Supernovae, Nucleosynthesis, and Constraints on Chemical Evolution

Jim Truran

QuickTime™ and a

TIFF (Uncompressed) decompressorare needed to see this picture.

Astronomy and Astrophysics Enrico Fermi Institute University of Chicago andArgonne National Laboratory

Ringberg Workshop on Nuclear Astrophysics March 12, 2008

Tracers of Star Formation Histories

The heavy element content of the Universe at any point in its history reflects the integrated nucleosynthesis contributions from earlier stellar generations. We can use this knowledge effectively as a tool both to probe its dynamical and star formation histories and to constrain models of stellar and supernova nucleosynthesis.

How might we unravel this history?

Since distinctive abundance patterns are identified with the nucleosynthesis products of stars of different masses (and lifetimes), constraints on the early nucleosynthesis and star formation histories of the Cosmos will be contained in the spectra of halo stars and QSO absorption line systems, as a function of [Fe/H] or redshift.

“Cosmic” Abundances of the Elements

Massive Stars & SNe

r-process s-process

Nucleosynthesis Sites and Production Timescales

Massive stars (M > 10 M) and SNe II: synthesis of most of the nuclear species from oxygen through zinc, and of the r-process heavy elements ( < 108 years)

Red Giant Stars (1 < M < 10 M): synthesis of carbon

and of the heavy s-process elements ( > 109 years)

SNe Ia: synthesis of the 1/2-2/3 of the iron peak nuclei not produced by SNe II ( > 1.5-2 x109 years)

Type Ia Supernovae: Theory

“Standard model” (Hoyle & Fowler 1960):

SNe Ia are thermonuclear explosions of C+O white dwarf stars.

Evolution to criticality:

Accretion from a binary companion (Whelan and Iben 1973) leads to growth of the WD to the critical (Chandrasekhar) mass ( 1.4 solar masses).

After ~1000 years of thermonuclear “cooking”, a violent explosion is triggered at or near the center.

Complete incineration occurs within two seconds, leaving no compact remnant.

Light curve powered by radioactive decay of 56Ni. (Nickel mass ≈ 0.6 M.) Peak luminosity M(56Ni).

Type II Supernovae: Theory

“Standard model” (Hoyle & Fowler 1960):

SNe II are the product of the evolution of massive stars 10 < M < 100 M.

Evolution to criticality:

A succession of nuclear burning stages yield a layered compositional structure and a core dominated by 56Fe.

Collapse of the 56Fe core yields a neutron star.

The gravitational energy is released in the form of neutrinos, which interact with the overlying matter and drive explosion.

Remnants: Neutron star and black hole remnants are both possible SNe II remnants.

Nucleosynthesis contributions: elements from oxygen to iron (formed as 56Ni) and neutron capture products from krypton through uranium and thorium. (nucleosynthesis < 108 yrs) Production of ≈ 0.1 M of 56Fe as 56Ni.

Courtesy Mike Guidry: guidry@utk.edu

SNe1054: Crab Nebula

SNe1987A Hubble Image

Supernova Nucleosynthesis Contributions Type Ia Supernovae: Thermonuclear explosions of CO white dwarfs. Type II Supernovae: Core collapse driven events in massive stars.

In both instances,the formation of iron peak elements in explosive nucleosynthesis occurs under neutron-poor conditions. This is reflected in the 56Ni56Co56Fe signatures in both Type Ia and Type II supernova light curves and in the isotopic compositions of iron-peak elements in solar matter.

Note the alpha-nuclei (Mg, Si, S, Ar, Ca) to iron-peak abundance ratios in SNe II ejecta.

(Iwamoto et al. 1999)

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

(Thielemann et al. 1992)

Type Ia Nucleosynthesis

Type II Nucleosynthesis

Explosive Nucleosynthesis

This behavior can extend well beyond mass A=56, perhaps even through mass A=72, viz: 52Fe, 56Ni, 60Zn, 64Ge, 68Se, and 72Kr.

The “freezing” of these patterns associated with the expansion and cooling of Types Ia and II supernova ejecta underscores the importance of experimental determinations of reaction rates as well as of masses and lifetimes for proton-rich isotopes near the a-line.

Trends in Low Metallicity Stars (Cayrel et al. 2005)

[Zn/Fe]

[Cr/Fe]

Synthesis of Nuclei Beyond Iron

Nuclei heavier than iron (A 60) are understood to be formed in neutron capture processes.

The helium shells of red giant stars ( 1-10 ) provide the s-process environment, with the 13C(,n)16O reaction providing neutrons. ( > 109 years)

Supernovae II provide the astronomical setting for the r-process. ( < 108 years)

Heavy Element Synthesis Processes

Z

N

184Os 186Os 187Os 188Os

185Re 186Re 187Re

180W 182W 183W 184W 185W 186W

180Ta 181Ta 182Ta

176Hf 177Hf 178Hf 179Hf 180Hf 181Hf

175Lu 176Lu 177Lu

174Yb 175Yb 176Yb

189Os

stable

> 1010 yrs

unstable

4j

7j

42j

115j

75j

91h

rs,r

s,r

s s,r s,r s,r s,r

s,r

s,r s,r s,r

s,r

s s s,r s,r

s

r

r

p

p

p

p

s

r r-process

s-process

p-process

r-process

r-process

r-Process and s-Process Synthesis

s-process in red giants

r-process in supernovae

C-O core

Heintershell

H-richconvectiveenvelope

He-burningshell

H-burningshell

Dredge-up

Flash-drivenintershell convection

Schematic structureof an AGB star(not to scale)

Calcium

Titanium

Halo Abundance Trends for -3 [Fe/H] -1Oxygen and -Elements R-Process Elements

(Truran et al. 2002)

These behaviors are compatible with nucleosynthesis predictions for SNe II.

