Nucleosynthesis in core-collapse supernova explosions triggered by a quark-hadron phase transition

Abstract

We explore heavy element nucleosynthesis in the explosion of massive stars
which are triggered by a quark-hadron phase transition during the early post
bounce phase of core-collapse supernovae. The present study is based on general
relativistic radiation hydrodynamics simulations with three-flavor Boltzmann
neutrino transport in spherical symmetry, which utilize a quark-hadron hybrid
equation of state based on the MIT bag model for strange quark matter. The
quark-hadron phase transition inside the stellar core forms a shock wave
propagating towards the surface of the proto-neutron star. The shock wave
results in an explosion and ejects neutron-rich matter which is piled up or
accreting on the proto-neutron star. Later, during the cooling phase, the
proto-neutron star develops a proton-rich neutrino-driven wind. We present a
detailed analysis of the nucleosynthesis outcome in both neutron-rich and
proton-rich ejecta and compare our integrated nucleosynthesis with observations
of metal poor stars.

Full text

Page 1

arXiv:1112.5684v1 [astro-ph.HE] 24 Dec 2011
Draft version December 30, 2011
Preprint typeset using LATEX style emulateapj v. 5/2/11
NUCLEOSYNTHESIS IN CORE-COLLAPSE SUPERNOVA EXPLOSIONS
TRIGGERED BY A QUARK-HADRON PHASE TRANSITION
Nobuya Nishimura1,2, Tobias Fischer2,3, Friedrich-Karl Thielemann1,2,
Carla Fr¨ ohlich4, Matthias Hempel1, Roger K¨ appeli1, Gabriel Mart´ ınez-Pinedo2,3,
Thomas Rauscher1, Irina Sagert5, and Christian Winteler1
e-mail: nobuya.nishimura@unibas.ch
1Department Physics, University of Basel, 4056 Basel, Switzerland
2GSI, Helmholtzzentrum f¨ ur Schwerionenforschung GmbH, 64291 Darmstadt, Germany
3Technische Universit¨ at Darmstadt, 64289 Darmstadt, Germany
4Department of Physics, North Carolina State University, NC 27695, U.S.A
5Department of Physics and Astronomy, Michigan State University, MI 48824, U.S.A
Draft version December 30, 2011
ABSTRACT
We explore heavy element nucleosynthesis in the explosion of massive stars which are triggered by
a quark-hadron phase transition during the early post bounce phase of core-collapse supernovae. The
present study is based on general relativistic radiation hydrodynamics simulations with three-flavor
Boltzmann neutrino transport in spherical symmetry, which utilize a quark-hadron hybrid equation
of state based on the MIT bag model for strange quark matter. The quark-hadron phase transition
inside the stellar core forms a shock wave propagating towards the surface of the proto-neutron star.
The shock wave results in an explosion and ejects neutron-rich matter which is piled up or accreting
on the proto-neutron star. Later, during the cooling phase, the proto-neutron star develops a proton-
rich neutrino-driven wind. We present a detailed analysis of the nucleosynthesis outcome in both
neutron-rich and proton-rich ejecta and compare our integrated nucleosynthesis with observations of
metal poor stars.
Subject headings: nuclear reactions, nucleosynthesis, abundances – supernovae: general – stars: neu-
tron – dens matter
1. INTRODUCTION
Nucleosynthesis
certainly responsible for the production of interme-
