Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
The Broad Fe Kα Linein
MCG−6-30-15
Jörn Wilms (U Warwick/IAA Tübingen)http://astro.uni-tuebingen.de/~wilms
in collaboration with
R. Staubert
C.S. Reynolds (UMD), M.C. Begelman (JILA), E. Kendziorra (IAAT),
J. Reeves (GSFC), S. Molendi (Milano)
Contents 1
Structure
• Kα Line Diagnostics
– Accretion geometry
– Compton reflection and Fe Kα generation
– Line transfer close to the BH
• MCG−6-30-15
– Simultaneous XMM-Newton/RXTE Observations
– Soft X-ray spectrum
– Observations of a Broad Line
– Interpretation and Caveats
• Summary
Testing Relativity in AGN 1
Kα Line Diagnostics
Hot Corona
Thin Accretion Disk
BLR
BH
RS
AU
10 100Energy [keV]
1
10
100
Flu
x [a
rbitr
ary
units
]
Γ=1.9
AGN X-Ray Spectrum:
• Comptonization of soft X-rays from
accretion disk in hot corona (T ∼ 108 K):
power law continuum.
Testing Relativity in AGN 2
Kα Line Diagnostics
Hot Corona
Thin Accretion Disk
BLR
BH
RS
AU
10 100Energy [keV]
1
10
100
Flu
x [a
rbitr
ary
units
]
Γ=1.9
AGN X-Ray Spectrum:
• Comptonization of soft X-rays from
accretion disk in hot corona (T ∼ 108 K):
power law continuum.
• Thomson scattering of power law photons in
disk: Compton Reflection Hump
Testing Relativity in AGN 3
Kα Line Diagnostics
Hot Corona
Thin Accretion Disk
BLR
BH
RS
AU
10 100Energy [keV]
1
10
Fe K
Fe K
100
Flu
x [a
rbitr
ary
units
]
Γ=1.9
α
β
AGN X-Ray Spectrum:
• Comptonization of soft X-rays from
accretion disk in hot corona (T ∼ 108 K):
power law continuum.
• Thomson scattering of power law photons in
disk: Compton Reflection Hump
• Photoabsorption of power law photons in
disk: fluorescent Fe Kα Line at ∼6.4 keV
Testing Relativity in AGN 4
Kα Line Diagnostics
∆Φ
0
2
4
6
8
I νo [
arb
itra
ry u
nits]
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0Energy [keV]
−0.15−0.05 −0.100.000.100.200.30
z=(E e/Eo) − 1
Total observed line profile affected by
• grav. Redshift
Testing Relativity in AGN 5
Kα Line Diagnostics
0
2
4
6
8
I νo [
arb
itra
ry u
nits]
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0Energy [keV]
−0.15−0.05 −0.100.000.100.200.30
z=(E e/Eo) − 1
5o
10o
20o
30o
40o
50o
60o
70o
80o
Total observed line profile affected by
• grav. Redshift
• Light bending
• rel. Doppler shift
Testing Relativity in AGN 6
Kα Line Diagnostics
Line
Em
issi
vity constant
power law
0.0
0.2
0.4
0.6
0.8
1.0
I νo [a
rbitra
ry u
nits]
4.5 5.0 5.5 6.0 6.5 7.0 7.5Energy [keV]
−0.15−0.05 −0.100.000.100.200.30
z=(E e/Eo) − 1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
io= 40o
Total observed line profile affected by
• grav. Redshift
• Light bending
• rel. Doppler shift
• emissivity profile
Testing Relativity in AGN 7
Kα Line Diagnostics
rotating Black Holeve
loci
ty
velo
city
nonrotating Black Hole
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
I νo [a
rbitra
ry u
nits]
4.5 5.0 5.5 6.0 6.5 7.0 7.5Energy [keV]
−0.15−0.05 −0.100.000.100.200.30
z=(E e/Eo) − 1
0.00
0.25
0.50
0.75
0.9981
io= 40o
Total observed line profile affected by
• grav. Redshift
• Light bending
• rel. Doppler shift
• emissivity profile
• spin of black hole
MCG−6-30-15 1
MCG−6-30-15, IL
ine
flu
x (
10
ph
/s/k
eV
/cm
)
−5
2
05
10
15
0.4 1.41.21.00.80.6
ν/ν 0
MCG−6-30-15 (z = 0.008): first AGN with relativistic disk line
Tanaka et al. (1995): time averaged ASCA
spectrum: line skew symmetric
=⇒ Schwarzschild black hole.
