X-ray Absorbing Outflows

1
X-ray Absorbing Outflows

• Astro 597 High Energy Astrophysics
• September 27, 2004
• Brendan Miller

2
The Big Picture

• Outflows carry off mass, energy, and angular
  momentum from the accretion disk
• Feedback could regulate growth of black hole and
  host galaxy, even pollute IGM
• Observed as blueshifted absorption lines (mostly
  in UV spectra)
• Probably all AGN have associated absorption,
  although the details vary by class and object

3
Where are we in the AGN zoo?

• Radio loud quasars UV absorption blueshifted by
  up to about 5000 km/s
• Radio quiet quasars 10 show broad absorption
  lines (max shift to about 0.2c)
• Seyfert 1 50 show narrower absorption features
  to about 1000 km/s

4
Why outflows are necessary (Arav 2003)

• Clouds require turbulence (thermal not broad
  enough)
• Clouds should have velocity independent
  absorption profiles
• Clouds dont obviously explain detached troughs
  (shifted absorption features)

5
Numbers

• To be self-consistent, Ill just use results from
  hydrodynamical modeling done by Proga, Stone, and
  Kallman (2000)
• For a central black hole accreting
  at a rate of , the wind launches
  from a radius of 1e16 cm and is accelerated by
  UV line radiation to a speed of 15,000 km/s at a
  distance of 1e17 cm. Mass loss from outflow is
  Column densities are a few times 1e23 and the
  covering factor is about 0.2.

6
Pretty pictures
7
Ionization in the outflows

• Nasty equations from Krolik reveal that
  ionization parameter is very large
• Need to prevent gas from becoming completely
  ionized (no line pressure)
• Murray et al (1995) postulate hitchhiking gas
  interior to wind to shield outflow
• Nasty equations from Murray describe properties
  of winds

8
Geometry of wind for BALQSOs
Gallagher 2002

• Outflow streams off accretion disk, driven by
  radiative UV line pressure
• Shielding gas protects the wind from being
  completely ionized

9
Model can explain

• The 10 of RQQ showing broad absorption lines is
  due to likelihood of viewing source through the
  wind
• This is supported by polarization fraction
  increasing in absorption troughs (photons can
  detour around outflow)
• The lack of RLQ with broad lines may be due to
  complete ionization of gas in the inner disk
  forcing outflow to launch further out

10
No RLQ with BAL
Non BAL quasars
BALQSOs
11
Velocity of the outflow

• The velocity of the outflow increases with radial
  distance
• Dashed line is approximate analytical function,
  top solid line is numerically integrated radial
  velocity, and bottom solid line is vertical speed
• See Murray 1995 for details

12
We Report, You DecideA Fair and Balanced Look
at Krolik

• UV absorption suggests column density of H atoms
  of about 1e20, or a density of 100-1000.

13
Force from a line

• The acceleration due to radiation line pressure
  can be evaluated by solving radiative transfer
  equation (looks like plane-parallel form because
  geometry is cylindrical, integrated over
  azimuthal angle)

14
Sobolev approximation

• Photons can interact with resonant line only at a
  localized region, since Doppler shifts from
  changing angle of outflow to ray and increasing
  speed with distance make the optical depth at
  other regions negligible

15
Punchline

• It is then trivially left as an exercise to the
  audience to show that the solution is given by

16
My head hurts

• Write total radiative acceleration as a force
  multiplier times the ordinary radiation
  acceleration and get equation of motion

17
Is there a point somewhere?

• Integrate equation of motion, pretending force
  multiplier is constant, get something pretty
  close to these expressions (Laor Brandt, 2002)
  and discover that to get a high outflow velocity
  you should start closer in to disk

18
Absorption line profile

• Optical depth decreases as velocity increases
• Also, ionization fraction decreases
• Combined effect is that absorption is strongest
  at lower velocity

Murray 1995
19
Discussion of Gallagher paper
20
Soft X-ray absorption means

• ?ox becomes more negative (steeper power law)
  as absorption depresses X-rays
• The hardness ratio (H-S)/(HS) increases as
  the soft X-rays are absorbed much more than the
  hard X-rays

21
Correlation with UV?

• Vmin describes angle at which you view wind
  relative to disk, since streams turn over
• Would expect absorption to increase (?ox becomes
  more negative) as Vmin decreases (line of sight
  closer to along disk)

22
On the other hand

• Low ionization quasars showing Mg II absorption
  have greater X-ray absorption
• This is because (discuss)

23
Nice job everyone

• The low ionization-state Mg II forms further out
  in the disk than the C IV, leading to a smaller
  covering factor for the Mg II wind
• So can see C IV without Mg II but if you see Mg
  II, must be looking fairly close to along disk
  (and youll definitely see C IV)
• Which means X-ray absorption should also be
  higher (although some might occur out beyond
  disk)
• Example of how the presence of UV absorbers and
  the presence of X-ray absorbers are linked

24
Complications

• PG 2112059 X-ray absorption variability not
  echoed in UV need different absorbers
• APM 082795255 Fe absorption features at speeds
  of 0.2 and 0.4c suggest very close launching
  point (Chartas 2002)
• Is highly ionized X-ray absorption coming from
  shielding gas? Shielding gas is expected to fall
  back in

25
Brandt paper X-ray spectroscopy

• NGC 3783 Nearby Seyfert 1
• Brandt Kaspi 10.4 day Chandra HETGS
  observation
• Resolve blueshifted absorption features

26
Absorption in AGN

• Equivalent width of C IV increases as ?ox
  decreases
• Correlation between presence of UV and X-ray
  absorption

Laor and Brandt, 2002
27
X-ray spectrum of NGC 3783
28
Multiple component outflows

• Combined line spectra show two absorption
  components in O VII in NGC 3783
• O VII and Ne X have different kinematic structure
• Purple lines show UV
• Hard to measure blueshift accurately

29
Outflow velocities for ions
30
Behind the scenes

• Kaastra et al (2002) required 3 distinct
  components, each with a different ionization
  phase, to fit X-ray spectrum
• Not immediately clear how the ionization
  components correlate to the UV velocity
  components
• Inclination of 45 deg

31
Whats going on?

• Maybe lower ionization states are more responsive
  to driving radiation (more lines, lower
  transition energies)
• Maybe line of sight coincides with direction of
  low ionization outflow
• (from class Ken suggests recombination becomes
  important makes sense)

Note that these lines could very well be clumps
or clouds, possibly the same material that
scatters radiation into line of sight in Seyfert
2 galaxies
32
Lack of X-ray resolution
33
Conclusions

• The presence of UV absorption and X-ray
  absorption is clearly linked
• UV and X-ray absorbers are not necessarily the
  same even in X-rays, require multiple ionization
  components to model absorption
• In BALQSOs, X-ray absorption associated with
  inner regions of disk can act to shield UV wind.
  Lack of correlation with UV properties, differing
  variability, and fast iron lines indicate that
  X-ray and UV absorption is probably from distinct
  components here as well