diate mass elements including the so-called alpha
elements and a certain fraction of iron isotopes and
their neighbors (the Fe-group nuclei).
alpha elements from oxygen to silicon have contri-
butions from hydrostatic burning which takes place
during stellar evolutions,
and the Fe-group isotopes originate from explosive
burning (Woosley & Weaver 1995; Thielemann et al.
1996; Woosley et al.2002;
Woosley & Heger2007;
A major open question is related to the source of heavy
elements beyond iron.
There have been strong expectations that the inner-
most layer of ejecta, close to the forming formed neu-
tron star, remains neutron-rich and is a possible site
for the r-process.For many years the late neutrino
wind, following the actual explosion, seemed an ade-
quate site in core-collapse supernovae that formed a neu-
tron star (Qian & Woosley 1996; Takahashi et al. 1994;
Woosley et al. 1994). Such investigations were followed
up by many parametrized calculations that explored the
sensitivity of the most relevant r-process parameter, the
neutron-to-seed ratio, to variations of entropy S, electron
fraction Ye and expansion timescale τ (Hoffman et al.
1997; Meyer & Brown 1997; Freiburghaus et al. 1999a;
Farouqi et al. 2010). However, steady state wind models
(Thompson et al. 2001; Wanajo 2006) showed that it is
very hard to attain the required entropies. These results
from core-collapse supernovaeis
The lighter
whilethe heavierones
Nomoto et al.
Thielemann et al.
2006;
2011).
have been recently confirmed by fully hydrodynamical
simulations Arcones et al. (2007) that in addition showed
that the presence of a reverse shock does not have any
major impact in the neutron-to-seed ratio.
Furthermore, recent investigations noticed that the
early neutrino wind turns matter proton-rich, pro-
ducing specific Fe-group isotopes and in the subse-
quent νp-process nuclei with a mass number up to
A ∼ 80 − 90 (Liebend¨ orfer et al. 2003; Pruet at al.
2005; Fr¨ ohlich et al. 2006a,b; Pruet at al. 2006; Wanajo
2006). While it was initially hoped that there exists a
chance that the late neutrino wind still turns neutron-
rich (after its initial, early proton-rich phase) all present
core-collapse calculations seem to indicate that the wind
becomes even more proton-rich in the longterm evolution
(Fischer et al. 2010; H¨ udepohl et al. 2010).
In fact, it was realized that there is a better chance to
eject neutron-rich matter, if it stems from the initial col-
lapse and compression, where electron captures turned
it neutron-rich, early in the explosion, before neutrino
interactions have the chance to convert it into proton-
rich matter in the neutrino wind. Electron-capture su-
pernovae, which explode without a long phase of accre-
tion onto the proto-neutron star, apparently provide such
conditions (Wanajo et al. 2011). However, the Ye’s ob-
tained under such conditions do not support a strong r-
process, which successfully reproduces the platinum peak
of r-elements around A = 195.
In the present paper, we investigate core-collapse
supernova explosions triggered by a quark-hadron
phase transition during the early post-bounce phase
(Sagert et al. 2009; Fischer et al. 2011) and their nucle-

