Proto-Neutron Star Winds with Magnetic Fields & Rotation (and - - PowerPoint PPT Presentation

Proto-Neutron Star Winds with Magnetic Fields & Rotation

(and other non-traditional r-process sites)

Brian Metzger

NASA Einstein Fellow, Princeton University with Todd Thompson (OSU), Eliot Quataert (UC Berkeley), Tony Piro (UC Berkeley)

Metzger, Thompson & Quataert 2007, 2008 Metzger, Piro & Quataert 2008, 2009

EMMI r-Process Workshop - July 16, 2010

Astrophysical R-Process Sites

1) Low Entropy < 1 kb nuc-1 , Ye ~ 0.1

⇒ Neutron Star Mergers (Dynamical Ejecta)

(Lattimer & Schramm 1974, 76; Eichler et al.1989; Freiburghaus et al. 1999; see talk by Goriely)

2) High Entropy > 102 kb nuc-1 , 0.4 < Ye < 0.5

⇒ Neutrino-Driven Wind

• Proto-Neutron Stars in Core Collapse Supernovae
• Hyper-Accreting Disks (Collapsars & NS Mergers)

3) “Intermediate” Entropy ~ 10 kb nuc-1, Ye ~ 0.2-0.4

⇒ Thermonuclear-Driven Winds

• Hyper-Accreting Disks (Late Times)

τn << τβ ⇒ NS or BH accretion disk

Astrophysical R-Process Sites

1) Low Entropy < 1 kb nuc-1 , Ye ~ 0.1

⇒ Neutron Star Mergers (Dynamical Ejecta)

(Lattimer & Schramm 1974, 76; Eichler et al.1989; Freiburghaus et al. 1999; see talk by Goriely)

2) High Entropy > 102 kb nuc-1 , 0.4 < Ye < 0.5

⇒ Neutrino-Driven Wind

• Proto-Neutron Stars in Core Collapse Supernovae
• Hyper-Accreting Disks (Collapsars & NS Mergers)

3) “Intermediate” Entropy ~ 10 kb nuc-1, Ye ~ 0.2-0.4

⇒ Thermonuclear-Driven Winds

• Hyper-Accreting Disks (Late Times)

τn << τβ ⇒ NS or BH accretion disk

Proto-Neutron Star Winds

(Duncan et al. 1986; Takahashi et al. 1994; Burrows et al. 1995; Qian & Woosley 1996)

Neutrinos Heat Proto-NS Atmosphere (e.g. νe + n ⇒ p + e-)

⇒ Drives Thermal Wind Behind Outgoing Supernova Shock Burrows, Hayes & Fryxell 1995

GMNSmn RNS ~ 200 MeV

• ~ 10 20 MeV
• Grav. Binding Energy

>> Avg. Neutrino Energy

⇒

(1) Final Ye ~ 0.5 set by competition btw (2) High Entropy

e +n &

e +p

S = dQ T

R NS

• > GMNSmn

RNS

• TNS

Conditions for 2nd/3rd Peak R-Process

(e.g. Meyer & Brown 1997; Hoffman et al. 1997)

• Low Electron Fraction Ye
• High Entropy S
• Short Dynamical Timescale τdyn

(fast expansion @ α formation) } α-rich Freeze Out

4He(αn)9Be(αn)12C

Bottleneck

(e.g. Woosley & Hoffman 1992)

Hoffman et al. 1997

Third-Peak Threshold:

S3 dyn > f (Ye)

R-Process Fail

(Qian & Woosley 1996; Hoffman et al. 1997; Otsuki et al. 2000; Thompson et al. 2001; Hudepohl et al. 2010) Thompson et al. 2001

⇒ 87Rb, 88Sr, 89Y, 90Zr

(see talks by Roberts and Arcones)

Path through S-τdyn Space

Dynamical Time

Alternative Ideas

Entropy

R-Process Fail

(Qian & Woosley 1996; Hoffman et al. 1997; Otsuki et al. 2000; Thompson et al. 2001; Hudepohl et al. 2010) Thompson et al. 2001

⇒ 87Rb, 88Sr, 89Y, 90Zr

(see talks by Roberts and Arcones)

Path through S-τdyn Space

Dynamical Time

• Very Massive NSs

(e.g. Cardall & Fuller 1997; Thompson et al. 2001)

• Neutrino Oscillations

(e.g. Duan et al. 2010; but see Hudepohl et al. 2010)

• Wind-SN Ejecta Interaction

(e.g. Wanajo et al. 2001; Arcones et al. 2007)

• Wave Heating

– Acoustic (Burrows et al. 2006) – MHD (Suzuki & Nagataki 2005; Metzger et al. 2007)

• Electron Capture SNe

(Ning et al. 2007; but see Hoffman et al. 2008)

• Magnetic Fields & Rotation

Alternative Ideas

Entropy

Magnetars

(Thompson & Duncan 1995; Kouveliotou et al. 1998; Woods & Thompson 2006)

Time (seconds) Counts per second SGR1806-20 Giant Flare December 4, 2004

(courtesy: A. Watts & T. Strohmayer)

• Soft Gamma-Ray Repeaters & Anomalous X-ray Pulsars
• Surface magnetic fields Bdip ~ 1014-1015 G
• Rapid rotation at birth as source of strong fields?

