Probing Quasar Outflows

1
Stratified Quasar Winds Revealed
Sarah Gallagher (UCLA) January 2007
(Urry Padovani 1995)
2
The Quasar Epoch z2-3
(Richards et al. 2006a Nagamine et al. 2006)
3
Normal Quasar
Ly??CIV ?MgII H?
H??
4
Quasars with Outflows BAL Quasars
voutflow (0.03-0.2)c
Ly?/NV SiIV CIV
Broad Absorption Lines (P Cygni profiles)
5
Infrared Optical-UV X-ray
big blue bump lt days
torus 1-10 pc
corona AUs
NASA/CXC
6
A Model for All Radio-Quiet Quasars
UV emission lines light yr (1017 cm)
X-ray continuum source light hrs (1014-15 cm)
UV/optical continuum source light hrs-days (1016
cm)
(Gallagher et al. 2002a Adapted from Königl
Kartje 1994 Murray et al. 1995)
7
Two Views Through the Wind
X-ray
UV
shielding gas
Broad Absorption Line (BAL) Quasars
BAL Wind
Does it exist?
8
X-ray Properties of Quasars
2 keV
2500 Å
(Laor et al. 1997)

• aox 0.384 log (f2 keV / f2500 Å)

9
UV Luminosity vs. aox
brighter in X-rays
Define Daox aox - aox (Luv)
fainter in X-rays
log(Luv) (ergs s-1 Hz-1)
228 SDSS Quasars with ROSAT (Strateva et al.
2005)
10
Bright Quasar Survey zlt0.5 Sample
BAL Quasars
Daox
X-ray weak (400x)
X-ray normal
55 BQS Quasars with zlt0.5 (Data from
Brandt, Laor, Wills 2000)
11
X-ray Absorption by Neutral Gas
NH
almost optically thick to Compton scattering
(assuming solar metallicity)
12
PG 2112059 (z0.466) First BAL Quasar X-ray
Spectra

• G 1.970.25 NH 1022 cm-2
• (Gallagher et al. 2001)

13
X-ray Spectroscopy of 12 RQ BAL Quasars

• normal underlying X-ray continua
• significant intrinsic absorption
• NH (0.1-5.0) x 1023 cm-2
• from gt5 keV continuum
• normal aox (UV/X-ray flux ratio)
• not just simple absorption
• partial coverage and/or ionized gas
• (e.g., Gallagher et al. 2002b Chartas et al.
  2002, 2003 Aldcroft Green 2003 Grupe et al.
  2003
• Page et al. 2005 Shemmer et al. 2005)

14
PG 2112059 Revisited Dramatic Spectral
Variability

• PG 2112059 ASCA Oct 1999 Chandra Sep 2002
• (Gallagher et al., 2004)

15
PG 2112059 Revisited Comparison of Models
FeKa
(Gallagher et al. 2004)
16
Major Change ? Increased NH


Chandra
ASCA
Partial-covering absorber model
17
Little Variability in UV BALs
18
Chandra BAL Quasar Survey
Large, well-defined sample

• 35 luminous BAL quasars
• z 1.42.9
• MB -26.1 to -28.4
• UV spectra for all (from literature)
• 47 ks exploratory observations
• 35 observed 27 detected (77)

(Gallagher et al. 2006)
Collaborators Niel Brandt, George Chartas,
Gordon Garmire (Penn State), Robert Priddey
(Hertfordshire), Rita Sambruna (Goddard)
19
Exploratory Surveys X-ray Data

• Rough continuum shape GHR
• from hardness ratio (hard-soft)/(hardsoft)
• (analogous to B-V ? spectral index)
• Relative UV-to-X-ray power Daox

