What Makes them Black Holes

• Other than the classical Rees argument about efficiency, size and luminosity what observational properties make these objects black holes ?

– High mass in a small volume via direct measurements

SGR A*, NGC4258 etc – Mass functions of stellar systems

For the vast majority of objects thought to be black holes such information is not

available

– We must use indirect observational data

• Spectra

• Timing

• Spectral/timing (reverberation mapping)

• Imaging (micro-lensing)

– E.g. things that look like other BHs are also BHs

Why are black hole interesting today

Black Holes and Strong Gravity–Spectral and timing probes of strong gravity–Astrophysics in the strong gravity region

AGN Winds & effect of BHs on cosmic structure–Outflows from AGN–Cooling flow and cluster entropy problems–Role of AGN in galaxy formation

Evolution of AGN and SMBH growth–Paradigm shift in AGN evolution

X-rays • I will focus on the x-ray properties for

several reasons– Many galactic black holes (GBH) optical

spectra and light curves show weak/complex evidence for the black hole- All (but 1?) GBH found by x-ray survey

– The space density of x-ray selected AGN is much higher than that of optically selected AGN (Hasinger)

126/286 Not detectedby HST in z’

Nuclei often 5 mags fainter than expected; at 3000A N(H)~1022 atm/cm2 for MW dust to gas ratio (Barger et al 2005)

X-rays show the highest amplitude/shortest timescales of variationOnly spectral feature from near the event horizon

Black Holes Across the Mass Spectrum

• This is a very large topic and I cannot cover everything

• In particular I will not cover radio emission and recommend the book

From X-Ray Binaries to Quasars: Black Holes on All Mass Scales Maccarone, Fender, Ho, and the review article by G. Fabbiano. in :Memorie della Societa Astronomica Italiana, v.77, p.728 (2006)

I will also not talk about spin- original or otherwise

Many of the topics in this talk have been the subject of whole meetings

Broad Fe K linePower density spectraSpectral transitions in black holes etc etc

And thus the the presentation here is ‘superficial’

JetsAccretion disk

Mass donor star

Blandford

AGN

Winds

Binary

Black Holes Across the Mass Scale

• There exist a wide range of black hole masses– MW ‘Stellar’ mass black

holes M< 20M (formed from

well understood stellar evolution processes- Fryer and Kalogera 2003) -talk by Casares

– “Massive’ Black holes at the centers of galaxies

M >3x105 M whose origin

is not clear

– Perhaps ‘intermediate mass’ black holes 20>M>103 M

whose existence is controversial and whose origin is not understood (Makishima etc)

Mass Distribution of Milky Way Black holes

Kubota 2005

No M>M 18 objects in Milky Way

Black Holes Across the Mass Scale

• There exist a wide range of black hole masses– For black holes at the centers

of galaxies there is a strong correlation of black hole mass and stellar velocity dispersion (Gebhardt et al, Ferrarese et al)

– For AGN- the masses of the black holes derived via ‘reverberation’ observation agree with the same scaling relations (Peterson et al 2005)

– Perhaps ‘intermediate mass’ black holes 20>M >105 M whose existence is controversial and whose origin is not understood (more later)

Mass of Black Holes in Center of GalaxiesBarth et al 2005 Precision masses (MW, M31, M32< NGC4258) show factor of 3 scatter in relation

109

107

105

103

Black Holes Across the Mass Scale • Using stellar velocity dispersion

as a scaling parameter the mass of AGN black holes and ‘passive galaxy’ black holes is very similar (Peterson 2005)

• It is secure that black holes exist (see talk by Genzel)

Mass of AGN Black Holes vs quiescent Galaxies Peterson 2005

109

107

105

104

Bla

ck h

ole

mas

s fr

om r

ever

bera

tion

AGN Mass- filled circlesPassive galaxies -open Mass function for a stellar mass GBH

M31model of BH vs data

How do we know how massive they are?