Page 2

2 Nishimura et al.
osynthesis features. Under such conditions, zones which
are initially neutron-rich, can be (promptly) ejected
without experiencing the strong effects of the neutrino
wind which comes from the central proto-neutron star.
In the following sections, we discuss the supernova ex-
plosion model and the nuclear physic inputs utilized for
the nucleosynthesis calculations in §2, the conditions ex-
perienced and the resulting ejecta in §3, and give a de-
tailed analysis of the ejecta composition §4, followed by
conclusions also in comparison to alternative r-process
sites in §5.
2. METHODOLOGY AND NUCLEAR PHYSICS INPUT
We explore nucleosynthesis in core-collapse supernovae
which are triggered by a deconfinement phase transition
during the early post-bounce phase. The explosion mod-
els have been discussed in Fischer et al. (2011) in detail.
In this section, we summarize the main features of the
explosion models and the input physics of the nuclear
reaction network utilized for nucleosynthesis.
2.1. Hydrodynamics and Equation of State
A self-consistent supernova explosion model, which we
adopt to investigate nucleosynthesis in the present work,
has been carried out by using the AGILE-BOLTZTRAN
code. This numerical code is based on general-relativistic
radiation hydrodynamics in spherical symmetry with an
adaptive grid and employs a three-flavor Boltzmann neu-
trino transport (Liebend¨ orfer et al. 2004) under detailed
micro physics e.g. nuclear equation of state (EOS), weak
processes and other nuclear reactions. For the current
study the standard weak processes were considered as
listed in Table 1 of Fischer et al. (2011).
The baryonic EOS in supernovae needs in general to
be known for the following three intrinsically different
regimes:
1. At temperatures below 6 GK (≃ 0.5 MeV), the
baryon EOS is dominated by heavy nuclei and their
abundances, determined by individual nuclear re-
actions which are not necessarily in equilibrium. In
Fischer et al. (2010) we included a nuclear reaction
network consisting of twenty nuclei, which permits
to simulate a large domain of the progenitor up
to the helium-layer and is utilized to predict the
amount of energy generation by explosive nuclear
burning.
2. At temperatures above 6 GK (≃ 0.5 MeV), nu-
clear statistical equilibrium can be applied, where
the baryonic EOS from Lattimer & Swesty (1991)
and Shen et al. (1998) are the commonly used in
supernova simulation studies.
3. Above the normal nuclear matter density, the
state of matter is highly uncertain. There exists
the possibility of the deconfinement phase transi-
tion. Therefore, we extended the hadronic EOS
by Shen et al. (1998) at high densities and tem-
peratures, making use of a quark EOS based on
the bag model for strange quark matter. For the
first order phase transition between hadronic and
quark phases we applied Gibbs conditions, lead-
ing to a mixed phase during the transition. This
results in a continuous phase transformation. De-
tails of the quark-hadron hybrid EOS are discussed
in Fischer et al. (2011).
Besides, the contributions from electrons and positrons
as well as photons and Coulomb corrections to the EOS
are added by the method of Timmes & Arnett (1999).
For the current nucleosynthesis predictions, we select
the explosion calculations of the 10.8 M⊙ progenitor
model, where a quark-hadron hybrid EOS was used with
an early phase transition to quark matter close to nor-
mal nuclear matter density (labeled EOS2, second line
of table 2 in Fischer et al. 2011). We chose this model
because it has an explosion energy consistent with the
expected order of magnitude of 1051erg (see the second
line of table 3 in Fischer et al. 2011). The maximum
gravitational mass for this EOS is 1.5026 M⊙. Though
it is in agreement with the highest precisely known mass
of a compact star, Hulse-Taylor pulsar of 1.44 M⊙, recent
mass limits for the physical EOS are based on the mil-
lisecond pulsars J1903+0327 with M = 1.67 ± 0.01 M⊙
(Freire & Wex 2010) and J1614-2230 with a high mass
of M = 1.97 ± 0.04 M⊙ (Demorest et al. 2010). The
inclusion of corrections from the strong interaction cou-
pling constant can stiffen the quark EOS and lead to
higher maximum masses (see e.g. Schertler et al. 2000;
Alford et al. 2005; Sagert et al. 2010; Weissenborn et al.
2011).
Therefore, Fischer et al. (2010) constructed a quark-
hadron hybrid EOS which includes corrections from the
strong interaction coupling constant. They showed that
such an EOS, with a maximum mass of 1.67 M⊙and a
phase transition to quark matter close to nuclear satura-
tion density (see EOS3 in table 2 and 3 in Fischer et al.
2011), leads to a qualitatively similar explosion scenario
in spherical symmetry as obtained in the explored mod-
els with several EOS which have lower maximum masses.
Furthermore, Weissenborn et al. (2011) recently showed
that it is possible to obtain a quark-hadron hybrid EOS
which allows for both, a maximum mass larger than 2 M⊙
and a low critical density for the appearance of quark
matter. Whether such quark-hadron hybrid EOS will
result in a similar dynamical evolution as discussed in
Fischer et al. (2011) will be examined in future simula-
tions.
2.2. Explosion Scenario
The supernova post-bounce evolution is characterized
by mass accretion causing a continuous rise of the cen-
tral density. Once the central density exceeds the crit-
ical density for the onset of deconfinement, the quark
hadron phase transition takes place, leading to the ap-
pearance of a quark-hadronmixed phase. Thereby, quark
matter appears in the supernova core where the highest
densities are experienced. The timescale for the appear-
ance of quark matter is given by the timescale for the
central density to rise. This depends on the progenitor
model, which determine mass accretion rates, and the
hadronic EOS. Note that the EOS in the mixed phase
is significantly softer than in hadronic and pure quark
phases. It is a consequence of the assumed first-order
phase transition. Mass accretion from the outer layers of
the progenitor onto the central supernova core leads to a
continuous rise of the central enclosed mass. When the