(e.g. α-Ω dynamo or magneto-rotational instability; Duncan & Thompson 1992; Akiyama et al. 2003)

• Fairly common (at least ~10% of neutron stars are born magnetars)

Effects of Strong Magnetic Fields

• Microphysics (EOS, Neutrino Heating & Cooling)

– Important for B > 1016 G (Duan & Qian 2005)

• Closed Zone Heating / Eruptions (Thompson 2003)
• Magneto-Centrifugal Outflows (Weber & Davis 1967)

“Helmet - Streamer”

Ω

Effects of Strong Magnetic Fields

• Microphysics (EOS, Neutrino Heating & Cooling)

– Important for B > 1016 G (Duan & Qian 2005)

• Closed Zone Heating / Eruptions (Thompson 2003)
• Magneto-Centrifugal Outflows (Weber & Davis 1967)

“Helmet - Streamer”

Ω

B2 8 > 12 vr

2

Outflow Co-Rotates with Neutron Star while

1) Magnetic Acceleration (lower τdyn) 2) Enhanced Mass Loss 3) Early Weak Freeze Out (lower Ye) ⇒ RA Rα Rheat

Top View

Proto-Neutron Star Winds with Magnetic Fields & Rotation

(BDM, Thompson & Quataert 2007, 2008) INPUT:

• NS Mass, Radius, Rotation Rate, Surface Field Strength
• Neutrino Luminosities & Spectrum
• Free Outer Boundary

OUTPUT:

• Steady-State Radial Wind Profile: ρ, T, vr , vφ , Bφ
• Captures 3 MHD Critical Points
• Eigenvalues: Mass, Angular Momentum, & Energy Loss Rate

+ Ω

“Equatorial” Flux Tube

“Normal” Thermally-Driven Wind

˙ M ~ 104 M s1; S ~ 70 kb nuc-1; dyn ~ 25 ms L e ~ 8 1051 ergs s-1; B0 =1013 G; P =100 ms

(sound speed)

Rs RA

Magnetically-Driven Wind

˙ M ~ 3103 M s1; S ~ 20 kb nuc-1; dyn ~ 0.5 ms L e ~ 8 1051 ergs s-1; B0 =1015 G; P =1.2 ms RA Rs

Dynamical Timescale BDM et al. 2007

1015 G 1013 G 1014 G

Bdip > 4 1013L ,52

5/ 6 ,10 5/ 3P ms 2 G

τdyn decreased if

Dynamical Timescale S3/τdyn BDM et al. 2007

1015 G 1013 G 1014 G

Bdip > 4 1013L ,52

5/ 6 ,10 5/ 3P ms 2 G

But…. MHD acceleration also decreases entropy τdyn decreased if

RA Rα Rheat Low τdyn Low S High τdyn High S Low τdyn High S ?

Latitude-Dependent Wind Properties?

RA Rα Rheat Low τdyn Low S High τdyn High S Low τdyn High S ?

Latitude-Dependent Wind Properties?

RYe

0.7 ms 1.6 ms

To appreciably reduce Ye

a ⇔

enhancement in ˙ M by a factor > GMNSmn RNS

~ 10

Electron Fraction (& Mass Loss Rate)

Electron Fraction

˙ M

BDM et al. 2008

Binary Compact Object Mergers

NS NS BH BH NS NS NS NS

˙ N

merge ~ 10-5 10-4 yr-1

Known Galactic NS-NS Binaries Known Galactic NS-NS Binaries

Hulse-Taylor Hulse-Taylor Pulsar Pulsar T Tmerge

merge

= 300 = 300 Myr Myr

( (Kalogera Kalogera et al. 2004) et al. 2004)

Credit: M. Shibata (U Tokyo) Credit: M. Shibata (U Tokyo)

Remnant Accretion Disk

• Disk Mass ~ 10-3 - 0.1 M & Size ~ 10-100 km
• Midplane Hot (T > MeV), Dense, & Neutron Rich
• Cooling via Neutrinos: (τγ >>1, τν ~ 0.01-100 )

˙ M ~ 102 10M• s-1 Short GRB Central Engine?

