Xray Observation of Black Hole Accretion Disks

1
X-ray Observation of Black Hole Accretion Disks

• Ken Ebisawa (NASA/GSFC, USRA)

2
Galactic Black Hole Candidates Spectral States
and Accretion Disk

• High/soft state (?1038erg/s)
• 2-10 keV X-ray spectrum is thermal (ultra-soft),
  power-law hard-tail above 10keV
• Optically thick and geometry thin disk
• Low/hard state(?1037erg/s)
• 2-10 keV X-ray spectrum is power-law
• Optically thin and geometry thick disk
• In this talk, I will concentrate on observation
  of optically thick accretion disk spectra

3
Accretion disk solution

• Abramowicz et al. (1995)

Optically thick ADAF
Standard disk is observationally established Slim
disk seems to be observed in Galactic BHC and ULXs
4
Standard Accretion Disk Model

• Gravitational energy is released as thermal
  emission (Shakura and Sunyaev 1973)
• Comptonization is significant in the inner region
  
• Local emission shows Wien peak with Tcol(r)
• Tcol/Teff 1.7-1.9 independent of radius and disk
  luminosity (Shimura and Takahra 1995 Ross and
  Fabian 1996).
• Standard accretion disk spectrum looks like
  super-position of blackbody spectra
• multi-color disk-blackbody approximation works
  (diskbb in xspec)

5
Optically thick accretion disk in the Soft state

• Theoretical optically thick accretion disk
  spectrum
• (Shimura and Takahara 1995)

Local emission stongly affected by Compton
scattering
Tcol color temperature Teff effective
temperature Tcol 1.9 Teff
6
High state of Galactic Black Hole Candidates

• Inner disk radius is constant at 3 Rs (last
  stable orbit in Schwarzschild metric)
• X-ray observation ? disk size ?Rs2GM/c2
• Black hole mass may be constrained from X-ray
  energy spectral analysis

7
Optically thick accretion disk in the Soft state

• LMC X-3 GINGA
• (Ebisawa et al. 1993)

Intensity vs. Hardness correlation
Optically thick accretion disk Ldisk ? Rin2 Tin4
8
Optically thick accretion disk in the Soft state

• LMC X-3 GINGA
• (Ebisawa et al. 1993)

• Spectral fitting with optically
• thick accretion disk
• with Schwarzschild metric,
• Rin 3 Schwarzschild radii

Distance, inclination fixed, M, dM/dt free
parameters.
M remarkably constant
M6 ((Tcol/Teff)/1.9)2 M?
Consistent with optical results
9
Relativistic effects

• Innermost disk radius Last stable orbit
• 3Rs in Schwarzschild metric
• 0.5 Rs in extreme Kerr case
• Kerr disk can be hotter than Schwarzschild disk
• Doppler effect, gravitational redshift, light
  bending
• Disk spectra calculation
• Schwarzschild disk spectrum is not very different
  from the Newtonian spectrum
• Face-on Kerr disk spectrum is not very difference
  either
• Inclined Kerr disk spectrum is significantly
  hardened
• Hard spectra of microquasars may be explained

10
Inclined Kerr disk is brighter in high energies!
Comparison of Schwarzschild and Kerr disk spectra
cos i
cos i
Kerr disk
Scwarzschild disk
Laor, Netzer and Piran (1990)
11
Ebisawa et al. (2003)

• Relativistic effects and inclination dependence
  of the disk spectra
• When the disk is face-on, relativistic effects do
  not change disk spectra significantly
• Edge-on Kerr disk has very hard spectrum
• Due to doppler boosts in the innermost region

12
Application of Kerr disk spectra to GRO J1655-40

• GRO J1655-40 and GRS191515 (microquasars) are
  highly inclined systems, and known to have
  unusually high disk temperature (e.g., Zhang, Cui
  and Chen 1997)
• ?Kerr disk model suggested instead of
  Schwarzschild disk
• GRO J1655-40 spectral fitting
• i70?, d3.2 kpc, Tcol/Teff1.7 fixed
• Schwarzschild disk model ? M1.8 M? (Ebisawa et
  al. 2003)
• Too small compared to M7 M? from optical
  observation
• Kerr disk model with a0.68 to 0.88 is consistent
  with M7 M? (Gierlinski et al. 2001 Ebisawa et
  al. 2003)
• Inclined Kerr disk model works to solve the too
  hot a disk problem!
• 450 Hz QPO (Strohmayer 2001) also supports
  presence of the disk around a spinning black
  hole (Abramowicz and Kluzniak 2001)

