MASSIVE BLACK HOLES: formation & evolution Martin Rees Cambridge University.

Post on 11-Jan-2016

217 views 0 download

Tags:

Transcript of MASSIVE BLACK HOLES: formation & evolution Martin Rees Cambridge University.

MASSIVE BLACK HOLES:formation & evolution

Martin Rees

Cambridge University

Themes of this symposium

1*. Radiation, ,accretion jets, winds, etc --- phenomenology and models.

3C31:

OpticalRadio

Themes of this symposium

1*. Radiation, ,accretion jets, winds, etc --- phenomenology and models.

2*. Do ‘holes’ obey the Kerr metric (testing strong-field GR, etc)?

* straightforward scaling laws between

stellar-mass and supermassive holes

Themes of this symposium

1*. Radiation, ,accretion jets, winds, etc --- phenomenology and models.

2*. Do ‘holes’ obey the Kerr metric (testing strong-field GR, etc)?

3. Population and demography of supermassive holes: how do they form and evolve?

* straightforward scaling laws between

stellar-mass and supermassive holes

Observational progress in demography and evolution of

holes(I) Ubiquity of holes in galaxies

Massive black holes?

Yes Yes but black hole mass scales with bulge mass not

total mass

Some at least

Maybe

Giant Ellipticals/S0s Spirals Dwarfs GlobularClusters

black hole mass scales with bulge mass

stellar velocity dispersion of the bulge

Kormendy 2003

Is this really tighter?

Observational progress in demography and evolution of

holes(I) Ubiquity of holes in galaxies

(II) Feedback from hole to galaxy

New clues from deep Chandra observations of Perseus

Fabian et al 03a,b

Observational progress in demography and evolution of

holes(I) Ubiquity of holes in galaxies

(II) Feedback from hole to galaxy

(III) Objects discovered at z > 6 .

A very early assembly epoch for A very early assembly epoch for QSOs QSOs

The highest redshift quasar currently knownSDSS 1148+3251 at z=6.4

has estimates of the SMBH mass MBH=2-6 x109 Msun (Willott et al 2003, Barth et al 2003)

As massive as the

largest SMBHs today,

but when the Universe

was <1 Gyr old!

THE HIGHEST-REDSHIFT QUASARS

Becker et al. (2000)

Fluctuation generator

Fluctuation amplifier

(Graphics from Gary Hinshaw/WMAP team)

Hot Dense SmoothCool Rarefied

Clumpy

Brief History of the Universe

QuickTime™ and a decompressor

are needed to see this picture.

BARYONS: need to COOLCOOL First ‘action’ happens in the the

smallest halos with deep smallest halos with deep enough potential wells to allow enough potential wells to allow

this this (at (at z~20-30)

Hierarchical Galaxy Formation:

small scales collapse firstand merge later to form more massive systems

courtesy of M. Kuhlen

First ‘seed’ black holes?First ‘seed’ black holes?

PopIII stars remnants(Madau & Rees 2001,

Volonteri, Haardt & Madau 2003)

Viscous transport + supermassive star (e.g. Haehnelt & Rees 1993, Eisenstein & Loeb 1995, Bromm & Loeb 2003, Koushiappas et al. 2004)

Efficient viscous angular momentum

transport + efficient gas confinement

First black holes in pregalactic halosFirst black holes in pregalactic halosz≈10-30z≈10-30

Simulations suggest that the first stars are massive M~100-600 Msun

(Abel et al., Bromm et al.)

Metal free dying stars with M>260Msun leave remnant BHs with Mseed≥100Msun (Fryer, Woosley & Heger)

Bar-unstable self-gravitating gas + large “quasistar” (Begelman, Volonteri & Rees 2006)

Transport angular momentum on the dynamical timescale, process cascades

Formation of a BH in the core of a low entropy quasistar ~104-106 Msun

The BH can swallow the quasistar

MBH~103-106 Msun MBH~100-600 Msun

Supermassive holes grow

from seedseed pregalactic BHs..

These seeds are incorporated

in larger and larger halos,

accreting gas accreting gas and dynamically dynamically

interacting interacting after mergers.

QuickTime™ and aGraphics decompressor

are needed to see this picture.