Page 3

Nucleosynthesis in Supernova Explosions Triggered by a Quark-Hadron Phase Transition3
critical mass (given by the hybrid EOS) of the config-
uration is obtained, it becomes gravitationally unstable
and the supernova core begins to contract. The contrac-
tion proceeds into a collapse which rises the density and
convert hadronic core into quark matter at around the
center. A massive pure quark core forms at the center,
where the EOS stiffens and the collapse halts, and an
accretion shock forms. The shock wave propagates out
of the high-density supernova core, remaining an accre-
tion front with no matter outflow. Once it reaches the
outer layers of the central core, where the density drops
over several orders of magnitude, the accretion front ac-
celerates and turns into a dynamic shock with matter
outflow. This moment determines the onset of an explo-
sion, also for supernova models which would otherwise
not explode in spherical symmetry, based on explosion
mechanisms discussed so far. Finally, at distances on
the order of 100 km the expanding shock wave merges
with the standing shock from the initial core bounce at
nuclear densities, which was unaffected by the dynamics
occurring in the supernova core.
When the (second) shock reaches the neutrino-spheres,
an additional millisecond neutrino burst is released. It
appears in all flavors, however dominated by ¯ νeand νµ/τ,
in contrast to the νe-deleptonization burst related to
the early bounce shock propagation across the neutrino-
spheres between 200 to 500 ms after core bounce. This
second neutrino burst is of particular interest for water-
Cherenkov neutrino detectors, which are more sensi-
tive to ¯ νe than to νe. Dasgupta et al. (2010) demon-
strated that the currently operating generation of water-
Cherenkov neutrino detectors (e.g. Super-Kamiokande
and IceCube detectors) can resolve such millisecond neu-
trino burst of the explosion models by Fischer et al.
(2011).
The matter considered for nucleosynthesis studies of
heavy elements (see section 3) belongs originally to the
inner parts of the silicon- and sulfur-layers of the 10.8 M⊙
progenitor from Woosley et al. (2002), with temperature
and electron fraction of 3 GK and Ye≃ 0.5, respectively,
at 800 to 1000 km from the center on the pre-collapse
phase. During the collapse and the explosion, material
contracts and is heated by the shock, and the temper-
ate exceed 100 GK and hence completely dissociate into
free nucleons. At high densities weak processes, mainly
electron captures, establish a very low proton-to-baryon
ratio Ye≃ 0.1 during the core-collapse. The further evo-
lution is discussed in detail in section 3.
2.3. Nuclear Reaction Network
The nuclear reaction network utilized for the follow-
ing nucleosynthesis simulations of the ejecta is an exten-
sion of previous ones, which have already been described
in detail (see, Nishimura et al. 2006; Fujimoto et al.
2008). The network includes more than 4000 nu-
clei from neutrons and protons up to fermium with
atomic number Z = 100 (for detail, see Table 1 in
Nishimura et al. 2006) and includes proton-rich isotopes
as well as neutron-rich ones far from stability.