13
Optically thick ADAF disk (slim disk)

• Standard disk breaks when Ldisk gt LEdd
• Energy advection will be dominant
• Slim disk solution
• Slim disk luminosity can exceed LEdd!
• May explain ULXs naturally
• 5 LEdd of 20 M? black hole ?1040 ergs/s

14
Slim disk can exceed LEdd!

• Optically thick and geometrically thick (slim)
  disk

For standard disk h/r0.1, Ldisk lt LEdd For slim
disk h/r 1 ? Slim disk luminosity can be 10
LEdd
15
Characteristics of the Slim disk
Standard disk (radiation dominant, T ? r -0.75)
Temperature profile(Watarai et al.2000)
T increase
Standard disk
Slim disk (optically thick ADAF)
Inner disk radius can be smaller and temperature
higher
Radial dependence of the disk temperature
changes p 0.75 ? 0.50
16
Slim disk in XTEJ1550-56
Kubota and Makishima (2004)
p-free disk model
Multicolor disk blackbody with T(r) r-p
Smaller p gives systematically better fits!
Slim disk is observed in Galactic BHC when
luminosity is extremely high
17
Slim disk in GRS1915105
Yamaoka (2001)
High Tin
T(r)?r -0.5 favored
? slim disk
Standard disk model fit T(r)?r -0.75
Disk oscillation between the standard disk and
slim disk
18
Slim disk in ULX (IC342 Source1)

• Slim disk model fit by Watarai et al. (2001)

• Standard disk fit

• Mass varies
• Unreasonable, bad model!

• Constant mass (23M?), only mass accretion rate
  changes
• Reasonable, better model!

Ebisawa et al. (2003)
19
Caveat in ULX spectral modeling

• Phenomenological standard diskpower-law model
  may mislead to very low disk temperature and
  large disk normalization (radius)
• Unexplained power-law dominates
• intermediate mass black hole accounts for soft
  excess
• Tin lt 0.1 keV, Mgt100 M?
• Slim disk with Comptonization (e.g. Kawaguchi
  2003) can explain intrinsically hard ULX spectra
• Disk component dominates
• power-law tail component hardly required
• Reasonable stellar black hole mass (M lt 40 M? )

20
Standard disk and slim disk comparison for M81 X-9
Application of Kawaguchi (2003) slim disk model
to XMM spectra (Foschini et al. 2004)
Slim disk
Power-law
Standard disk
Power-law
Slim disk explains most of the spectrum,weak
hard-tail required
Unexplanied power-law component is dominant, Tin
0.3 keV, large Rin
M150 M?
M18 M?
21
Standard disk and slim disk comparison for
NGC1313 X-2
Application of Kawaguchi (2003) slim disk model
to XMM spectra (Foschini et al. 2004)
Slim disk
Power-law
Standard disk
Power-law
Tin0.27 keV M120 M?
M4 M?
22
Likely origin of ULXs

• Universal luminosity function of X-ray
  binaries(Grimm et al. 2003)
• Luminosity cut-off 1040 erg/s
• Presumably, ULXs are X-ray binaries with possible
  maximum stellar-size black holes
• 40 M? black holes theoretically possible (Fryer
  1999)
• Such a massive black holes may not exist in our
  Galaxy, but not uncommon in external galaxies

23
Conclusion

• Standard accretion disk model is established for
  Galactic BHC when Ldisk ? LEdd
• Slim disk seems to be observed in Galactic BHC
  and ULXs when Ldisk ? LEdd
• ULXs are presumably slim disk around massive
  stellar black holes (lt 40 M? ) shining at
  super-Eddington luminosities
• Theoretical work on slim disk spectra and
  application to observed spectra urgently needed

Acknowlegement to my colleagues A. Kubota, P.
Zycki, K. Watarai, K.Yamaoka, L. Foschini, and T.
Kawaguchi