20
Spectral Shape vs. X-ray Weakness X-ray Faint ?
X-ray Hard
All RQ BAL quasars show evidence for X-ray
absorption (complex with NH (180)x1022 cm-2)
normal RQ quasars G 2.0 0.25 Daox 0.0
0.15
Softer (less absorbed)
Harder (more absorbed)
covered
Daox
Fainter
Brighter
21
Bright Quasar Survey zlt0.5 Sample
BAL Quasars
Daox
X-ray weak (400x)
X-ray normal
55 BQS Quasars with zlt0.5 (Data from
Brandt, Laor, Wills 2000)
22
BQS Chandra BAL Quasars
Chandra BALQs
Daox
No correlation for BAL quasars.
23
UV SpectraX-ray Bright vs. X-ray Weak
CIV
CIV
SiIV
SiIV
X-ray BRIGHT
X-ray FAINT
24
vmax vs. Daox
te1
Daox
High velocity appears to require large NH.
25
Conclusions I X-ray Observations

• compact thick X-rayonly absorbers
• X-ray UV absorption not consistent
• NH and variability properties
• some may be Compton-thick!
• (te 1 NH1.5x1024 cm2)

(Gallagher et al. 2006)
26
X-ray and UV Continua Emitting Regions Are Not
Cospatial

• Implied by Compton-thick X-ray absorption.
• Consistent with results from gravitational
  micro-lensing.
• (Kochanek et al. 2006)

27
Conclusions I X-ray Observations

• compact thick X-rayonly absorbers
• X-ray UV absorption not consistent
• some may be Compton-thick!
• (te 1 NH1.5x1024 cm2)
• likely correlation of vmax Daox
• first UV/X-ray correlation found
• ? supports radiative driving of UV outflows

(Gallagher et al. 2006)
28
Link Between Shielding Gas and vmax
thicker shield ? more X-ray weak
thinner shield ? less X-ray weak
larger Rlaunch ? lower vterm
smaller Rlaunch ? higher vterm
(Gallagher et al. 2006 Gallagher Everett 2007)
vterm (GMBH/Rlaunch)1/2
(cf. Chelouche Netzer 2000 Everett 2005)
29
IR Opt-UV
X-ray
big blue bump lt1 pc
torus 1-10 pc
corona AUs
disk wind from here
NASA/CXC
30
The Dusty Outflow
The inner radius of the dusty outflow is set by
the temperature at which dust sublimates
r1.3(Luv/1046)½(T/1500)-2.8 pc
(Adapted from Königl Kartje 1994)
31
Sample SEDs of Non-BAL Quasars
mid-infrared
optical
8 ?m
5000 Å
spectral indices (L?? ??) ??,ir ??,opt
32
IR-Luminous Quasars Typically Have Steeper
Spectral Slopes
(Gallagher et al., submitted)
33
IR-Luminous Quasars Show Spectral Curvature
34
Sample Hot Bump
(Rodríguez-Ardila Mazzalay 2006)
35
Changing Shape of Dusty Outflow?
Less Luminous
More Luminous
36
BAL Quasar IRS Spectra
silicates
(Gallagher et al. in prep see also Shi et al.
2006)
37
Clouds and a Torus
(Urry Padovani 1995)
38
Multiwavelength Synthesis The Stratified Wind
Picture
39
Physical Parameters of the Stratified Wind
40
Stratified Wind Driving Mechanisms

• UV BAL wind
• Radiation pressure on UV resonance lines
• Dusty Outflow
• Radiation pressure on dust
• Shielding Gas
• Continuum radiation pressure
• Magneto-hydrodynamic driving?

41
Future Work

• 1. Accepted Spitzer Cycle 3 theory project.
• Illuminating the Dusty Wind 3D Modeling of
  Quasar Silicate Emission
• Collaborators John Everett (Wisconsin) Dean
  Hines (SSC)
• 2. Chandra follow-up of vmax vs. ??ox
  correlation.
• 3. Suzaku (10-30 keV) follow-up of nearby
  Compton-thick BAL quasars.
• 4. Long term Constellation-X spectroscopy of
  X-ray outflows in absorption.

42
Resolving the Shielding Gas with Constellation-X
Model Two component outflow with different
ionization states, velocities, and velocity
dispersions. (Flux matches BAL quasar
PG1115080.)