MCG−6-30-15 2
MCG−6-30-15, IIL
ine
flu
x (
10
ph
/s/k
eV
/cm
)
−5
2
05
10
15
0.4 1.41.21.00.80.6
ν/ν 0
0.4 0.6 0.8 1.21.0 1.4
ν/ν 0
010
15
Lin
e flu
x (
10 ph/s
/keV
/cm
)
−5
2
5
MCG−6-30-15 (z = 0.008): first AGN with relativistic disk line
Tanaka et al. (1995): time averaged ASCA
spectrum: line skew symmetric
=⇒ Schwarzschild black hole.
Iwasawa et al. (1996): “deep minimum
state”: extremely broad line
=⇒ Kerr Black Hole.
MCG−6-30-15 3
MCG−6-30-15, IIIL
ine
flu
x (
10
ph
/s/k
eV
/cm
)
−5
2
05
10
15
0.4 1.41.21.00.80.6
ν/ν 0
0.4 0.6 0.8 1.21.0 1.4
ν/ν 0
010
15
Lin
e flu
x (
10 ph/s
/keV
/cm
)
−5
2
5
MCG−6-30-15 (z = 0.008): first AGN with relativistic disk line
Tanaka et al. (1995): time averaged ASCA
spectrum: line skew symmetric
=⇒ Schwarzschild black hole.
Iwasawa et al. (1996): “deep minimum
state”: extremely broad line
=⇒ Kerr Black Hole.
Later confirmed with BeppoSAX (Guainazzi et al., 1999) and RXTE (Lee et al., 1999).
XXX
MCG−6-30-15 5
XMM-Newton Data, II
Two long XMM-Newton Observations of MCG−6-30-15 available:
June 2000 : (rev 108), 100 ksec
(Wilms et al. 2001, Reynolds et al., 2003)
• overall flux ∼ ASCA “deep minimum”
• source strongly variable
• broad Fe Kα line
July/August 2001 : (rev 301–303), 315 ksec
(Fabian et al. 2002, Vaughan et al. 2003, Ballantyne et al. 2003)
• source 70% brighter than in June 2000
• similar variability
• broad Fe Kα line
MCG−6-30-15 6
Data Extraction
0
2
4
6
8
10
12
14
Cou
nt r
ate
[cps
]
0 6 12 18 0 6 12 182000 July 11 2000 July 12
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2
JD−2451736
XMM−Newton EPIC pn (singles)
RXTE PCA (normalized to 1 PCU)
XMM-Newton and RXTE lightcurves, ∆t = 100 s.Total XMM-Newton exposure∼100 ksec.
Overall flux ∼ ASCA “deep minimum” (F2−10 = 2.3 × 10−11 erg s−1 cm−2)To avoid possible pile-up, use data from < 14 cps in EPIC pn onlyfor MOS: Ring 10′′, 35′′, pattern 0–12.
MCG−6-30-15 7
Spectral Analysis
10-2
10-1
10 0
10 1
coun
ts s
-1 k
eV-1
Single Events
Double Events
Energy [keV]
0.60.81.01.21.41.61.8
data
/mod
el
1.0 5.0 10.0
MCG-6-30-15EPIC pn
X-ray spectrum very
complex:
• warm absorber at
. 1.5 keV
• power-law continuum
• reflection continuum
• relativistic iron line
• narrow iron line
Low energy end of
spectrum: warm absorber
=⇒ RGS.
Then Fe line analysis with
EPIC data.
MCG−6-30-15 8
RGS Data
0
1
2
3
RG
S c
ount
s s-1
keV
-1
Energy [keV]
-5
0
5
∆χ2
0.5 1.0
MCG-6-30-15RGS
RGS1
RGS2
(red: RGS1, blue: RGS2; χ2red = 1.4).
