Formation of the Galaxies: Current Issues Joe Silk University of Oxford Gainesville, October 2006.

Formation of the Galaxies:

Current Issues

€

Joe Silk

University of Oxford

Gainesville, October 2006

Some remarks about star formation…mass, light, chemistry control galaxy evolution Low mass stars control M

Solar mass stars control light in a spheroidal galaxy

The most massive stars dominate the light in a disk galaxy

Intermediate mass stars control chemical evolution

THE INITIAL STELLAR MASS FUNCTION What determines the characteristic

mass of a star?

Is the IMF universal?

Kroupa 2004Kroupa 2004

Stars

Fundamental theory applied to a diffuse interstellar cloud that is collapsing under self-gravity

Minimum fragment mass

a robust but wrong result! Resolution: continuing accretion of cold gas, eventually halted by feedback that taps stellar energy via MHD turbulence

first stars were massive

In addition IMF most likely also involves fragmentation

M 01.0~~ 2/3pg m−α

€

˙ M gas ~vs

3

G⇒

3 PROCESSES PLAY A ROLE:FRAGMENTATION, ACCRETION, FEEDBACK

Shu 2006

NGC1333: Quillen et al. 2006

Pudritz et al. 2006Shu 2006

Klessen 2006

Ellipticals are old because infall is quenched….by AGN outflowsEfficient early star formation occurred in massive spheroids and ellipticals

There are likely to be two modes of star formation: disks/pseudobulges AND elliptical/spheroid formation

Disk galaxy star formation is inefficient, due to SN feedback Accretion and minor mergers renew gas supply

Accretion, mergers and AGN outflows are key ingredients

Galaxies

Gas cooling time-scale

Dynamical time-scale

A necessary condition for star formation is cooling:

)(Lφ

luminosity

theory (CDM-motivated)

observations

LL 10103~ ×∗

too many Dwarfsbut they are fragile

too many Giants:a problem!

2)(~

nT

nkTtcool Λ

nGmt

p

dyn

1~

€

Mcooled −baryons ~ α g−2α 3 mp

me

⎛

⎝ ⎜

⎞

⎠ ⎟tcool

tdyn

⎛

⎝ ⎜ ⎜

⎞

⎠ ⎟ ⎟T

1+2β

So the BIG ISSUE is astrophysical feedback

€

β ~ −0.5

Ultraluminous infrared galaxies and the galaxy luminosity function

Sanders 1999

The red sequence evolves

Bell et al. 2004 Blanton 2006

Star formation was efficient in the most massive galaxies

Papovich et al. 2006

More evidence for a shorter timescale

Maraston 2006

AN EFFICIENT MODE OF STAR FORMATION IS NEEDED FOR SPHEROID FORMATION: THE CASE FOR POSITIVE FEEDBACK

D. Thomas

D. Thomas 2006

DISK MODE: motivated by gravitational instability of cold disksstar surface densitygas surface density

SFE = gas

vcool m*,SN

ESN initial

0.02

€

σ

€

≈

Star formation efficiency

THERE ARE PLAUSIBLY TWO MODES OF STAR FORMATION: REGULATED BY GAS SUPPLY, DYNAMICAL TIMESCALE ….

SPHEROID MODE: motivated by gas-rich mergers

A GLOBAL STAR FORMATION LAW FOR DISKS

Need cold gas accretion via infall and/or minor mergersto maintain global disk instabilityNeed low efficiency: due to SN feedback

SFR=0.02 (GAS SURFACE DENSITY)/tdyn

Sajina et al. 2006

fits quiescent and starburst galaxies

NGC 891

HI contours

Oosterloo et al. 2005Boomsma et al 2005

NGC 6946

LOCAL COLD GAS FEEDING BY INFALL

The Rate of Star Formation

( ) ⎟⎟⎠

⎞⎜⎜⎝

⎛×

⎟⎟⎟

⎠

⎞

⎜⎜⎜

⎝

⎛×

⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜

⎝

⎛

1.36(pressure)

1rateformation star ~

pressure ISMambient

by limited bubble a of

Volume-4 maximum

unit timeper

generated

bubbles SN

ofnumber

~porosity

Three-phase ISM

Perhaps porosity self-regulates!

SFR with SN feedback in a multiphase ISM

Slyz et al. (2005)

HISTORY OF STAR FORMATION

Rocha-Pinto 2000: solar vicinity

Allard et al. 2006: M100

Star Formation Rate Simulation

The Mice (NGC 4676 a,b)old stars + gasdensity-dependent SFR shock-induced SFR

Barnes (2004)

Bower et al. 2006

space density of galaxies

GALAXY LUMINOSITY FUNCTION

AGN Feedback

luminosity

Massive spheroids form first

K. Bundy et al. 2006

Cimatti et al. 2006

Bouwens, Illingworth et al 2006

Build-up of luminosity and star formation rate

AGN ARE ANTI-HIERARCHICAL

Hasinger et al. 2006

SMBH formation/feedback in galaxy spheroid formation

Fits observed normalisation and slope

King (2003), Silk & Rees (1998)

Supernovae provide feedback in potential wells of low mass galaxies SMBH outflows provide positive

feedback in massive protospheroids Blowout occurs/star formation

terminates when SMBH- relation is saturated

€

M• = 3 ×109 Msun σ

300 km

s

⎛

⎝

⎜ ⎜ ⎜

⎞

⎠

⎟ ⎟ ⎟

4

€

€

σ

LEdd/c=GMMgas/r2

LEddMSMBH

black holemass

spheroid velocity

dispersion

Triggered global star formation? OUTFLOWS

FROM SMBH OVERPRESSURE ISM CLOUDS star formation timescale tjet<<tgal

yields high efficiency

Labiano et al. 2005z=0.27 radio galaxy

Saxton et al. 2005

star formation rate compared to renormalised black hole feeding rate

Silverman et al. 2006

jet-enhanced star formation in spheroids

redshift

comoving star formation rate

comoving SMBH accretion rate

x 10-3

suppressionby ouflows

gravity-induced star formation

feedback

at z~2, SMBH fall below the relation

Star formation suppressed

Star formation triggered

Borys et al 2006

€

˙ M sfrAGN = ˙ M sfr

SN (tdyn / t jet )

≈ εMgas(v jet /σ ) / tdyn ∝ vw3

AGN-induced outflows & star formation

Boost by ~10! Observed scaling!

€

˙ M gasoutflowAGN ~ LAGN /cvw ∝σ 3

˙ M gasoutflowSN ∝ ˙ M sfrσ

−2.7

OUTFLOWS FROM ULIRGSC. Martin 2005: KI and NaI line profiles

Morganti et al. 2005: HI absorption

Swinbank et al. 2006 a SCUBA galaxy at

z=2.385

multiplicative factor of AGN-triggered SN

Everett & Murray 2006:

extended injectionof energy needed for NGC 4151 outflow

X-ray absorbed QSOs in ULIRGs

Ultraluminous starburstsassociated with AGN absorptionby ionised wind

M. Page et al. 2006

A UNIFIED THEORY

NEGATIVE

POSITIVE

FRESH THEORETICAL INGREDIENTS ARE NEEDED!