• In ~20 nearby galaxies the velocity of gas (including masers) and/or stars near the black hole has been measured- velocity law implies a central black hole

• Using Kepler’s law the masses have been measured.-precision masses exist for only a few objects.

Genzel et 2005

Velocity of Gas near M87 nucleus (Harms et al)

Mass Model for ctr of MW

Most of the nearby black holes are quiescent

• Many nearby black holes are extremely dim in x-ray and optical band (Soria et al 2006)- the bolometric corrections are not known, but are typically ~10-20

• many x-ray black hole transients in their quiescent state are 104-107 times dimmer than in ‘on’ state

• It is not yet clear what property makes a black hole active- either for AGN or GBHs

Log

Lx/

LE

dd

Log M(dot)/MBondi

Micro-lensing can constrain size of ‘light’ source • Multi-wavelength lensing

measurements (Pooley et al 2006) allow measurements of the size of the emission line region in different wavelength bands - in principle test the actual physical size of quasars

•In practice one measures the size of the emitting regions

–The x-ray emission region is smaller than the optical –The optical emission region is larger than that predicted by standard disk models

M=3x108 standard accretion disk size

Orbiting Spots Near the Black Hole ( Matt, Reynolds) Spin parameter

20000 60000 10000

sec

?

• Recent evidence has show that some AGN show ‘rapid’ changes in part of the Fe K line profile (Turner et al 2003, Iwasawa et al 2004)

• If these changes can be interpreted as rotating ‘hot spots” (Dovciak et al 2005,Goosman 2006) the mass can be estimated

Reynolds et al

Typical X-ray spectra in different GBH ‘states’

• There is a strong correlation between the Eddington ratio of a GBH and its x-ray spectra

• As one moves from ‘low-hard’ to ‘high’ to ‘ultrasoft’ the Eddington ratio increases and the spectra gets ‘more’ thermal in nature (e.g. more of the bolometric luminosity is carried by a component which can be associated with an optically thick disk, thermal disk)

• This transition has never been seen in an AGN-may have been seen in ‘AGN’ transients (Komossa et al 2005)

• ULX sometimes show transitions in the opposite sense

• In the very high state the spectrum is highly Comptonized

• Some of these properties are unique to black holes and are never seen in neutron star spectra

AD?

Open circles black holes , colors Neutron stars Hatched region never visited by NS

• In a large x-ray spectral survey of x-ray spectra in the MW Done and Gierlinski have shown that there is a spectral region which NS never visit.

– They interpret this as

evidence that BHs do not have a surface and NS do

• GBHs never have x-ray bursts (nuclear burning on the surface) or x-ray pulses

• All AGN spectra are in the GBH regions, some ULX have spectra in ‘NS’ region (Winter et al 2006)

X-ray Spectra of Accreting Compact objects

Neutron Star and Black Hole Spectra

NS BH

Blacks Holes in Quiescence (104 -108 times dimmer )

• McClintock et al (2004) – black holes in ‘quiescence’( a

situation in which the accretion rate is very low) are

– much dimmer than NS in ‘quiescence”

– because they have a horizon through which they ‘swallow’ the advected energy? -

– e.g Black holes are black e.g Black holes are black

• So far not seen in AGN or ULXs - impossible today to follow AGN or ULX at 105 lower flux !

X-ray luminosity vs orbital periodBlack holes are black circles

(the mass accretion rate is highly sensitive to the orbital period)

Porb

L

og L

min

30

3

2

34

Scaling relations

• Seyfert galaxies (AGN) and galactic black holes both show high amplitude rapid, aperiodic x-ray variability spanning several orders of magnitude in frequency.

• This noise shows a relatively steep PDS at high frequencies (red noise) and is flatter at low frequencies

•The physical origin of this noise is not known( but see Goosmann et al 2006 and Lyubarskii and Kotov, Churazov

& Gilfanov ) , but both its PDS shape and amplitude differ only subtly from neutron star x-ray binaries (Uttley 2006) •Frequently the hard photons lag the soft photons in both AGN and GBHs

Black hole grand unification- Uttley 2006

Time Variability

• All accreting compact objects vary. • Is there some signature of a black hole and of its

mass in the time variability domain?