• absorption edges of
C V (400), O VII (740),
O VIII (870), and
Ne X (1362 eV),
• blended resonance
absorption lines below O VII
and O VIII edges,
• four moderately broad, weak
Gaussian emission features
at 0.885 keV, 0.935 keV,
0.995 keV, and 1.06 keV
(lab frame; no interpretation).
(model similar to Lee et al. 2001)
Analysis à la Branduardi-Raymont et al. (2001) gives similar results for Fe line.
Iron Line 1
Iron Line, I
1 10Energy [keV]
0.8
0.9
1.0
1.1
1.2
data
/mod
el
χ2/dof = 2852/1718
Continuum model:
1. empirical “warm absorber model”,
2. power law (Γ = 1.94 ± 0.02),
3. relativistically smeared reflection from a
possibly ionized accretion disk (i = 30,
Ω/2π = 1–1.5).
4. narrow Fe line (EW ∼ 40 eV)
=⇒Clear deviations in the 2–7 keV band
=⇒ relativistically broadened iron line.
Iron Line 2
Iron Line, II
1 10Energy [keV]
0.8
0.9
1.0
1.1
1.2
data
/mod
el
χ2/dof = 2852/1718
Energy [keV]
0.8
0.9
1.0
1.1
1.2
data
/mod
el
2 10
χ2/dof = 1506/1312
Continuum model:
1. empirical “warm absorber model”,
2. power law (Γ = 1.94 ± 0.02),
3. relativistically smeared reflection from a
possibly ionized accretion disk (i = 30,
Ω/2π = 1–1.5).
4. narrow Fe line (EW ∼ 40 eV)
=⇒Clear deviations in the 2–7 keV band
=⇒ relativistically broadened iron line.
5. relativistically broadened Gaussian:
E = 6.95+0−0.15 keV, WKα = 582 eV; with
emissivity IFe(r) ∝ r−β: β = 4.6 ± 0.3
Iron Line 3
Observational Summary
0.6
0.8
1.0
1.2
1.4
1.6
1 100.8
0.9
1.0
1.1
1.2
1.3
data
/mod
el
2 3 4 5 6 7 8 9Energy (keV)
6
8
10
12
14
16
ν f ν
(10
-3 k
eV c
m-2
s-1
)
a
b
c
a pure PL fit,
b Setting the normalization of the lines
to zero reveals the extreme width
and skewed profile of the Fe line,
c Components of the final fit
Iron Line 4
Other XMM Observations
Lin
e fl
ux (
10 keV
/cm
/s
/keV
)2
−4
82
46
4 6 8
0
Energy (keV)
after Fabian et al. (2002)
July/August 2001 315 ksec
observation
(Fabian et al., 2002)
• Strong narrow line
• broad line clearly present
• emissivity profile very steep forradii close to rin
IFe Kα ∝
r−5.5±0.3 for r < 6.1+0.8−0.5rg
r−2.7±0.1 outside
(Fabian & Vaughan, 2003)
Caveats 1
Caveats
Broad line obviously depends on assumed continuum
• ionized reflection models not yet very realistic!but would require big change at low E, rather improbable. . .
=⇒ Ballantyne & Fabian models give same result for line!
• absorption in ionized gas? attempts of including this generally unsuccessful. . .
• partial covering? so far, all partial covering models we tried failed in the RXTE-PCA. . .
• simple PL continuum usedexpected from Comptonization; fitting realistic Comptonization models to XMM/RXTE data
results in large β and kTe & 70 keV =⇒ PL
• Continuum variabilityΓ changes with flux, might mimic broadening
=⇒ time dependent analysis shows line remains broad
Caveats 2
Alternative Continuum Models
• Partial covering à la Pounds et al.
(2002) does not work with PCA!
• ionized reflection, as assumed
here, works with PCA!
Caveats 3
Observational Summary
3 4 5 6 7 8Energy [keV]
0
5
10
15
Pho
ton
Flu
x [1
0−5 p
h cm−
2 s−
1 ke
V−
1 ] MCG−6−30−15Tanaka et al. (1995)
1 10 100Energy [keV]
10−3
10−2
ν f ν
[keV
2 c
m−
2 s−
1 keV
−1
Tanaka et al., 1995 Wilms et al., 2001, 2003
XMM-Newton: MCG−6-30-15 has broadest Fe Kα line seen in all AGN
Width of the line: Kerr Black Hole is required (Confirming ASCA and BeppoSAX )
The emissivity index, β, is large =⇒ Line from centralmost regions!