Lee et al. (2004) Lee et al. (2004)

Accretion Rate Accretion Rate

Magnetized Accretion Disks

Hawley & Hawley & Balbus Balbus (2002) (2002)

J Md (GMBHRd )1/ 2 Rd J Md

2

BH BH

MHD Turbulence MHD Turbulence Redistributes Redistributes Angular Angular Momentum Momentum

Accretion ⇔ Expansion to Larger Radii

Md,0 = 0.1 M, rd,0 = 30 km, α = 0.3

1D Height-Integrated Disk Evolution

Local Disk Mass πΣr2 (M)

Angular Momentum Angular Momentum Transport Transport (Viscous Spreading) (Viscous Spreading) Entropy Entropy Nuclear Composition Nuclear Composition

Heating Heating Cooling Cooling

1) High Thick Disk: H ~ R

• Optically Thick; Matter Accretes Before Cooling
• Neutrino-sphere Deeper (and Hotter) than Neutrino-sphere
• ν-Driven Wind with low Ye ⇒ r-Process? (e.g. Surman et al. 2006, 2008)

2) Neutrino-Cooled Thin Disk: H ~ 0.2 R

• Optically Thin; Neutrino Luminosity Lν ~ 0.1 c2
• ν-Driven Wind with Ye > 0.5 ⇒ νp-Process? (Kizivat et al. 2010)

3) Low Thick Disk: H ~ R

• Low Temperature ⇒ inefficient neutrino cooling & weak freeze-out

Three Accretion Phases

(Metzger, Piro & Quataert 2008)

˙ M ˙ M

˙ M

1 2 3

• e

e

Late Time Winds Late Time Winds

• Recombination: n + p ⇒ He
• Thick Disks Marginally Bound

After t ~ After t ~ 0.1-1 seconds, R ~ 500 km & 0.1-1 seconds, R ~ 500 km & T < 1 T < 1 MeV MeV

E EBIND

BIND ~ GM

~ GMBH

BHm

mn

n/2R ~

/2R ~ 3 3 MeV MeV nucleon nucleon-1

• 1

Δ ΔE ENUC

NUC ~

~ 7 7 MeV MeV nucleon nucleon-1

• 1

Late Time Winds Late Time Winds

• Recombination: n + p ⇒ He
• Thick Disks Marginally Bound

⇒

}

After t ~ After t ~ 0.1-1 seconds, R ~ 500 km & 0.1-1 seconds, R ~ 500 km & T < 1 T < 1 MeV MeV

E EBIND

BIND ~ GM

~ GMBH

BHm

mn

n/2R ~

/2R ~ 3 3 MeV MeV nucleon nucleon-1

• 1

Δ ΔE ENUC

NUC ~

~ 7 7 MeV MeV nucleon nucleon-1

• 1

~20-50% of Initial Disk ~20-50% of Initial Disk Ejected Back into Space! Ejected Back into Space!

BH

Powerful Winds Blow Apart Disk

(see also Lee et al. 2009)

Neutron-Rich Freeze- Out Composition

(Metzger et al. 2008, 2009)

Mej ~ Mdisk/3 ~ 10-3 - 10-2 M

Thickening / Freeze-Out Begins at the Outer Disk and Moves Inwards

Ye

eq

Ye

Weak Interactions Weak Interactions Drive Y Drive Ye

e

⇒ ⇒ Y Ye

e eq eq

Until Freeze-Out Until Freeze-Out

Neutron-Rich F Freeze-Out (“Little Bang”)

e + p e + n e+ + n

e + p

Robust Neutron-Rich Freeze-Out

M per bin

Mtot = 0.02 M Mtot = 0.02 M

M per bin

Mtot= 2 10-3 M

• Mej ~ 10-3-10-2 M with Ye ~

0.1-0.3 and S ~ 10 kb nuc-1

• Galactic production rate:

M0 = 0.1 M, α = 0.3 M0 = 0.1 M, α = 0.03 M0 = 0.01 M, α = 0.3

˙ M

r ~ 106 M yr-1

˙ N

NSNS

104 yr1

• Mej

102 M

Conclusions

• Proto-NS Winds with Magnetic Fields & Rotation

1. At least ~10% of Galactic NSs are born “Magnetars” 2. Magnetic acceleration decreases τdyn (good) , but also reduces asymptotic entropy (bad). Lower Ye possible only for extremely rapid rotation (P ~ 1 ms) 3. No r-process succes yet, but future work must address more realistic magnetospheric geometry (e.g. latitude dependence). 4. Future work: place nucleosynthesis constraints on distribution of B & Ω at birth

• Binary Neutron Star Mergers

1. Disk formation is a common feature of NS mergers 2. Early disk evolution ⇒ neutrino-heated winds as a possible site for r-process and νp-process. (Relatively low total ejecta mass.) 3. As disk spreads, neutrino cooling shuts off and α-particles form ⇒ Thermo-nuclear-powered outflows eject ~10-3-10-2 M with Ye ~ 0.1-0.4 and S ~ 10 kb baryon-1