Black holes: NeXT steps….

Chris DoneUniversity of Durham & ISAS

• Appearance of BH should depend only on mass and spin (black holes have no hair!)

• Plus mass accretion rate – L/LEdd

• 104-1010 M: Quasars • 10-1000(?) M: ULX • 3-20 M: Galactic black holes• How does accretion flow

change as function of M/M

spin and L/LEdd ?

Black holes

Big questions: First BH & galaxies • stellar bulge ( or MB) gives BH mass – link between galaxy and BH

• QSO peak at z~2 L/LEdd ~1 onto ~108 M0 (local NLS1)

• First SMBH = first galaxies. QSOs z~6, so first BH ~106 M0 z~10?

Bla

ck h

ole

mas

s

Stellar system mass

103

109

1061012

Big questions:ULX

• Ultra - Luminous X-ray sources in spiral arms of nearby starforming galaxies – ULX

• L~1039-41 ergs s-1 M~10-1000 M for L<LEdd

• Hard for stellar evolution to make BH > 50 M

• How to form bigger BH? How to get L/LEdd>1?

• Need to know mass!!!

Gao et al 2003

• Nature of the jet=AGN feedback on cluster cooling! link to spin ?

• MRI simulations say powerful jets for moderate spin – don’t need a=0.998. (weak) jet a=0

• LMXB: gravitational waves in core collapse may limit spin to <0.8 (Gammie et al 2002)

• AGN: high spin for BH in major mergers (big galaxies) but not from multiple small mergers Radio loud/quiet?

• Need to measure spin!!!

Big questions: BH jet – spin?

Big questions: strong gravity

• Accreting BH: huge X-ray luminosity close to event horizon

• Bright emission from region of strong spacetime curvature

• Spectral distortions depend on velocity, geometry and GR

• Observational constraints on strong gravity if we know velocity/geometry!

• Need to understand accretion!

• Differential Keplerian rotation

• Viscosity B: gravity heat

• Thermal emission: L = AT4

• Temperature increases inwards

• GR last stable orbit gives minimum radius Rms

• a=0: Tmax = (M/M )-1/4 (L/LEdd )

1/4

• 1 keV (107 K) for 10 M

• 10 eV (105 K) for 108 M

• a=0.998 Tmax ~ 2.2 Tmax (a=0)

• AGN: UV disc, ISM absorption, mass more uncertain. XRB…

Spectra of accretion flow: disc

Log

Log

f

(

• Disc in X-ray band so not so heavily absorbed as UV

• Huge amounts of high s/n data• Variability timescales

• ms – year (observable!)

• hours – 108 years in quasars • Observational template of

accretion flow as a function of L/LEdd onto ~10 M BH

• Need ASM…

Galactic Binary systems

7 years

• Bewildering variety of spectra from single object in XRB!

• Underlying pattern

• High L/LEdd: soft spectrum, peaks at kTmax often disc-like, plus tail

• Lower L/LEdd: hard spectrum, peaks at high energies, not disc!

• All GBH in RXTE archive consistent with SAME spectral evolution 10-3 < L/LEdd < 1

• How is hard X-ray tail produced?

• Why change with L/LEdd ?

Done & Gierlinski 2003

X-ray Spectra

1.53.04.5

Observed GBH spectra• RXTE archive of many GBH • Same spectral evolution 10-3 < L/LEdd < 1• Need this sort of data for AGN

Done & Gierliński 2003

1.5

4.5

3.0

6.4-

16)

3-6.4)

1.5

4.5

3.0

6.4-

16)

• Disc models assumed thermal plasma – not true at low L/LEdd

• Instead: hot, optically thin, geometrically thick inner flow replacing the inner disc (Shapiro et al. 1976; Narayan & Yi 1995)

• Hot electrons Compton upscatter photons from outer cool disc

• Few seed photons, so spectrum is hard

Accretion flows without discs

Log

Log

f

(

Qualitative and quantitative models: geometry

Log

Log

f

(

Log

Log

f

(

Hard (low L/LEdd)

Soft (high L/LEdd)

Done & Gierliński 2004

Moving disc – moving QPO

• Low frequency QPO – very strong feature at softest low/hard and intermediate and very high states

• Moves in frequency correlated with spectral slope

• Moving inner disc makes sense of this. Disc closer in, higher f QPO. But also more soft photons from disc so softer spectra (di Matteo Psaltis 1999)

Qualitative and quantitative models: geometry

Log

Log

f

(

Log

Log

f

(

Hard (low L/LEdd)

Soft (high L/LEdd)

Done & Gierliński 2004

And jet….• ALL produce jet and are consistent

with same radio/X ray evolution • No special QSO class• Depends on spectrum - maximum

radio at brightest low/hard, intermediate and very high states (like QPOs) i.e. L/LEdd (Merloni et al 2003; Falke et al 2004)

• Strongly quenched in disky spectra….

Gallo et al 2003

• And link to jet behaviour! Sudden quenching of jet for disky spectra!• Need geometrically thick flow to get large scale B field for jet? • MRI gives jet from accretion flow – more powerful at higher spin but

doesn’t require extreme a=0.998 (even present at a=0: Kataoka’s talk)• Need to MEASURE spin

Big questions: X-ray tail – jet? F

ender Belloni &

Gallo 2004

Observed disc spectra: spin?

