 µm flux the X/O10 samples
27
Bulge-dominated sources (in KS)

• Elliptical galaxy
• nuclear component
• match the data within a factor of 2
• Nuclear comp.
• consistent with KS
• upper limits
• Nucleus starts dominating at 8.5 µm

NUCLEAR COMPONENT not a SED fitting! ?
normalized to X-ray observables LX and NH
28
Bulge and black hole masses
Bulge-dominated sources

• KS-band flux mostly from the host galaxy
• ? estimate of the stellar mass Mbulge assuming
• Mstar/LK0.3-1 (valid for old stellar
  populations at z0.9-2 and t1-6 Gyr
  Bruzual Charlot 03)

massive host galaxies Mstar 1011_1012 M?
109 M?
assuming the local LK-MBH relation (Marconi
Hunt 03 see also Peng et al. 06)
29
Summary

• We have observed with Spitzer (3.6-24?m) a sample
  of obscured X-ray selected AGNs. For almost all
  of them the optical NIR emission is dominated
  by the host galaxy
• All the sources have been detected in IRAC
  MIPS24
• dusty torus revealed?
• Yes SEDs well reproduced by 2 components host
  elliptical
• galaxy dusty torus emission (normalized to
  X-ray luminosity)
• Bolometric luminosity Black Hole masses
  derived.
• ? Eddington ratios.
• "Fading" very massive Black
  Holes (108-109 M?)
• at high-z post rapidly accreting phase
• (Lbol/Ledd 0.01-0.03)