Gierlinski and Zdziarski 2003

Days

sec 2 AGN Edelson and Markowitz

AGN Light curves in different wavelengths

radio

x-ray

UV

Power density Spectra (figures from v.d. Klis)

Fastest phenomenon in BHs is at 100-450Hz do not change and have high harmonic content In ULXs the QPO frequency changes so cannot be analog of HF QPO of GBHs

break frequency vs. low QPO) Black holesBlack holes (dots), various ‘types of NS (z’s, atolls etc) )

PDS of NS (left) BH (right)

At low frequencies ( <100 Hz) the PDS of NS and GBHs are very similar correlations of QPO components are also similarHighest v show most differences-inner part of accretion disk? Only NS have ‘paired’ kilo-hertz QPO- they show a lot of variation and low harmonic content

Black Hole Power Density Spectra

• In galactic black holes (and AGN???) there is a strong connection between the photon spectrum and the power density spectrum (McClintock and Remillard 2003)

• This is thought to be related to the physics of the inner accretion disk

In the low and very high state there is a change of slope - a break frequency

High state

Low state

Very High state

Power Density Spectrum Photon spectrum

Power density Spectra

L<0.2 LEdd L>0.2 LEdd

AGN PDS

It is not yet clear if we can assign ‘high’ state or ‘low’ state PDS to AGN

“Power-law” emission via thermal Comptonization of seed disc (UV) photons

Soft excess - hard tail of thermal disc emission in EUV (big blue bump)

Warm absorber/Emitter - ionized gas outflowing from nucleus (lightdays - parsec scale)

Iron line emission - accretion disk, BLR, torus, NLR

Compton Reflection - off optically thick matter (disc, torus)

AGN X-ray Spectral Components

Reeves 2005

Broad Band Spectrum of Black Holes

• Qualitatively the broad band spectra of AGN and GBHs are similar- however the effective temperature of the ‘soft black body’ is much lower in AGN appearing in the optical/UV band

Cyg X-1 ratio of a Power Law

soft excess

Cutoff(Te=110keV)

narrow Fe (6.4keV)

Zdziarski et al

Suzaku data

For GBHs kTsoft~0.7-2 keVAGN kTsof~0.1 keVULX kTsoft ~0.1-1 keV

Esin et al 2001

14 16 18 20Log

AGN SED

High Z absorbed QSOHasinger

SED of AGN

Van Duyne et al. 2006SED sorted by L(x)

• Recent very large samples of AGN show systematic trends in the SED- but, so far, no good equivalents of the high state GBH spectra

• A ‘special’ class of AGN (NLSY1s ) are thought to be the cognate of High state GBHs but their SED is not a good match

Barger et al 2005

Elvis 1994

Spectral Variability

• Even in a given ‘state’ the spectral shape can vary

• Thus values like the PL slope and kTBB are not unique for a given object and ‘state’

Gilfanov

Ratio of x-ray spectrum ofCyg X-1 to a power law

Possible Spectral Signature of IMBH

• The Effective temperature of the accretion disk scales as

J.Miller et alTwo sources in NGC 1313

kTcol<0.4 keV M> 700 M for L=LEdd

Only if L<< LEdd can the temperature be low for a lower mass object

The Temperature Luminosity Relation

• Miller et al 2004 show– galactic black hole candidates have

BB component temperatures appropriate for their dynamical masses and observed luminosities

– If the BB components in ULXs are not indicative of their mass then additional (unknown) physics is required to account for the BB component- no sign of warm absorber in ULX spectra- clearly see ‘cold’ ISM features due to O and Fe

– R. Soria presentation

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

L/LEdd~0.01; M=7M L/LEdd~1; M=7M

L/LEdd~0.3,M=1600M

X-r

ay lu

min

osity

(0.