• Pick ONLY disky ones

• L/LEdd T4max (Ebisawa et al 1993; Kubota

et al 1999; 2001)

• Proportionality gives Rms i.e. a

• Bit tricky as disc not blackbody. Some Compton distortion in upper layers of disc – fcol

• Good models now (full vertical disc structure + radiative transfer with metals + full GR) Davis et al 2005

• RANGE a = 0.1-0.8 but NOT extreme (depend on i, M, D)

Davis, Done Blaes 2006

Gierlinski & Done 2003

Observed disc spectra

• Pick ONLY disky ones

• L/LEdd T4max (Ebisawa et al 1993; Kubota

et al 1999; 2001)

• Proportionality gives Rms i.e. a

• Bit tricky as disc not blackbody. Some Compton distortion in upper layers of disc – fcol

• Good models now (full vertical disc structure + radiative transfer with metals + full GR) Davis et al 2005

• RANGE a = 0.1-0.8 but NOT extreme (depend on i, M, D)

Relativistic effects

Fabian et al. 1989

Energy (keV)

flux

• Relativistic effects (special and general) affect all emission (Cunningham 1975)

• Hard to easily spot on continuum components

• Fe Kline from irradiated disc – broad and skewed! (Fabian et al 1989)

• Broadening gives an independent measure of Rin – so spin if ISO (Laor 1991)

• Models predict increasing width as go from low/hard to high/soft states

Problems: extreme Fe lines• Broad iron lines are

common in GBH. Some indications of increasing width with spectral softness

• But some are extremely broad, indicating high spin (if disc to ISO) and extreme emissivity – tapping spin energy of black hole? (Miller et al 2002; 2003; 2004; Minuitti et al 2004)

Miller et al 2004

• BUT these are the same objects for which a<~0.7 from disc spectra

• Mainly in VHS and softest low/hard states (intermediate states) flow extends below ISO? B field means not force free so LSO distorted…

XTE J1650-500: real conflict

• QPO and broadband data say that disc not down to ISO (RXTE)

Done & Gierlinski 2005

•10-3 < L/LEdd < 1

Extreme line in bright LH state of J1650

• MECS data (moderate resolution from BeppoSAX)

• pexriv+narrow line. Best fit is extreme spin (Rin=2) and emissivity q=3.5 (Miniutti et al 2004)

• But ionised reflection – line is intrinsically comptonised (Ross, Fabian & Young 1999)

Done & Gierlinski 2005

• Ionisation balance in all these from Done et al 1992 and is VERY approximate!

• Slab temperature given rather than calculated!

• Constant density ionised disc (CDID/reflion) models of Ballantyne, Iwasawa & Fabian 2001; Ross & Fabian 2005

• Irradiated slab so and temperature drops as go deeper

• Also includes compton UPSCATTERING! SMEARS edge Ross, Fabian & Young 1999

Reflection from a slab: constant n

pexriv: =3x103

CDID: =103

Extreme line in bright LH state of J1650

• Proper reflection models: very different shape for ionised reflection – line and edge are intrinsically broadened by Comptonisation in the disc (Ross, Fabian & Young 1999)

• Still extreme but relativistic effects only 2=13 not 190 as for pexriv. Significance much reduced

Done & Gierlinski 2005

Extreme line in bright LH state of J1650

• Absorption lines – outflow. Rin~10 with normal emissivity. Nh~1023, log =2-3

• Need strong dip at 7.5keV so either fast outflow at v~0.1c or else strong abs at 7.5 from K and/or Ni K

Done & Gierlinski 2005

Absorbers now commonly seen• GRS1915+105 and J1655-40

in ASCA data (Ueda et al 1998;

Kotani et al 2000) • XMM 3 disky spectra. BIG

dips - 7.5keV! Also Suzaku x1630 Sala et al 2006, Kubota et al 2006

• Chandra HETG Miller et al 2006

• Chandra absorption lines – outflow of 1500 km/s, broadening by velocity dispersion of similar speeds

• Probably not line, thermal or radiation driven wind, so magnetic from inner disc?

• What is the effect of this on the disc structure?? Miller et al 2006

High frequency QPO’s• 2 tiny features – give small radius• ..frequencies harmonically related

2:3 RESONANCE. but which??• Orbits in GR: (Stella & Vietri 1998)

– v(circ, plane)– v(circ, vert)– v(ellipse, plane)

• Coupled by gas dynamics so can get ‘parametric’ 2:3 resonance between vertical and radial perturbations – requires a>0.95 for GBH where disc says lower…

• But if large scale height flow then also compressive modes then radius larger, and no strong constraint on a.

Depend on M, a, R

High frequency QPO’s• 2 tiny features – give small radius• ..frequencies harmonically related

2:3 RESONANCE. but which??• Orbits in GR (stella & vietri 1998)

– f(circ, plane)– f(ellipse, plane)– f(circ, vert)

• Coupled by gas dynamics so can get ‘parametric’ 2:3 resonance between vertical and radial perturbations – a>0.95 (Abramovicz et

al 2000) BUT disc says lower…• But if large scale height flow then

also compressive modes then radius larger, and no strong constraint on a (Blaes et al 2006)

Depend on M, a, R

Big questions: summary

• Understand accretion flow at 10-3 L/LEdd < 1 in GBHtruncated disc, hot inner flow with jetdisk to last stable orbit, no hot flow, no jet (extreme)

• Measure spin to test how accretion flow produces jetsdisk spectra - L T4 – depends on vertical structurebroad iron lines – depends on absorption high frequency QPO’s – depends on identifing QPO

• Use understanding of flow to test GR – even with extremely broad lines then its fairly Einsteinian. No more than factor 2, strongly limits higher order curvature modifications to gravity

Big questions: summary

• Need comparable data quantity and quality on AGN in this range of L/LEdd – how does flow properties scale with mass?

maximum hard X-rays from brightest low/hard state – CXB?

• Understand accretion flow at L/LEdd > 1

nature of ULX (GBH at high L/LEdd show low T disc)

How do winds affect the disc structure

NLS1’s (T Boller)

use to predict AGN spectra at peak of QSO activity z~2

use to predict first BH spectra in early universe