X-ray spectroscopy of bright AGN

1
X-ray spectroscopy of bright AGN
Giorgio Matt Stefano Bianchi
(Dipartimento di Fisica, Università Roma Tre)
2
Bright AGN
With bright AGN I mean sources with enough
photons to allow for a decent spectral analysis
in the 0.2-4 (0.1-6) keV energy range
About 60000 AGN with more than 5000 counts
(0.5-2 keV) will be detected by WFXT
Population studies over very wide ranges of the
main parameters will be possible Optical
coverage (SLOAN, etc) of the same fields will
provide parameters like the mass of the BH
3
Some open problems (which can be addressed by
WFXT)

• The nature of the soft X-ray emission in
• unobscured sources
• The properties of the warm absorber
• beyond the local Universe
• The IT effect at high z (gt1) and L

4
The nature of the soft excess
The soft X-ray excess was first discovered in the
EXOSAT spectrum of Mkn 841 (Arnaud et al. 1985).
The thermal disc interpretation of the soft
excess has problems the derived temperatures are
too high and always the same (T ? M-1/4). WFXT
will stretch considerably the range of masses
probed. Its soft X-ray band will be also very
useful.
Mkn 841 (Petrucci et al. 2007)
5
The warm absorber beyind the LU
Evidence for warm absorption in 50 of
Seyferts. Usually outflowing (v hundreds
km/s). Location, mass and energetics largely
unknown. Dependence of ? on L, Z, BH mass ??? ?
WFXT !!
Blustin et al. 2005
6
The IT effect
The Iwasawa-Taniguchi (a.k.a. X-ray Baldwin)
effect is the anticorrelation between the EW of
the iron line (narrow core) and the X-ray
luminosity (first discovered by Iwasawa
Taniguchi 1993)
7
The IT effect
The IT effect may be due to a decrease with L of
the covering factor of the reflecting matter (a
similar effect has been found by Maiolino et al.
2007 using infrared data).
8
The IT effect
The anti-correlation of the EW with the Eddington
ratio is highly significant
log(EWFe)(1.610.05) (-0.190.05) log(Lbol/Edd)
9
The IT effect
No significant dependence of the iron EW on the
BH mass is instead found
log(EWFe)(1.730.04) (-0.070.04) log(LBH,8)
10
The IT effect
This effect is not well known at high
luminosities and it is probed only in the local
Universe. WFXT can expand enormously the range
of L and of z probed. Of course, the local
Universe is forbidden
11
Waiting for WFXT where we are now
The XMM-Newton Bright Serendipitous Survey (Della
Ceca et al. 2004 Caccianiga et al.
2008) Flux-limited sample (7x10-14 erg/s/cm2
in the 0.5-4.5 keV energy range) 400 sources
(80 AGN)
X-ray selection! Elusive AGN, etc.
Type 2
Type 1
Caccianiga et al. 2007
12
Waiting for WFXT where we are now
CAIXA (Catalogue of AGN In the XMM-Newton
Archive) is a catalogue of pointed XMM
observations of AGN in the public archive up to
March 2007 (Bianchi et al. 2009) To be included
in CAIXA, a source must have at least 200 (0.5-2
keV) 200 (2-10 keV) counts In practice, most
sources have more than a thousand counts. Biased
towards well-known sources. Undefined selection.
13
CAIXA A Catalogue of AGN in the XMM-Newton Archive
156 sources All radio-quiet AGN in targeted
XMM-Newton pn observations gt200 counts in either
of the (rest-frame) bands of 0.5-2 and 2-10
keV lt1 pileup log(R)lt1 (quasar) log(R)lt2.4 and
log(RX)lt-2.755 (Seyfert)? NHlt2x1022 cm-2

• 6 and 20 cm fluxes (100 of the catalog)?
• Hß FWHM (64 of the catalog)? only from BLR
• (lt2000 km/s -gt narrow-line objects)
• BH mass (52 of the mass)?
• Luminosity dependent bolometric correction
  (Marconi et al. 2004)?

Bianchi et al. 2009
14
Sy
QSO
General properties
15
Powerlaw index
The average 2-10 keV spectral index for the whole
catalogue is 1.730.04, with a large spread of s
0.450.03
G steeper in quasars (1.800.05, consistent with
PG quasars Piconcelli et al., 2005) than in
Seyferts (1.660.05). A Kolmogorov-Smirnov test
yields that the two populations are different at
the 99 level The difference is even larger if
we consider narrow-line objects (1.940.07) with
respect to broad-line objects (1.620.04). In
this case, a K-S test gives a NHP of 2x10-5
BLS1
NLS1
It is important to stress here that the photon
indexes in CAIXA, in principle, are not the
intrinsic ones, because we do not include a
Compton reflection It is possible that the
different photon indexes found for different
populations in CAIXA do not only (or at all)
reflect differences in the intrinsic G of the
sources, but also in the amount of Compton
reflection
16
Soft excess
Hß FWHM is correlated with the hard X-ray
luminosity and anti-correlated with the ratio
between the optical V and the X-ray flux Objects
with narrow-lines seems to be X-ray weaker than
broad-line objects
17
Soft excess
123 out of 156 sources need a second and steeper
powerlaw to model the soft X-ray emission
If we replace the second power law with a thermal
model, the soft excess in unobscured radio-quiet
AGN is characterized by a temperature which is
constant across the range of luminosities and BH
masses, as already shown by several authors
(Gierlinski Done 2004 Crummy et al. 2006)
18
Correlations with BH mass
19
The iron lines

• The average EW of the neutral (narrow) Ka iron
  line is around 80 eV, as expected for Type 1
  objects.
• And, as said before, we confirmed the
• IT effect!

We do not find different rates of detection for
ionised iron lines for narrow- and broad-line
objects
20
The iron lines
Both the Fe XXV and Fe XXVI lines are detected in
24 sources Their EWs and the ratios between them
are consistent with a production in distant
matter photoionised by the AGN, with large
ionization parameters and large column densities
(up to 1023 cm-2 for Fe XXV, larger for
FeXXVI) We cannot exclude a contribution from an
ionised accretion disc for the larger EWs
21
Conclusions
Thanks to XMM-Newton we have a decent idea of the
X-ray average properties of (unobscured) AGN in
the Local Universe. WFXT will extend this to
high luminosities and redshifts