cludes two- and three-body reactions, decay channels,
and electron as well as positron capture (for details,
see network A in Fujimoto et al. 2007) and screening
effects for all relevant charged particle reactions. Ex-
perimentally determined masses Audi & Wapstra (1995)
It in-
and reaction rates are adopted if available.
wise, theoretical predictions for nuclear masses, reaction
rates and beta-decays are applied, based on the Finite
Range Droplet Model FRDM mass model (M¨ oller et al.
1995). Spontaneous and beta-delayed fission processes
(Staudt & Klapdor-Kleingrothaus (1992)) are taken into
account. We adopt the empirical formula for fission frag-
ments by Kodama & Takahashi (1975).
We also employ neutrino interactions with matter in
order to include dominant weak interactions affecting the
evolution of the overall proton/nucleon ratio Ye.
(anti-)electron-neutrino captures by nucleons we adopt
reaction rates derived by Qian & Woosley (1996) but
ignore any reactions with heavy isotopes because the
amount of neutrino capture is negligible for the early
phase of nucleosynthesis. The neutrino fluxes, result-
ing from the detailed neutrino-radiation hydrodynam-
ics calculation described in the previous subsection, are
utilized to determine the actual rates as a function of
time. These reaction rates depend on the distance from
the proto-neutron star and the mean energy and lumi-
nosity of neutrinos emitted from the proto-neutron star.
It depends on the structure and the evolution, which is
sensitive to the EOS, and the precise evolution history
of ejected matter including the early phase of the core
bounce. Thus, this is different from the treatment of
other nuclear reaction rates, which are determined only
by local thermodynamic conditions and density and tem-
perature.
Other-
For
3. NUCLEOSYNTHESIS IN THE EJECTA
3.1. Dynamic Evolution of Mass Zones
In order to calculate the nucleosynthesis evolution of
ejected matter within a postprocessing approach, the
dynamic evolution is required in radial Lagrangian mass
zones. For this reason the evolution determined with
the radiation hydrocode AGILE/BOLTZTRAN (see
Mezzacappa & Bruenn1993a,b,c;
2001a) which is based on an adaptive grid, was mapped
on a Lagrangian grid of 120 mass zones. This provides
the Lagrangian evolution of physical quantities, such as
density, temperature, electron fraction and velocity of
the ejected material, and in addition the neutrino fluxes
experienced as a function of time.
Liebend¨ orfer et al.
TABLE 1
Summary of mass zones and their properties
zone #
001 – 014
015 – 019
020 – 050
051 – 120
M#[10−2M⊙]
0.000 − 0.208
0.210 − 0.216
0.217 − 0.232
0.250 − 1.482
∆M#[M⊙]
1.496 × 10−4
1.474 × 10−5
1.063 × 10−5
1.786 × 10−4
Ye,NSE
0.20
∼ 0.55
∼ 0.33
0.33 ∼ 0.50
tej
—
1.5 ∼
∼ 0.5
∼ 0.5
aM#: mass coordinates relative to the innermost zone of 1.48M⊙
b∆M#: the averaged mass of the zone
cYe,NSE: Ye at the end of NSE (below T = 9 GK)
dtej: ejection time after the bounce
The mass zones which are ejected in the explosion are
classified in three different categories, related to their
ejection process and thermodynamic quantities. As listed
in Table 1, zones #001 to #120, given with the ejection