All models for first BHs predict

a biased formation: in the

HIGHEST PEAKS OF DENSITY HIGHEST PEAKS OF DENSITY

FLUCTUATIONSFLUCTUATIONS at z~20-30

Quasar host

Dark matter Galaxies

Mh= 5 x 012M

Mh= 51012M M*

= 1011M

SFR = 235 M /yr MBH

= 108M

Quasars end up in cD galaxies at centres of rich galaxy clusters

today

Mh= 2 x 1015M

Descendant

Mh= 21015M

Formation and evolution of supermassive binaries

1. Dynamical friction

2. Binary hardening due to stars

or accretion of gas

3. Gravitational radiation

t a4

t a

Do they merge?

LISA

Will see mergersof 105 –107 Msol

black holes

2011?

Lisa sensitivity to massive black hole binaries

Dependence of merger rate on mass of minihalos in which first holes form

DWARF GALAXIES/MINIHALOS

ELLIPTICAL GALAXIES

mass

Vesc

(km

/s)

1000

100

10

109 1013

VVre

coil

reco

il

(km

/s)

(km

/s)

Gravitational rocketGravitational rocketbinary center of mass recoil during coalescence due to binary center of mass recoil during coalescence due to

asymmetric emission of GW asymmetric emission of GW (e.g. Fitchett 1983, Favata et al 2004, Blanchet et al 2005, Baker et al 2006)

vvrec rec ≤ 250 km/s≤ 250 km/s

««vvesc esc from today galaxies from today galaxies

≈≈vvesc esc from high-z onesfrom high-z ones

GR SIMULATIONS

at z >10 more than 80% of merging MBHs can be kicked out of their halo(Volonteri & Rees 2006)

the gravitational rocket effect is a

threat at the highest redshifts, when host halos are small and

have shallow potential wells

Can the merger process start early enough toCan the merger process start early enough toAllow build-up of supermassive holes?Allow build-up of supermassive holes? Can the merger process start early enough toCan the merger process start early enough toAllow build-up of supermassive holes?Allow build-up of supermassive holes?

Build-up of holes by accretionBuild-up of holes by accretion

(a (a) Is there a continuous gas ) Is there a continuous gas supply from host halo?supply from host halo?

(b) When supply is super-critical:, (b) When supply is super-critical:, is ’excess’ radiation trapped and/or?is ’excess’ radiation trapped and/or?

accretion inefficient, allowing rapid growth accretion inefficient, allowing rapid growth in hole’s mass ? in hole’s mass ?

Or is there a radiation-driven outflow?Or is there a radiation-driven outflow?

(spin?)(spin?)..

NOTE; Classic argument of Soltan (1982), which compares total mass of holes with total radiative

output, implies that most of the mass is gained via ‘efficient’ accretion.

But most ot the ‘e-folds’ (eg first 10% of mass) could be gained rapidly via inefficient accretion

from Yu & Tremaine 2002

SMBH

=2.5-3.5x105M Mpc-3

~0.2 @ z<5

qso(0)

=2.1x105[0.1(1-)/]M Mpc-3

Elvis, Risaliti & Zamorani 2002

Eddington accretion

=0.2

super-Eddington accretion

Are massive black holes rapidly spinning? (affects maximal accretion

efficiency, minimum variability timescale, importance of Blandford-

Znajek energy extraction, etc)

Spin is modified by BH mergers and Spin is modified by BH mergers and the coupling with the accretion discthe coupling with the accretion disc

mergers can spin BHs either up or down;

alignment with the disc spins up

spin evolution by BH mergers only

spin evolution by BH mergers AND accretion

1. Mass of the BH 1. Mass of the BH seedseed

PopIII stars remnants

MBH~100-600 Msun

Gas collapse via Post-Newtonian instability

MBH~105-106 Msun

2. BH mergers2. BH mergers

Positive contribution:build-up of high masses

Negative contribution(gravitational rocket)

3. Accretion rate 3. Accretion rate

Eddington-limited(continuous or intermittent)

Super-Eddington(Excess swallowed or expelled?)

QuickTime™ and aYUV420 codec decompressor

are needed to see this picture.