[/Fe] in Halo Stars and Dwarf Galaxies

(Tolstoy et al. 2005)

Sculptor

Globular Cluster StarsType Ia and Type II histories yield complicated abundance histories of stellar populations.

DLAs: Abundance Evolution with Red Shift

(Pettini 2003)

(Lu et al. 1996)

Lower bound on metallicities due to masses of typical clouds in which first stars formed.

Silicon Abundance History in the Cosmos

Figure Credit: Francesca Primas (2003)

s-Process/r-Process Chemical Evolution

(Truran et al. 2002)

Abundances in Dwarf Spheroidal Galaxies

(Shetrone et al. 2003)

Abundance Trends & Chemical Evolution: [Fe/H] > -3

Extremely metal-deficient stars of [Fe/H] ~ -2 to –3 are characterized by both high O/Fe and (Ne-Ca)/Fe ratios and an r-process heavy element pattern

SNe II production ( 108 years) Signatures of an increasing s-process contamination first appear

at [Fe/H] -2.5 to –2.0 first input from AGB stars ( 109 years)

Evidence for entry of SNe Ia ejecta first appears at [Fe/H] -1.5 to –1.0, as evidenced in the [O/Fe] and [(Ne-Ca)/Fe] histories input from SNe Ia on timescales > 1.5-2 x 109 years

Supernova Ia: Progenitors and Sites

(Oemler and Tinsley 1979)

(Sullivan et al. 2006)

Evidence for SNe Ia in the Early Galaxy

Thin Disk

Thick Disk

Observations by Bernkopf and Fuhrmann (2006) reveal distinctive abundance evolution in the thick disk and thin disk components of our Galaxy. Truran and Burkert (2008) argue that this reflects the combined heating and nucleosynthesis contributions from SNe Ia over a period of order 109 years at the end of the thick disk star formation epoch.

Trends at the Lowest Metallicities

Truran et al. (2002)

r-Process Scatter

Cayrel et al. (2005)

[Mn/Fe]

[Cr/Fe]

[Zn/Fe]

Abundance Trends for [Fe/H] < -4 ??Frebel et al. (2005)

The abundances in the two most “iron-deficient” stars known do not trace the smooth trends found (Cayrel et al. 2004) above [Fe/H] = -4. The details of their evolutions remain uncertain.

Abundance Trends/Chemical Evolution: -4 <[Fe/H}< -2.5

Evidence for increasing scatter exists in the (r-process/Fe) ratio below metallicity [Fe/H] ~ -2.5, suggesting both that only a small fraction of massive stars form r-process nuclei - and that

the ISM was highly inhomogeneous at that epoch. In contrast, the scatter in abundance ratios of nuclei from Mg to Zn with

respect to iron is remarkably small. Given the level of inhomogeneity reflected in the r-process/Fe ratio, this quite strongly implies the massive stars responsible for these early products were extremely

robust in their synthesis of nuclei through iron. (Keep in mind that the heavy elements introduced into stars formed at metallicities [Fe/H] ~ -4 are most likely to have come from a single progenitor.)

The paucity of DLAs with metallicities below [Fe/H] ≈ -3 is compatible with their having been enriched by only a very few stars - but in star forming regions typically ~ 106 M.

(Note that the introduction of 10 M of metals from a ~ 20-30 M star is sufficient to enrich a ~ 106 M cloud to a metallicity ≈ 10-3 Z.)

Look-back Times versus Redshift

Red Shift Age of the Universe Look-back Time

0 Gyr 14.5 Gyr

10 (First Stars -

0.5 14.0

6 - SNe II) 1.0 13.5

5 1.2 13.3

4 (AGB Stars) 1.6 12.9

3 (SNe Ia) 2.3 12.2

2 3.5 11.0

1 6.2 8.3

0.5 9.1 5.4

0.4 Birth of Sun

9.9 4.6

0 14.5 0(Ho= 65 km s-1 Mpc-1; baryons= 0.022 h-2; M = 0.3; =0.7; cosmos= 14.5 Gyr )

Concluding Remarks Based upon existing observations of abundances in our Galaxy, other

galaxies, and QSO absorption line systems, we might conclude: Only normal stars in a Salpeter-like initial mass function

are required to produce the elements seen in the oldest stars.

While contributions from massive stars clearly dominate at early epochs, this is more likely a consequence of their shorter production timescales rather than of an altered IMF.

Our present knowledge of the abundance history of the Universe provides no clear evidence for an earlier Population (III?).

The collective trends in halo stars, disk stars, globular clusters, dwarf spheroidal galaxies, and DLAs are generally compatible with our understanding of stellar evolution and supernova nucleosynthesis.