Page 4

4Nishimura et al.
timescale after bounce (tej) and the final Ye,NSE which
is Ye at the end of NSE (below T = 9 GK), cover the
material from the surface of the inner core at 1.48000
M⊙to layers with a corresponding mass of 1.49482 M⊙.
They are also shown with respect to their mass as well
as Ye-distribution in Fig. 1.
0.1
0.3
0.5
0.7
0.9
0 20 40 60 80
0.1
0.3
0.5
0.7
0.9
mass, M# [10-2 M⊙]
electron fraction, Ye, NSE
mass zone, #
NSNDWdelayedprompt
mass zone
Ye
Fig. 1.— Initial distribution of mass and Yeas a function of mass
zone number. Thick lines and dashed lines relate to Lagrangian
mass coordinate (starting at 0 for zone #001) and electron frac-
tion, respectively. Masses are measured from the surface of the
proto-neutron star (starting at 0 for zone #001 of Table 1) and the
electron fractions are adopted at the time when the temperature
decreases down to T = 9×109K for ejected matter and the end of
the hydrodynamic simulation for inner non-ejected zones). Though
the plot is shown at the range of #001 to #090, mass and Ye
for the mass zones #90 to #120 are proportional to mass zone
and almost invariable, respectively.
These zones coincide with the matter discussed at the
end of subsection 2.2, where at high densities a Ye de-
crease down to even 0.02 has been noticed. The evolu-
tion of these mass zones in time is displayed in Fig. 2.
Zones #001 to #014 are not ejected within 0.5 s after
the core bounce. These zones preserve the original low
Yeobtained during collapse and shock wave propagation.
As they are not ejected, we ignore them in the further nu-
cleosynthesis discussion, plus all matter originating from
regions at smaller radii. In Fig. 2, they are displayed in
black. Zones #015 to #019 are ejected in the so called
neutrino driven wind, shown in red. Their Yeis strongly
affected by neutrino interactions, turning this matter
proton-rich. Zones #020 to #050 (displayed in green)
have stalled from infall after shock formation and are
ejected thereafter due to neutrino heating and dynamic
effects (Fischer et al. 2011). The adjacent zones #051
to #120 are ejected in a prompt way, due to the shock
wave originating from the deconfinement phase transi-
tion (displayed in blue). We clearly see the division of
matter which is ejected in a prompt fashion (blue), mat-
ter which is coasting and falling in again, but gets reac-
celerated outward by neutrino energy deposition (green),
matter which falls back onto the neutron star, but be-
comes part of the neutrino wind ejecta (red), and finally
matter which stays on the neutron star and will never be
ejected (black).
As shown in the bottom part of Fig. 2, the blue and
green zones are neutronized during the collapse via elec-
101
102
103
104
1.0
radius [km]
0.5
0.5
2.0
2.0
4.0
4.0
0.5
0.5
2.0
2.0
4.0
4.0
0.1
0.2
0.3
0.4
0.5
1.0
electron fraction, Ye
time after bounce [s]
Fig. 2.— top: Radial trajectories of mass elements as a func-
tion of time after bounce. The colors indicate the properties of
these mass elements: black, red, green and blue lines refer to mat-
ter which is either (black) not ejected, (red) part of the neutrino
driven wind, (green) initially stalled matter which gets boosted
by the wind and (blue) matter which experiences a prompt ejec-
tion. bottom: Evolution of Ye as a function of time after the
core bounce. The deconfinement phase transition happens about
0.4s after this initial bounce, causing the explosion and ejection of
matter. The colors are the same as in the top panel.
tron capture to various degrees, depending on the max-
imum density attained. Their Ye-values range from 0.35
to 0.5. The prompt or quasi-prompt ejection does not
change this value (with minor effects on the innermost
zones, being partially affected by the neutrino wind).
The mass zones in red experience a similar effect in their
early evolution during collapse, but the later evolution
leads to values of Ye exceeding 0.5. This is similar to
recent studies of the neutrino driven wind (see the in-
troduction and Fischer et al. 2010), where similar neu-
trino and anti-neutrino spectra and flux intensities favor
proton-rich matter due to the neutron-proton mass dif-
ference, resulting in different energies available for the
neutrino/anti-neutrino captures. As the neutrino lumi-
nosity is still high in the ejection phase, we expect νp-
process nucleosynthesis. Finally non-ejected mass zones
(black) can initially also experience interaction with the
neutrino flux and turn proton-rich while still at larger
radii and small(er) densities. Once they settle on the
surface of the neutron star at high densities, capture of
degenerate electrons dominates over the neutrino effects
and they turn neutron-rich again.
The thermodynamic conditions (density, temperature
and entropy), which are responsible for the nucleosyn-
thesis results, are shown in Fig. 3 for the ejected mass

