Transcript OUTFLOW INFALL AND ROTATION IN HIGH

Outflow, infall, and rotation in
high-mass star forming regions
Riccardo Cesaroni
Osservatorio Astrofisico di Arcetri
1) High-mass vs low-mass: the dividing line
2) The formation of high-mass stars:
accretion vs coalescence
3) Observations: infall, outflows, and disks
4) Kinematical evidence supports accretion
Low-mass vs High-mass
Theory (Shu et al. 1987): star formation from
inside-out collapse onto protostar
Two relevant timescales:
accretion  tacc = M*/(dM/dt)
contraction  tKH = GM*/R*L*
 Lowmass (< 8 MO): tacc < tKH
 Highmass (> 8 MO): tacc > tKH  accretion on
ZAMS
PROBLEM:
High-mass stars “switch on” still accreting 
 radiation pressure stops accretion 
 stars > 8 MO cannot form!?
SOLUTIONS
Yorke (2003): Kdust< Kcrit  M*/L*
1) “Reduce’’ L*: non-spherical accretion
2) “Increase’’ M*: large accretion rates
3) Reduce Kdust: large grains (coalescence of lower
mass stars)
Possible models
• (Non-spherical) accretion: Behrend &
Maeder (2001); Yorke & Sonnhalter
(2002); Tan & McKee (2003)
ram pressure > radiation pressure
• Coalescence: Bonnell et al. (2004) many
low-mass stars merge into one massive
star
Implications & predictions
• Accretion
 infall  disks  outflows
 isolated star formation possible
 massive stars form at cluster centre:
(dM/dt)  FWHM3 (Shu et al. 1987)
 massive stars form with lower mass stars:
t  M*1/4 (Tan & Mc Kee 2003)
Implications & predictions
• Coalescence
 infall, low-mass disks, multiple outflows
 isolated star formation impossible
 massive stars form at cluster centre:
large n* (108 */pc3 !)  many collisions
 massive stars form after lower mass stars
 best
discriminant between models:
kinematics of molecular gas
 infall: accretion  large accretion rate
coalescence  small(?) accretion rate
 outflow: accretion  single massive flow
coalescence  multiple low-mass flows
 rotation: infall+ang. mom. conservation 
 rotating disks (“only’’ in accretion model)
 best
discriminant between models:
kinematics of molecular gas
 infall: accretion  large accretion rate
coalescence  small(?) accretion rate
 outflow: accretion  single massive flow
coalescence  multiple low-mass flows
 rotation: infall+ang. mom. conservation 
 rotating disks (“only’’ in accretion model)
 best
discriminant between models:
kinematics of molecular gas
 infall: accretion  large accretion rate
coalescence  small(?) accretion rate
 outflow: accretion  single massive flow
coalescence  multiple low-mass flows
 rotation: infall+ang. mom. conservation 
 rotating disks (“only’’ in accretion model)
 best
discriminant between models:
kinematics of molecular gas
 infall: accretion  large accretion rate
coalescence  small(?) accretion rate
 outflow: accretion  single massive flow
coalescence  multiple low-mass flows
 rotation: infall+ang. mom. conservation 
 rotating disks (“only’’ in accretion model)
Discriminating between models:
observations
• Observational problems:





IMF  high-mass stars are rare
formation in clusters  confusion
rapid evolution: tacc=20 MO /10-3MOyr-1=2 104yr
large distance: >300 pc, typically a few kpc
parental environment profoundly altered
• Advantage:
 very luminous (cont. & line) and rich (molecules)!
High-mass star
forming region
0.5 pc
NIR J+H+K
10 pc
G9.62+0.19
G9.62+0.19
350 micron
0.5 pc
Hunter et al. (2000)
Testi et al.
Cesaroni et al.
Infall
Difficult to reveal: Vff 
• direct evidence:
R 0.5
 red-shifted (self)absorption: ambiguous…
 position-velocity plots/channel maps
• indirect evidence:
 lack of support: Mgas > Mvir
 model fit to SED
 dM/dt=10-3—10-2 MOyr-1  accretion possible
 insufficient resolution: infall on single star?