5-10

) ke

V x

1040

kTBB

The Continuum form of AGN, ULX and GBHs

• The low state of GBHs and most AGN spectra can be well characterized by a power law with additional spectral features

Wilms et al 1999 GX 339-4

Wilms et al 2006 Cyg x-1

Spectral correlations

• Cyg X-1 shows the a spectral index - intensity relation in the low state as do many AGN do.

However the sign of the effect depends on energy and object

Softer when Brighter-AGN

Lamer et al 2003- NGC 4051

The trend (with exceptions) for AGN : spectral index increase when the source

is brighter in the 2-10 kev band

Gliozzi et al 2004- 3c 390.3

Markowitz and Edelson 2004

Spectral Index Correlations in AGN

• There is little relationship between x-ray spectral index and optical properties for broad line objects

• There may be a correlation with Eddington ratio

• Redshift dependance ? - (Chartas)

Indirect Imaging of Black Holes Using Spectroscopic De-convolution

• ASCA discovered a relativistically broadened iron line that come from close to the event horizon of black holes in the nucleus of nearby galaxies

• This feature also occurs in GBHs - but so far not in ULXs

• This line provides a unique probe of the inner sanctum near black holes, observing the effects of GR in the strong gravity limit

• Recent results have show not only emission but absorption

Deconvolved image from ASCAline profile

Relativistic lines in Galactic BHs Jon Miller

Many galactic black holes,often show broad Fe K lines -

XMM AGN Fe K line profiles- Reeves et al NGC4151, Fairall 9 Mrk335 NGC7213

NGC4151 NGC7213

Fairall 9 *

MKN335 *

EW=206

XMM Fe K line profilesIIZW2, NGC7469, Mrk 841, NGC3227

NGC7469IIIZw2

MKN841 NGC3227

While many BHs have Broad lines- so do some NS (Asai et al 2001)

However it is not common and in general EW of Fe K emission is low and the line widths are narrower then found in AGN and GBHs

The variety of line shapes and physical effects thus make a broad

Fe K line not unique to BHs

Fe K linesIn Neutron stars

J. Miller et al 2004,2005

J. Miller et al 2004,2005However there is not yet a detailed1:1 comparison of the Fe K line shapes in AGN and GBHs

There is a very wide range of accretion rates in AGN

• Based on SDSS [OIII] data there is a very wide range of Eddington ratios in AGN

• We now need work on connecting the SED and x-ray spectra to the Eddington rate

• So far there seem to be only subtle effects - more work needed

Heckman 2004)

‘Unification’ of Black holes

• Whole meetings have been devoted to this subject but is is fair to say that there are strong resemblances in the both the photon spectra of AGN and GBHcs in the low state and between the PDS in AGN and GBHs

• SUPERUNIFICATION OF ACTIVE GALACTIC NUCLEI:Black Hole Mass, Spin and Accretion RateElba Island (Italy), May 25-28, 2005

Radio, X-ray and Mass

• There is an apparent relation between the radio and x-ray luminosity and the mass of the black hole.

• Unfortunately I do not have time to discuss the radio emission - see Markoff et al

Merloni 20053 orders of magnitude

Absorption features

• Both AGN and GBHcs (but so far not ULXs ? NGC1313 Makishima) show Fe in absorption as well

• However these features are also seen in neutron star binaries and thus are indications of winds and ionized material and not a black hole signature

AGN MCG-6-30-15Young et al 2006

GBH

Resonant Abs Lines • Low velocity resonant abs lines from a wide variety of

ionization stages are seen in both GBHs, AGN and NS • Associated with winds • Probably not line, thermal or radiation driven wind, so

magnetic from inner disc?

