Star Formation inthe Central Molecular Zone

Sungsoo S. Kim 金 成洙

Takayuki SaitohMyoungwon Jeon

David MerrittKeiichi Wada

(Kyung Hee University & Rochester Institute of Tech-nology)

(National Astronomical Observatory of Japan)

(University of Texas)

(Rochester Institute of Technology)

(Kagoshima University)

Molecular Gas Distribution in the Milky Way

Longitude-velocity map of CO emission in -2º < b < 2º (Dame, Hartmann & Thaddeus, 2001).

Molecular ring at ~4 kpc

Central Molecular Zone

3-kpc arm

Gap!

The Central Molecular Zone

l-v diagram of CO emission in the CMZ using data of Bally et al. (1987). Figure from Rodriguez-Fernandez &

Combes (2008).Schematic diagram of face-on view

of the inner Galactic bulge byRodriguez-Fernandez & Combes

(2006).

CMZ

±200 pc

Total Gas Mass in the CMZ

SCUBA Galactic center survey at 450 μm and 850 μm (Pierce-Price et al. 2000)

“Thermal dust continuum emission at submillimeters gives optically thin map to

trace essentially all the mass in the CMZ.”

The total mass in the cen-tral 400 pc measured by in-tegrating the total 850 μm emission is (5.3±1.0)x107

M⊙.

Optically thin molecular line(C18O) maps gives a~3 times smaller value.

Star Formation in the CMZ

The Arches (~Myr, ~104 M⊙)

Figer et al. 1999

The Quintuplet (~Myr, ~104 M⊙)

Figer et al. 1999

G359.43+0.02 Cluster of YSOs (~102 M⊙)

Yusef-Zadeh et al. 2009

24μm(MIPS)/8μm(Spitzer) Ratio Im-age

Yusef-Zadeh et al. 2009

Typical recent SFR of 0.04~0.08 M⊙/yr

Sustained Star Formation

• Serabyn & Morris (1996) argued that star formation has been occurring in the CMZ throughout the lifetime of the Galaxy, and formed the “r-2 central star cluster”.

• With a constant SFR of the current value, the inner bulge would have accumulated a total mass of ~109 M⊙.

2.2 μm emission from COBE/DIRBE(Launhardt et al. 2002)

Enclosed mass profiles of different components (Launhardt et al. 2002)

GB

NSC+NSD

200 pc

~109 M⊙

Bulge = Nuclear Bulge + Galactic BulgeNuclear Bulge = Nuclear Stellar Cluster + Nuclear Stellar Disk

Sustained Star Formation

• Serabyn & Morris (1996) argued that Star formation has been occurring in the CMZ throughout the lifetime of the Galaxy, and formed the “r-2 central star cluster”.

• With a constant SFR of the current value, the inner bulge would have accumulated a total mass of ~109 M⊙.

2.2 μm emission from COBE/DIRBE(Launhardt et al. 2002)

Enclosed mass profiles of different components (Launhardt et al. 2002)

GB

NSC+NSD

200 pc

~109 M⊙

Bulge = Nuclear Bulge + Galactic BulgeNuclear Bulge = Nuclear Stellar Cluster + Nuclear Stellar Disk

Disk

log r

log ρ

Bulge

GB

NSD

NSC-2

~3 kpc

~200 pc

~100 pc

“Nuclear Bulges” in Other Spiral Galaxies

• Many (>50%) spiral galaxies have identifiable cores that are photometrically and morphologically distinct from the surrounding bulge (e.g., Böker et al. 2002; Carollo et al. 2002).

• The sizes of these cores range from several tens to hundreds of parsecs.

Hughes et al. (2005)

Nuker pro-file

Inward Transport of the Gas

• Processes for angular momentum loss– shear viscosity– Compression and shocks associated with bar potential– dynamical friction– magnetic field viscosity– dilution of specific angular momentum by stellar mass loss material from outer

bulge (Jenkins & Binney 1994)

Transition from Atomic to Molecular Form

Morris & Serabyn (1996)

• Bar potentials generally have (at least) two distinct stable orbit families, X1 (along the bar) and X2 (perpendicular to the bar) orbits.

• Gas is compressed at the cusps or self-in-tersecting tips of the inner X1 orbits, loses orbital energy, and falls inward to settle onto an X2 orbit.

• This compression and subsequent cooling will cause a transition from atomic to molecular form.

The Questions

• Does this model really work?– Is SF really dominant in the CMZ region?– Is SFR consistent with the observations?