Interpretation 1
Interpretation
Shape of line profile depends on
radial line emissivity
XMM-Newton: IFe Kα ∝ r−4.6
line emissivity ∝ flux irradiated on disk
∝ downward coronal flux
∝ local disk emissivity
∝ local energy dissipation rate1 10 100
Energy [keV]
10−3
10−2
ν f ν
[keV
2 c
m−
2 s−
1 keV
−1
BUT: Accretion theory: dissipation rate ∝ r−3!
XMM-Newton observations show that more energy dissipated in disk
than predicted from theory =⇒ additional energy source required!
Interpretation 2
Interpretation
Hypothesis: Magnetic couple between BH
and accretion disk:
1. Magnetic connection between inner disk
and plunging region (Agol & Krolik, 2000)
2. Magnetic connection between inner disk
and rotating event horizon (Li, 2000, 2002,
2003).
“torqued disks”
Complications:• Magnetic field structure for r < rin: connection
btw. plunging region and disk too weak forenergy release required? (Li, 2003) [but:contradicts numerical simulations, e.g., Hawley& Krolik, 2001]
• Corona might be suppressed by torquing (Merloni & Fabian, 2003)• gravitational focussing of X-rays? Petrucci & Henri (1997) show that in this case central source needed
as well (but see Martocchia, Matt & Karas 2002).• relativistic outflow? Proposed by Titarchuk, no comparison with data yet. . .• limb darkening law of accretion disk? (Beckwith & Done, 2004, no fits yet).
Line Profiles 1
Realistic (?) Line Profiles
Payne & Thorne (1994) accretion disk, torqued at
r = rms (Agol & Krolik, 2000), including returning
radiation
χ2/dof = 2075/2104
=⇒Fits need large increase of inner disk radiative
efficiency to explain line profile, are dominated
by torquing.
Reynolds, Wilms, et al. (2004)
Summary 1
Summary
• XMM-Newton observations show extremely broad line.
• Line profile implies very steep emissivity profile.
• Steep emissivity profile also seen in longer observation of Fabian et al.
(2002).
• Shakura & Sunyaev type disks cannot explain inferred emissivity profile.
• Plausible explanation by elimination of other alternatives:
spin extraction from a magnetized black hole
=⇒More theoretical work on line emissivity profiles from (Kerr) black
holes is needed!
And the future?
• more deep AGN observations with XMM-Newton
• Simbol-X
• XXX-XEUS
=⇒ enough to do for age 65 – 85. . .
1–1
*
Bibliogr
1–1 BIBLIOGRAPHY
Agol, E., & Krolik, J. H., 2000, ApJ, 528, 161
Branduardi-Raymont, G., Sako, M., Kahn, S. M., Brinkman, A. C., Kaastra, J. S., & Page, M. J., 2001, A&A, 365, L140
Fabian, A. C., et al., 2002, MNRAS, 335, L1
Guainazzi, M., et al., 1999, A&A, 341, L27
Iwasawa, K., et al., 1996, MNRAS, 282, 1038
Lee, J. C., Fabian, A. C., Brandt, W. N., Reynolds, C. S., & Iwasawa, K., 1999, MNRAS, 310, 973
Lee, J. C., Ogle, P. M., Canizares, C. R., Marshall, H. L., Schulz, N. S., Morales, R., Fabian, A. C., & Iwasawa, K., 2001, ApJ, 554, L13
Martocchia, A., Matt, G., & Karas, V., 2002, A&A, 383, L23
Petrucci, P. O., & Henri, G., 1997, A&A, 326, 99
Tanaka, Y., et al., 1995, Nature, 375, 659
Wilms, J., Reynolds, C. S., Begelman, M. C., Reeves, J., Molendi, S., Staubert, R., & Kendziorra, E., 2001, MNRAS, 328, L27