Page 5

Nucleosynthesis in Supernova Explosions Triggered by a Quark-Hadron Phase Transition5
10
100
1.0
entropy [kB]
time after bounce [s]
0.5
0.50.5
2.0
2.02.0
4.0
4.04.0
0.1
1
10
1
10
100
temperature [MeV]
temperature [GK]
104
107
1010
1013
density [g/cc]
Fig. 3.— Evolution of temperature (top), density (middle), and
entropy (bottom) of mass zones as a function of time after bounce.
zones. These figures indicate a temperature, density and
entropy maximum when the quark-hadron phase transi-
tion occurs, which causes a second core bounce (about
0.4 s after the first bounce at nuclear densities; see Fig. 2)
and an outgoing shock front forms. The expansion fol-
lows a close to constant entropy, i.e. is adiabatic, once
matter is ejected (after 0.4 s for the prompt ejection and
after 1.5 s for the delayed ejection). The matter which
initially fell back onto the proto-neutron star and is fi-
nally ejected by the neutrino driven wind, experiences
heating and an entropy rise due to this energy deposi-
tion by neutrinos.
3.2. Neutrinos from the Proto-neutron Star
The properties of the neutrino and anti-neutrino flux
(luminosities, average energies and neutrino sphere radii)
can be found in Fig. 4. It is clearly seen that the initial
bounce (0 s) at nuclear densities leads to a neutrino burst
due to electron captures, while the second shock wave -
caused by the quark-hadron phase transition (0.4 s) - also
produces antineutrinos. From that point on in time the
neutrino and anti-neutrino luminosities are comparable
(slightly smaller for anti-neutrinos). The average ener-
10
54
50
90
130
170
radius [km]
Eνe
Eanti-νe
52
53
luminosity, log10 [erg/s]
10
20
30
40
0.00.51.0
mean energy [MeV]
time after bounce [s]
(#018)
(#018)
Fig. 4.— The radius of neutrino spheres for electron neutrinos
and electron anti-neutrinos (top). Luminosities (middle) and mean
energies (bottom) experienced by mass zone #018, which belongs
to the neutrino-driven wind with proton-rich ejecta, as a function
of time after bounce. The difference between the mean energy of
anti-neutrinos and those of neutrinos is of the order 3 MeV, i.e.
less than 4∆, where ∆ is the neutron-proton mass difference.
gies are larger for anti-neutrinos than for neutrinos, but
the difference remains less than 4 MeV. Neutrino and
anti-neutrino captures determine the neutron/proton ra-
tio due to the reactions
¯ νe+ p → n + e+
νe+ n → p + e−.
Based on the neutron/proton mass difference of ∆ =
1.293 MeV, (Fr¨ ohlich et al. 2006a) could show (see their
Eq.4), that with the use of Eqs.
Qian & Woosley (1996), ˙Ye > 0 in the case that the
difference between the mean antineutrino and neutrino
energies fulfills ǫ¯ ν − ǫν < 4 ∆.
is obtained if the timescale for neutrino/anti-neutrino
captures is shorter than the dynamic timescale. Thus,
for all conditions discussed here, where neutrino and
anti-neutrino captures are responsible for the n/p ratio,
proton-rich conditions are attained, i.e. Ye> 0.5. That
64a and 64b in
Therefore Ye > 0.5