• We have performed global hydrodynamic simulations for these questions.

Previous Hydrodynamic Simulations of the CMZ Region

• Mostly 2-dimesional, isothermal, and/or no SF/feedback/self-gravity.• Not many studies specifically for the CMZ region.

Jenkins & Binney (1994)2D Sticky Particle, N=8k, no self-grav

Englmaier & Gerhard (1999)2D isothermal SPH, N=10–200k

Regan & Teuben (2003)2D isothermal grid, no self-grav

±5 kpc

Rodriguez-Fernandez & Comes (2008)2D hydro grid, 3D motion, no self-grav, N=1M

Namekata et al. (2009)2D isothermal grid, no self-grav/vis-cosityDouble-bar potential

Our Simulations

• Code developed by T. Saitoh: ASURA– 3D SPH with self-gravity– Parallelized tree scheme– Uses GRAPE boards or Phantom-GRAPE (accelerated gravity routine on a machine-language

level)– Cooling function for 10–108 K from Spaans & Norman (1997)– Heating by uniform FUV radiation (1–30x solar neighborhood values)– Star Formation: Spawns collisionless particles when

• n > nth

• T < Tth

• Converging flow• No heat increase by nearby SNs

– SN feedback in forms of thermal energy addition

• Galactic Potential– Power-law bulge with a m=2 bar

– Miyamoto-Nagai disk

• Performed a series of simulations for the central ±1.2 kpc region for up to 700 Myr.

)0( v

)0( dQ

2cos)(cos1 2222

75.1

00 Pbr

r

ASURA Simulations of SF in Galactic Disks

• Saitoh et al. (2008) successfully reproduced the complex structure of the gas disk.• Found that nth is the most important SF criterion, but SFE is not.

)0( v

)0( dQ

2cos)(cos1 2222

75.1

00 Pbr

r

nth

=100cm-3

Tth =100 KTcut=10 KC* =0.033

nth =0.1 cm-

3

Tth =15000 KTcut=10000 KC* =0.033

nth =0.1 cm-

3

Tth =15000 KTcut=10 KC* =0.033

nth

=100cm-3

Tth =100 KTcut=10 KC* =0.5

N=106

mSPH=3500 M⊙

ε=10 pc

dyn

gas

tC

dt

d *

*

SF in the CMZ : Standard Run (Run #1)

N=100kMgas=5x107 M⊙

mSPH=500 M⊙

ε=3 pc

nth =100 cm-3

Tth =100 KTcut=10 KC* =0.033

Gas Influx and Star Formation Rates

Yusef-Zadeh+(2009)

Equilibrium SFR ~ 0.04 M⊙/yrEquilibrium gas mass within ±400 pc

~ 0.3-0.5 M0

→ Equilibrium SFR for the observed gas mass within ±400 pc ~ 0.1 M⊙/yr

GC

Schmidt–Kenni-cutt with slope 1.4

Yusef-Zadeh et al. (2009)

The l-v Diagram

Rodriguez-Fernandez & Comes (2008)

Inclination of the Disk (Standard Potential)

Jeon, Kim & Ann (2009)

Disk Evolution in Bar-like Triaxial Potential

Inclination of the Disk (Thinner disk potential)

Christopher et al. (2005)

Circumnuclear Disk in HCN

Response of the inner disk to the bar-like triaxial potential may be responsible for the apparent in-clination of the CND.

SFR Dependence on the SF Criteria

nth =1000 cm-3

nth =1000 cm-3

G0 =10

C* =0.01

C* =0.1

nth =100 cm-3

Tth =100 KTcut=10 KC* =0.033

Standard Parame-ters

SFR Dependence on the Bar Elongation

b22=0.1

2cos)(cos1 2222

75.1

00 Pbr

r

b22=0.14

b22=0.07

SFR Dependence on the Initial Cloud Distribution

Summary

• Main results from the first hydrodynamic simulations for the CMZ with SF/feedback/self-gravity are

– SF takes place mostly in the inner ±200 pc.– Obtained SFR, ~0.1 M⊙/yr, is consistent with recent YSO observations.

• These support the suggested connection between the CMZ and the stellar nuclear bulge.

– (Obtained SFR) x (Lifetime of the Galaxy) ~ 109 M⊙.

– Consistent with the current NB mass, 1-2x109 M⊙.

• Triaxial bulge potential can cause the CND to be inclined from the Galactic plane, and may also be responsible for the inclined stellar disks in the cen-tral parsec.