Suzaku data for GBH (4U1630Kubota et al 2006 v~1000km/sec

High Velocity Resonant Abs Lines

• So far high velocity Fe abs lines have only been reported In AGN (Pounds et al 2003, Reeves et al) PG1211+143 etc Pounds

• If this is more general it implies that AGN winds can carry a large amount of energy- so far not yet seen in GBHs or ULXs

• PG1211- blue shifted resonance Fe absorption feature V~0.08c (Reeves et al 2003)

PG1211 Chandra LETG data-If features are Fe v ~0.26, 0.4c

AGN High Velocity Lines

• There are now several other examples of high velocity blue shifted Fe absorption lines in AGN

• These indicate a fast wind with a mass flow rate greater than the accretion rate.

• If this is a general BH phenomenon have a strong influcence on galaxy formation, ISM and IGM

Gibson et al 2005

MR2251-17 Highly blue shifted featureV=12,700km/sec

1.75 1.8 1.85 1.9 Wavelength (A)

Power Density Spectra

• power density spectra , for many galactic black holes and AGB are flat at low frequencies steep at high frequencies

• This form seems to be ‘ubiquitous’ in black holes (not seen in high state GBHs)

• The break frequency scaling as mass of object

Hayashida et al

NGC3516 Nandra and Edelson

NGC4559 x-7

Break timescale days

Bla

ck h

ole

mas

s

How is Time Variability Characteristics Related to Mass

• Measurement of the break frequency in the PDS seems to be related to the black hole mass.

• However the shape of the PDS in galactic black holes depends on state:

• Not clear if this applies to AGN or not

• Situation for ULXs is unclear since only 2-3 breaks in the PDS have been measured

Mass Measurements from Timing

• Comparison of BH mass from x-ray timing vs other methods seems to agree well

Power density Spectra ULX • In galactic black holes there is a

correlation between the photon spectra and the QPO frequency

• For objects of known mass there is a strong relation which may allow ‘calibration’ of the mass- QPO Frequency

• In the case of M82 (Fiorito and Titarchuk,Dewangan et al 2006,Muciarelli et al

2006 ) this indicates M~25-500M

Spe

ctra

l ind

ex

QPO frequency

In M82 the QPO freq varies from 50-166 mHz

M82

GBH

Power Density Spectra ULXs • NGC5408- Soria et al observed a break in the

PDS for NGC5408 X-1.• A deep (130ks) XMM observation has

confirmed this and discovered QPOs at ~20mHz and perhaps a break at low frequencies

• The absolute power in the source is very high rising to ~3 at 0.001 10X ~10x larger than that seen in galactic black holes

• Most ULX do not have detectable power (Feng and Kaaret 2006)

NGC5408

Comparison of M82 and NGC5408 PDS with AGN and Galactic Black holes

M82

4U 1543-47

NGC5408

AGN

GBH

Relationship Between Fe K line width and QPO- Gilfanov et al

Systematic Changes with Eddington Ratio

Tomsick et al 2005

Yamaoka et al 2004- galactic black holes

Nature of the X-ray Spectrum

Sspectra of Milky Way black holesfall into 2 broad classes

–Powerlaw spectra (low state)–Disk Black body +power law (high state)

•The x-ray spectra of the ULXs can be different

• ~1/4 of bright objects are better fit by a very hot disk black body model or comptonized spectrum than a power law, •A significant fraction ~1/3 of ULX require a soft black body (kT<0.5 keV) component (Winter et al 2005, Miller et al 2004)

Colors of galactic sources Done et al-ULX

Conclusion

• There are multiple similarities between BHs across the mass scale but also differences

• In particular the ULX seem to have different detailed properties

• The relationship between spectral and timing properties of GBHs is not seen in AGN and may not be true for ULXs.

• ULXs seem to have very dim disks (soft x-ray, UV, optical, IR (?)

• No state transitions have been observed in AGN, and in ULXs the state transitions maybe different than in GBHs

• Have we actually sampled the same physics across all mass scales- not yet clear.