XMMNewton spectroscopy of four Xray selected BALQSOs discovered in a survey field

1
XMM-Newton spectroscopy of four X-ray selected
BALQSOs discovered in a survey field
Alexander J. Blustin
STFC Postdoctoral Fellow, UCL Mullard Space
Science Laboratory
In collaboration with Tom Dwelly (Southampton),
Mat Page (UCL-MSSL), Ian McHardy (Southampton),
Derek Moss (Southampton), Nick Seymour (Spitzer
Science Centre)
X-ray surveys, Rodos Island, Greece, 2-6 July 2007
2
Introduction
Most Broad Absorption Line Quasars (BALQSOs) are
extremely X-ray faint, probably due to the
presence of ionised absorption associated with
their high-velocity BAL-producing outflows
(Gallagher et al. 2006). Most are therefore
discovered through optical surveys. Some X-ray
selected BALQSOs do exist, though
How do X-ray absorption properties of X-ray
selected BALQSOs differ from those of optically
selected BALQSOs?
Are the absorbing columns simply lower?
3
Black hole galaxy feedback
Energy output blasts away gas clouds, stopping
star formation
Black hole grows, stars form
Are the BALQSO outflows part of this process?
Silk Rees 1998, Fabian 1999, Granato et al.
2004, Page et al. 2004
How does any trend in mass and energy output
through quasar outflows relate to the history of
star formation?
4
The sources
Four broad absorption line quasars appeared
serendipitously in the 1H survey field. X-ray
spectra of the objects extracted by method used
for sources in 13H field (see Page 2006) Work
presented here to be published by Blustin et al,
in preparation
XMM-Newton X-ray image of the 1H deep field Red
0.2-2 keV Green 2-5 keV Blue 5-10 keV
5
X-ray spectra
z 2.63
z 1.793
309 counts
220 counts
background
z 1.40
z 2.64
240 counts
188 counts
6
Optical spectra
Cha 106
Cha 123
Cha 134
Cha 137
Trump et al. 2006 Absorption Index versus
Balnicity Index (as defined by Weymann et al.
1991)
7
Defining the optical X-ray spectral continuum

• Intrinsic Spectral Energy Distribution (SED)
  based on that of Marconi et al. 2004.
• X-ray powerlaw, G 1.9
• Reflection hump spectrum calculated using refl
  model in SPEX 2.00.11.
• aox for each QSO calculated from unabsorbed 2500
  Å flux using relation of Vignali, Brandt and
  Schneider 2003

• SED used for
• Calculating photoionised absorber models (SPEX
  xabs)
• Continuum for X-ray fits (variable normalisation)

aox -0.11 log Ln(2500 Å) 1.85
8
The dynamics of the outflowing gas revealed by C
IV absorption
Fitting gaussian absorption lines in specfit
(IRAF) gives
z 2.63
z 1.793

• outflow speeds of the multiple dynamic phases in
  each object
• velocity width of the absorption lines (required
  for X-ray absorber fitting)
• relative optical depths of absorption associated
  with different dynamic phases used to set
  ratios of column densities of phases

z 1.40
z 2.64
9
X-ray spectral fits

• Model consists of
• continuum with Galactic neutral absorption
• ionised absorber phases corresponding to
  dynamical phases seen in optical
• NH for different phases linked together by
  optical depth ratio
• ionisation parameter x the same for all phases in
  a source

Three free parameters continuum normalisation,
NH, and x
10
Results the properties of the ionised X-ray
absorbers
X-ray selected BALQSOs
11
Results the properties of the ionised X-ray
absorbers
X-ray selected BALQSOs
Optically selected BALQSOs
PG1115080 (Chartas 2007) Q1246-057 (Grupe
2003) SBS1542541 (Grupe 2003) PG1254047 (Sabra
2001) PG2112059 (Gallagher 2004)
12
Results the properties of the ionised X-ray
absorbers
X-ray selected BALQSOs
Optically selected BALQSOs
Non-BAL QSOs and Seyferts (Blustin et al. 2005
sample)
13
Results mass transport through the outflows
(Blustin et al. 2005 method)
X-ray selected BALQSOs
Non-BAL QSOs and Seyferts (Blustin et al. 2005
sample)
14
Results kinetic energy of the outflows
(Blustin et al. 2005 method)
X-ray selected BALQSOs
Non-BAL QSOs and Seyferts (Blustin et al. 2005
sample)
15
Conclusions
X-ray selected BALQSO ionised absorption
properties? Within our limited samples of X-ray
and optically selected BALQSOs, the X-ray
absorber properties are similar NH and x are
perhaps slightly higher for the XBALQSOs
(counterintuitive), but large error bars on
fitted values. There are significant biases in
this comparison the OBALQSOs are the X-ray
brightest in their class, hence observable, but
this may mean that they have atypical absorption
properties. Better signal-to-noise and more
sources required. Evolution of outflow
mass/energy transfer versus star formation? The
mass outflow rates in our BALQSOs are broadly
similar to those in nearby AGN. There is some
evidence, though, that the kinetic energy output
via the outflows is greater at redshifts
corresponding to the peak of star formation
activity than in nearby AGN. Need X-ray
spectroscopy of more sources to confirm this.
16
Extra material
17
New black hole in star forming clouds
Energy output blasts away gas clouds, stopping
star formation
Black hole grows, stars form
A black hole/galaxy co-evolution scheme
Black hole accretes remaining gas unobscured
Silk Rees 1998, Fabian 1999, Granato et al.
2004, Page et al. 2004
Dormant black hole, fuel used up
18
Assumptions in calculating the mass outflow rate
19
Mass outflow rate of the wind
volume filling factor
Ionising luminosity (1-1000 Ryd)
outflow velocity
solid angle of shell 1.6 sr
1.23 mproton Lion Cv v W
x
20
Volume filling factor of the outflow
(Labs Lscatt) x
Cv
1.23 mproton c Lion v2 W
Kinetic luminosity of the wind
21
(Labs Lscatt)
c
22
Physical properties of X-ray absorbing winds in
nearby AGN

• In our survey of nearby AGN, we found that
• The winds have low volume filling factors (lt10)
• They account for lt1 of the bolometric
  luminosities of the AGN
• The median mass outflow rate for the Seyferts
  (including NLSy1s) in our sample was 0.3 Solar
  masses/year, eight times the median mass
  accretion rate.

• Over its lifetime, a black hole that builds up a
  mass of 107 MSun will have pushed out 108 MSun
  into its host galaxy
• This is of the order of, say, the amount of hot
  ISM in a Seyfert galaxy bulge

23
How much matter, in total, could be ejected from
the nucleus of a galaxy by a BAL-type wind?
Assuming typical BAL properties, and that the BAL
phase lasts for 10 of the AGN lifetime, a crude
re-scaling of the value for winds in nearby AGN
gives 1010 solar masses
So the winds in distant AGN may be of
cosmological importance
24
Properties of winds in more distant AGN

• Principal population that we know about is the
  Broad Absorption Line quasars (BALQSOs redshifts
  of 0.2 - 6)
• Winds moving at much higher speeds than in
  nearby AGN (2000 60000 km/s)
• Much higher absorbing columns than in nearby
  AGN NH 1023 cm-2
• Similar range of ionisation to nearby AGN
• Multiple ionisation and velocity phases

25
Cha 106
Cha 123
log x 2.85 log NH total 23.3
log x 2.82 log NH total 24.0
Cha 134
Cha 137
log x 2.83 log NH total 23.8
log x 3.23 log NH total 24.7