Infall
Difficult to reveal: Vff 
• direct evidence:
R 0.5
 red-shifted (self)absorption: ambiguous…
 position-velocity plots/channel maps
• indirect evidence:
 lack of support: Mgas > Mvir
 model fit to SED
 dM/dt=10-3—10-2 MOyr-1  accretion possible
 insufficient resolution: infall on single star?
Infall
Difficult to reveal: Vff 
• direct evidence:
R 0.5
 red-shifted (self)absorption: ambiguous…
 position-velocity plots/channel maps
• indirect evidence:
 lack of support: Mgas > Mvir
 model fit to SED
 dM/dt=10-3—10-2 MOyr-1  accretion possible
 insufficient resolution: infall on single star?
Outflow
Easy to detect even with low angular resolution
• single-dish (>10” i.e. >0.5 pc) CO surveys of
UCHIIs, IRAS sources, masers (Shepherd &
Churchwell 1996; Zhang et al. 2001; Beuther et
al. 2002, etc.), H2 (shocked) 2.2m emission
 outflows in high-mass stars do exist
 typical parameters: 1 pc, 5—5000 MO ,
10-4—10-2 MO yr-1, dM/dt  L0.7
BUT… are these from the most massive (proto)star?
CO(2-1) outflow
&
1mm continuum
Beuther et al. (2002)
Outflow
Easy to detect even with low angular resolution
• single-dish (>10” i.e. >0.5 pc) CO surveys of
UCHIIs, IRAS sources, masers (Shepherd &
Churchwell 1996; Zhang et al. 2001; Beuther et
al. 2002, etc.), H2 (shocked) 2.2m emission
 outflows in high-mass stars do exist
 typical parms.: 1 pc, 5-5000 MO , 10-4-10-2 MO yr-1
 dM/dt  L0.7  continuity from low- to high-mass
BUT… is outflow from one massive (proto)star?
CO outflows in YSOs
Churchwell (2002)
dM/dt  L0.7
Outflow
Easy to detect even with low angular resolution
• single-dish (>10” i.e. >0.5 pc) CO surveys of
UCHIIs, IRAS sources, masers (Shepherd &
Churchwell 1996; Zhang et al. 2001; Beuther et
al. 2002, etc.), H2 (shocked) 2.2m emission
 outflows in high-mass stars do exist
 typical parms.: 1 pc, 5-5000 MO , 10-4-10-2 MO yr-1
 dM/dt  L0.7  continuity from low- to high-mass
BUT… is outflow from one massive (proto)star?
• interferometric (>1” i.e. 0.05 pc) observations
of selected targets in CO, HCO+, SiO, etc.
(PdBI, OVRO, BIMA, NMA)
 “single-dish’’ outflows resolved into (massive &
collimated) multiple outflows (Beuther et al. 2002)
 precession of outflow complicate interpretation
(Shepherd et al. 2000; Gibb et al. 2003)
 powering source difficult to identify
 infall/outflow insufficient to prove model
CO(2-1)
& mm cont.
Beuther et al.
(2002)
single-dish
(12’’ beam)
05358+3543
Beuther et al.
(2003)
interferometer
(4’’ beam)
• interferometric (>1” i.e. 0.05 pc) observations
of selected targets in CO, HCO+, SiO, etc.
(PdBI, OVRO, BIMA, NMA)
 “single-dish’’ outflows resolved into (massive &
collimated) multiple outflows (Beuther et al. 2002)
 precession of outflow complicate interpretation
(Shepherd et al. 2000; Gibb et al. 2003)
 powering source difficult to identify
 infall/outflow insufficient to prove model
IRAS 20126+4104
blue lobe
Shepherd et al. (2000)
H2 knots
red lobe
IRAS 20126+4104
jet in H2 line
IRAS 20126+4104
Cesaroni et al. (in prep.)
• interferometric (>1” i.e. 0.05 pc) observations
of selected targets in CO, HCO+, SiO, etc.
(PdBI, OVRO, BIMA, NMA)
 “single-dish’’ outflows resolved into (massive &
collimated) multiple outflows (Beuther et al. 2002)
 precession of outflow complicate interpretation
(Shepherd et al. 2000; Gibb et al. 2003)
 powering source difficult to identify
 infall/outflow insufficient to prove scenario
Disks
Circumstellar accretion disks predicted only by
accretion model! Any evidence?
