the role of massive stars in the turbulent infancy of galactic ...the role of massive stars in the...

The role of massive stars in the turbulent infancy of

Galactic globular clusters:

Nucleosynthesis, superbubble dynamics and timeline

Corinne CharbonnelGeneva Observatory & IRAP CNRS

Krause, Charbonnel, Decressin, Prantzos, Meynet, Diehl (12, A&A 546, L5)

Krause, Charbonnel, Decressin, Prantzos, Meynet (13, A&A 552, A121 )

C.Charbonnel (Eurogenesis –June 2013)

Lind, Charbonnel, Decressin, Primas, Grundahl, Asplund (2011)NGC 6397


C-N, O-Na, Mg-Al anticorrelations

Prantzos, Charbonnel & Iliadis (07)

T ≥ 15 x 106 K : CN

T ≥ 25 x 106 K : CNO, 22Ne 23Na

T ≥ 40 x 106 K : CNO, 20Ne 23Na25,26 Mg 26 Al, 27Al

T ≥ 70 x 106 K : 24Mg (and 25, 26 Mg) 26 Al, 27Al

H-burning through CNO, NeNa, MgAl at T ~ 72 to 78 MK


Long-lived low-mass stars

The observed patterns pre-existed

in the material out of which

the presently surviving stars formed

Implies pollution of the intra-cluster gas

by a first generation of more massive

rapidly evolving stars

in which H burns at ~ 72-78MK!

Formation of second generation stars





Long-lived low-mass stars









Possible « polluters » :

Massive stars


Possible « polluters » :

Massive stars and massive AGBs






H-burning through CNO, NeNa, MgAl at T ~ 72 to 78 MKPrantzos, Charbonnel & Iliadis (07)



C-N, O-Na, Mg-Al, F-Na, Li-Na anticorrelations

H-burning devoid of light elements

(LiBeBF)

H-burning ashes mixed with pristine gas}2d generation

Smith et al. (2005)

NGC 6397

Lind, Primas, Charbonnel, Grundahl & Asplund (09)

Turnoff stars

M4

30 % of pristine gas

& 70 % of stellar ejectaC.Charbonnel (Eurogenesis –June 2013)


[(C+N+O)] ~ constant within experimental errors

[Fe/H] constant

H-burning through CNO, NeNa, MgAl

H-burning ashes mixed with pristine gas

No recycling of He-burning products

}2d generation



[(C+N+O)] ~ constant within experimental errors

[Fe/H] constant

}2d generationH-burning through CNO, NeNa, MgAlH-burning ashes mixed with pristine gasNo recycling of He-burning products

No recycling of supernovae ejecta,

except in some rare (most massive) cases (e.g., Ω Cen or M22)


FRMS

≥ 25M

Massive

AGB

~ 5 – 8M

If 1G polluters follow a standard IMF

(Salpeter or Kroupa)

today’s ratio 1G:2G should be ~ 90:10Decressin et al. (07), D’Ercole et al. (08)

Observed ratio 1G:2G ~ 30:70Carretta et al. (10)

Flat polluter IMF

X ~ 0.6 - 0.8 (≥ 20 M

)

X < 0.45 (4 - 9 M

)

X < -0.65 (5 - 6.5 M

)

Prantzos & Charbonnel (06)

See also Smith & Norris (82, C-N data)

D’Antona & Caloi (04)

Downing & Sills (07)

Standard IMF + Much higher initial GC mass

Important loss of 1st generation low-mass stars

Prantzos & Charbonnel (06), Decressin et al. (07)

D’Ercole et al. (08, 10), Vesperini et al. (10)

Schaerer & Charbonnel (11), Conroy (12)

For Salpeter polluter IMF

Mini ~ 8-10 x Mobs if e2GLL = 0

Prantzos & Charbonnel (06), D’Ercole +(08), Carretta+(10)

Mini ~ 25 x Mobs if e2GLL = 0.64 Schaerer & Charbonnel (10)

Minimum values (conservative assumptions) !!!!

e2GLL = 0

e2GLL = 0.41(1% of 2d gen.* in

the halo)

e2GLL = 0.64 (2.5% of 2d gen.* in

the halo)Martell & Grebel (10), Carretta+(10)

Second generation stars: only ≤0.8M

100% of relevant polluters’ ejecta are used to form stars


Important loss of 1st generation low-mass stars (> 95%)

Schaerer & Charbonnel (10)


For Salpeter polluter IMF

Mini ~ 8-10 x Mobs if e2GLL = 0

Prantzos & Charbonnel (06), D’Ercole +(08), Carretta+(10)

Mini ~ 25 x Mobs if e2GLL = 0.64 Schaerer & Charbonnel (10)

Minimum values (conservative assumptions) !!!!