• Large scale (1 pc)
rotating clumps seen in medium density tracers
e.g. NH3 in G35.2-0.74 (Little et al. 1985)
• Small scale (<0.1 pc)
many claims of rotating “disks’’…
Disks
Circumstellar accretion disks predicted only by
accretion model! Any evidence?
• Large scale (1 pc)
rotating clumps seen in medium density tracers
e.g. NH3 in G35.2-0.74 (Little et al. 1985)
• Small scale (<0.1 pc)
many claims of rotating “disks’’…
CH3OH masers ATCA, Ellingsen et al., Walsh et al.
EVN
Minier et al.
OH masers
Merlin outflow sources
Cohen et al. (2003)
SiO & H2O
VLA, e.g. Orion source I
masers
VLBA Greenhill
NIR, mm & cm BIMA, jets/outflows in massive stars
continuum
VLA
Hoare et al., Gibb et al.
NH3, C18O, CS, PdBI, UC HIIs, Hot Cores
C34S, CH3CN OVRO, Keto et al., Cesaroni et al.,
BIMA, Zhang et al., …
NMA
• CH3OH masers: stellar mass too low; H2 jets
parallel to CH3OH spots (De Buizer 2003)
• SiO & H2O masers: outflow or disk ?
• NIR-cm cont.: confusion between disk and wind
emission
• Molecular lines: kinematical signature of disk &
outflow
CH3OH masers
W48
Minier et al. (2000)
M*=6 MO
H2O masers
Cep A HW2
Torrelles et al. (1996)
• CH3OH masers: stellar mass too low; H2 jets
parallel to CH3OH spots (De Buizer 2003)
• SiO & H2O masers: outflow or disk?
• NIR-cm cont.: confusion between disk and wind
emission?
• Molecular lines: kinematical signature of disk &
outflow
outflow
core
disk
outflow
G192.16-3.82
Shepherd & Kurtz (1999)
2.6mm cont.
disk
CO outflow
G192.16-3.82
Shepherd & Kurtz (1999)
3.6cm cont. & H2O masers
NGC7538S Sandell et al. (2003)
IRAS 20126+4104
M*=7 et
Mal.;
O Moscadelli et al.
Cesaroni
H2O masers prop. motions
Disks & Tori
B stars
O stars
L
Mdisk
Ddisk
M*
(LO) (MO)
(AU)
(MO)
IRAS20126 104
4
1600
7
G192.16
3 103
15
1000
6-10
NGC7538S 104 100-400 30000
40?
G24.78 (3) 7 105 80-250 4000-8000 20…
G29.96
9 104 300
14000
G31.41
3 105 490
16000
-
Gibb et al. (2002)
Olmi et al. (2003)
Beltran et al. (2004)
Beltran et al. (2004)
Beltran et al. (2004)
Gibb et al. (2002)
Olmi et al. (2003)
Beltran et al. (2004)
1200 AU
Hofner priv comm.
Disks & Tori
B stars
O stars
L
Mdisk
Ddisk
M*
(LO) (MO)
(AU)
(MO)
IRAS20126 104
4
1600
7
G192.16
3 103
15
1000
6-10
NGC7538S 104 100-400 30000
40
G24.78 (3) 7 105 80-250 4000-8000 20…
G29.96
9 104 300
14000
G31.41
3 105 490
16000
-
Results
• “Circumcluster’’ (massive) tori in O (proto)stars
• Circumstellar (Keplerian) disks in early-B
(proto)stars
 Are disks in O (proto)stars short lived?
Assuming (dM/dt)acc  (dM/dt)outflow
and Mdisk  M*
disk life time
B stars
O stars
disk ioniz.
&
accr. rates
B stars
O stars
Conclusions
• Circumstellar (Keplerian) disks in early-B
(proto)stars  disk accretion likely
• Circumcluster (unstable) tori in O
(proto)stars  large accretion rates make
them long-lived
ACCRETION SCENARIO MORE LIKELY
G45.47+0.05
red-shifted
absorption
NH3(2,2)
Hofner et al. (1999)
systemic
velocity
blue-shifted
emission
Fontani et al. (2001)
n  R-2.6
CO outflows
in YSOs
Beuther et al. (2002)