with MGCnow / M*halo ~ 2%

from Freeman & Bland-Hawthorn (02)

+ halo stars produced from destroyed BGCs and other lower mass clusters

that formed at high redshift

(e.g. Hut & Djorgovski 1992; Parmentier & Gilmore 2007; Bell et al. 2008; Boley et al. 09)

Contribution to the Galactic halo of the

~ 180 MW GCs

Meje ≥ 3 x Mobs

≥ 6 % M*halo if e2G

LL = 0

Meje ≥ 10 x Mobs

≥ 20% M*halo if e2G

LL = 0.64



Important loss of 1st generation low-mass stars (> 95%)


Multiple stellar generations in GCs

Towards a global scenario


Krause, Charbonnel, Decressin, Prantzos, Meynet, Diehl (12, A&A 546, L5)

Krause, Charbonnel, Decressin, Prantzos, Meynet (13, A&A 552, A121 )

Standard case :

NGC 6752 (today’s M ~ 3 x 105 M

)

Proto-GC cloud of Mtot = 9 x 106 M

Mass-segregated cluster (Hillenbrand 97; de Grijs+02; Klessen 01; Bonnel+01)

N-body models (Decressin et al. 10)

Plummer profile for mass distribution (eg Baumgardt+08)

Half-mass radius r1/2 = 3pc

Average (gas) ~ 106mp cm-3

SFE = 1/3

Salpeter IMF for 1G stars with Mi>0.8M

~ 5700 massive stars between 25 and 120 M

log-normal IMF for 1&2G low-mass stars

Stellar parameters (lifetimes, winds, …) (Decressin et al. 07b)


Important loss of 1G low-mass stars

Cluster is impacted by the stellar winds


Stellar winds unable to lift any

noteworthy amount of gas out of the GC

on a relevant timescale

Cluster is impacted by the stellar winds


Formation of hot, overlaping bubblesaround massive starsSpongy structure for ISM


Formation of hot, overlaping bubblesaround massive stars

High ultraviolet radiation Conroy & Spergel (11)

Lyman-Werner photons

QLW(M) = 7 x 1043 (M/M

)2.9 s-1

Photodissociation of molecular H

Tgaz ~ 100K

No « usual » star formation

(confirm Conroy & Spergel 11)

Formation in the decretion discs

of individual massive stars

(Decressin et al. 07)

Stellar evolutionSpongy structure for ISM

Wind of Fast Rotating Massive Stars scenario

Higher rotational velocities in young massive clusters

than in the field

(Huang & Gies 06; Strom et al. 05; Dufton et al. 06)

Transport of angular momentum and chemicals

by meridional circulation and shear turbulenceZahn (92), Maeder & Zahn (98), Meynet & Maeder (00)

Same physics successfully applied to

Massive stars : HeBCN anomalies (Maeder & Meynet 00)

Intermediate-mass stars : Primary N production at low Z (Chiappini et al. 06)

Low-mass stars : Hot side of the Li dip, Li in subgiants (Charbonnel & Talon 99,Palacios et al.03, Pasquini et al. 04)

Prantzos & Charbonnel (06), Decressin et al. (07a,b,09,10)

Schaerer & Charbonnel (10), Krause et al. (12a,b)


Early main sequence

Meridional circulation and turbulence extract angular momentum

from the fast-rotating core

The star reaches the break-up velocity (Centrif. acc. compensates gravity)

Equatorial matter released in a keplerian orbit

Formation of a slow outflowing disk (Be stars)

Green : pristine [O/Na]=0.6

Blue : H-burning

products [O/Na]=-2

Decressin et al. (07)C.Charbonnel (Eurogenesis –June 2013)

Main sequence and LBV phase at break-up :

Transport of H-burning-products from the core to the surface and disk

Green : pristine [0/Na]=0.6

Blue : H-burning

products [O/Na]=-2

Red : He-burning

products [O/Na]=3

Star formation in the “decretion disc”

Clumps or protostars

observed in the disk

of the Be star MWC 1080

(Wang et al. 07)

Rotation (mechan.wind)

M ~ 20M

Standard (rad.wind)

M ~ 1M

60 M

, Z = 5x10 -4

Decressin et al. (07) C.Charbonnel (Eurogenesis –June 2013)

After the LBV phase, the star moves away from break-up

The disk is disconnected from the star,

and the classical radiatively-driven fast winds ( ≥ 1000 km.sec-1) take over

No recycling of the stellar ejecta

of more advanced phases (He-burning products and metals)

Green : pristine [0/Na]=0.6

Blue : H-burning

products [O/Na]=-2

Red : He-burning

products [O/Na]=3

Decressin et al. (07) C.Charbonnel (Eurogenesis –June 2013)

Slow equatorial mass ejection

at critical rotation velocity

Total mass output by equatorial

mechanism

averaged over total ejection time

(main sequence and LBV stage)

and IMF:

104 M

/Myr

Equatorial mass ejection

Shadowing of the disc frees the equatorial

region from radiation pressure

Establishment of an accretion flow of

surrounding dense pristine gas

Time and orbit averaged Bondi accretion

rate

Accretion onto the discs

Viscous processes transport

material within

the disc

Disc fed both by stellar processed and

pristine material

Mixture of gas within the disk:

~ ½ pristine – ½ ejecta (on average)

Equatorial mass ejection

vs accretion

[O/Na]Observed abundance

ratios in individual

NGC 6752 stars(Carretta et al. 2005)

Theoretical distribution

histograms of the matter

composed

of the wind ejecta mixed

with 30% of pristine material,

in M

Decressin, Meynet, Charbonnel, Prantzos, Ekström (07)

Mixture of gas within the disk:

Lind, Primas, Charbonnel, Grundahl & Asplund (09)

C.Charbonnel (Eurogenesis –March 2013) C.Charbonnel (Eurogenesis –June 2013)

Self-gravitating discs

(mass similar to the central star) (eg Armitage 11)

Toomre criterion (Shu 92)

The disc reach the critical mass for

gravitational instability on timescale

of ≤ 106 yrs

Formation of 2G low-mass stars

Studied in the context of planet

formation (eg review by Kley & Nelson 12)

Very complex problem, lots of physics:

Transport/exchange of matter and angular

momentum, role and influence of disk self-gravity

and magnetohydrodynamic turbulence, …

More to come (Hennebelle, Teyssier, Fromang)

Gravitational instability

and star formation in the disk

Mass limit for stars to explode as SNe ?

M ≥ 25 M

may to turn to silent black holes(Portegies Zwart et al. 97; Ergma & van der Heuvel 98;

Kobulnicky & Skillman 97; Fryer 99; Belczynski et al. 12)


Stellar winds (from Decressin et al. 07b)

9-120M

SNe: 1051ergs each, =0.2 (efficiency parameter)

Power sources: Stellar winds, SNe, accretion onto stellar BH, NS

Fast gas expulsion and loss of 1st generation stars ?

Krause, Charbonnel, Decressin, Prantzos, Meynet & Diehl (12a) C.Charbonnel (Eurogenesis –June 2013)

Growth of the supperbubble via spherically symmetric thin shell approximationBrown, Burkert & Truran (91,95)

Shell momentum given by the applied forces:

p : bubble pressure

depends on energy injection law E(t) and η

g : gravitational acceleration

M : mass in the shell, v : shell velocity, A : surface area of the shell

Plummer profile for mass distribution (eg Baumgardt+08)

Fast gas expulsion and loss of 1st generation stars ?

Krause, Charbonnel, Decressin, Prantzos, Meynet & Diehl (12a)

Supperbubble


gravitational accelerationshell acceleration

(blue:

SNe phase Expanding superbubble, Rayleigh-Taylor instable

Shell fragments fall back (gas expulsion fails)

ICM develops strong SNe-driven turbulence

with rms-velocity ~ 50 km s-1 (< escape velocity)Kritsuk et al. 01, Krumholtz et al. 06


Turbulent ISM Formation of 2G stars stops

No accretion onto the remaining disks

nor on the dark remnants

Turbulence decays

Gas accretion on dark remnants


Rayleigh-Taylor scale

half-mass radiusbubble radius

escape speedat the current bubble radius

shell velocity

gravitational acceleration

shell acceleration (blue: 25M

3M

BH, accretion of local gas adds energy to the gas

at a rate of 20% of Eddington L

10-25M

1.5 M

neutron stars, contribute 20% of Eddington L


Rapid gas expulsion Turbulence decays

Gas accretion on dark remnants

Coherent onset of accretion of local ISM onto the stellar remnants

is required to expel cold gas and unbind 1st generation stars

Limiting initial cloud mass ~ 107 M

Observed intrinsic dispersion in Fe

from GIRAFFE spectra

vs visual total magnitude of 19 GGCs

Carretta et al. (09)

Ω Cen

M22

Marino et al. (11)

Da Costa et al. (09)

Krause, Charbonnel, Decressin, Prantzos, Meynet & Diehl (12)




Important role of the physics

of the intra-cluster medium

Need for multiD hydro simulations

Star formation in decretion disks around

massive stars

Dark remnant activation to evacuate

the bulk of gas and of 1G starsC.Charbonnel (Eurogenesis –June 2013)



Different IMF in different environments?

Were most halo stars ejected from very

dense stellar clusters? GAIA

Dominant contribution to reionize the IGM

at high z ?

GC formation in a cosmological context



the role of massive stars in the turbulent infancy of galactic ...the role of massive stars in the...

Documents