Page 6

6 Nishimura et al.
is exactly what is seen in Figs. 2 and 4 for mass zones
which experience essential neutrino fluxes (weighted by
1/r2) at radii of about 100km. For matter at larger radii
(about 1000km), the timescale for this process is too long
and minor Yechanges occur, i.e. the initial Yefrom the
collapse phase is retained. Matter at smaller radii, on top
of the neutron star, experiences high densities and elec-
tron Fermi energies, where electron captures dominate
which make matter neutron-rich.
30
40
50
60
70
80
90
20 40 60
mass zone #
80 100 120
0.30
0.40
0.50
0.60
0.2090.2171.482
entropy, sNSE [kB]
electron fraction, Ye,NSE
mass, M# [10-2 M⊙]
NDWdelayed prompt
entropy
Ye
Fig. 5.— Yeand entropy S, which set the conditions for explosive
nucleosynthesis at the the time of matter ejection. The innermost
ejected zones are proton-rich due the effect of the neutrino wind,
which also heats matter efficiently, leading to high entropies. The
outer mass zones, ejected in a more prompt fashion keep their orig-
inal (slightly neutron-rich) Ye from the infall/compression phase.
Fig. 5 underlines this effect due to neutrino interactions
or electron capture. All outer mass zones keep their orig-
inal Ye, which is due to electron capture at high densities
during the collapse, and ranges from about 0.48 (further
out) to 0.32 (for the inner quasi-prompt ejected matter).
Material which fell in initially onto the surface of the neu-
tron star and is then ejected via the neutrino wind, has
been turned proton-rich by neutrino and anti-neutrino
captures with values up to Ye= 0.55. The neutrino wind
also leads to energy deposition and an entropy increase
to maximum values of about 85 kbper baryon. The en-
tropy in the outer ejected regions is of the order 30-50 kb
per baryon, caused by shock heating during the passage
of the ejection shock wave.
3.3. Nucleosynthesis Results
In the following we show final nucleosynthesis results
for a number of typical mass zones. In order to get a
rough idea about the results of explosive nucleosynthesis,
one can utilize either maximum densities and tempera-
tures prior to an adiabatic expansion or the entropies at-
tained in the expanding matter (in radiation-dominated
regimes S ∝ T3/ρ).
Comparing entries in Fig.5 of (Thielemann et al. 1990)
and Fig.3 of (Thielemann et al. 1996) leads to the conclu-
sion that (a) these are typical conditions for an alpha-rich
freeze-out from explosive Si-burning and (b) one would
expect remaining alpha mass-fractions after charged-
particle freeze-out of the order 20-100%. One should con-
sider, however, that those calculations were performed
-8
-6
-4
-2
0
2
20 40 60
mass zone, #
80 100 120
-2
-1
0
0.2090.2171.482
n/seed, log10(Yn/Yseed)
mass fraction, log10 Xα
mass, M# [10-2 M⊙]
delayedprompt
n/seed
no boost
Xα
Fig. 6.— Neutron/seed ratio and remaining mass fractions of
4He (α-particles) after charged-particle freeze-out. Both properties
result from the original Ye and entropy S in these mass zones. In
turn they determine the fraction of heavy elements and whether
those experience further neutron capture after charged-particle
freeze-out, which is the key to the pattern of heavy nuclei and the
maximum mass number attained. In the inner part (green lines)
an effect is seen, which results from an intermediate fallback before
final ejection. These mass zones experience first (due to the shock
from the deconfinement phase transition) a maximum temperature
and density, expand afterwards close to adiabatically, heat up dur-
ing the intermediate fallback, and then expand freely. The line
indicated with ”no-boost” gives the initial neutron/seed ratio after
the first expansion. The alpha-fraction in the inner zones results
from the reheating phase. The initial expansion at low Ye’s would
result in vanishing alpha-fractions.
for hydrodynamic (i.e. free fall) expansion timescales,
which can differ from the actual simulation, and a value
of Ye= 0.4988, i.e. matter neither neutron nor proton-
rich. Therefore we expect the following changes (i)
higher/lower entropies within the given variety will lead
to higher/lower remaining alpha-fractions, (ii) higher
˙Ye, i.e.more proton-rich matter causes an alpha-rich
charged-particle freeze-out with remaining free protons,
which can thereafter lead to a νp-process, if a sufficient
flux of electron anti-neutrinos is still present, (iii) smaller
Ye’s, i.e. more neutron-rich matter permits to bypass the
slower triple-alpha reaction via the faster ααn-reaction,
in order to produce heavier nuclei and a reduction in the
remaining alpha-fraction is expected. In addition, free
neutrons are remaining after the charged-particle freeze-
out. With respect to this latter aspect, we expect also
the additional behavior: higher entropies and lower Ye’s
lead to a larger neutron/seed ratio, seed nuclei being the
heaviest nuclei formed after charged-particle freeze-out,
and permit therefore more neutron captures on these
seed-nuclei.Dependent on the neutron to seed ratio,
this could lead to light, medium or strong r-processing,
producing nuclei in the first, second or third r-process
peaks, around A=80, 130 or 195, depending on the ac-
tual n/seed ratio attained.
In Fig. 6 the properties of mass zones #020 to #120
are summarized, those with a Ye < 0.5 which experi-
ence neutron-rich conditions to a varying degree. Prop-
erties (i), i.e. the degree of alpha-rich freeze-out and
(iii), the resulting neutron/seed ratio, are displayed. We
see a complex dependence of these properties on en-
tropy S, Ye, expansion timescale τ, plus further com-

Page 7

Nucleosynthesis in Supernova Explosions Triggered by a Quark-Hadron Phase Transition7
plications from reheating and re-expansion which do not
fit to a simple expansion interpretation. For similar Ye’s
the remaining alpha-fraction increases with entropy, as
expected, when looking at the behavior of mass zones
#080 to #120. Then, following mass zones further in,
the decrease in Ye dominates, which permits to pass
the alpha-to-carbon bottle-neck more efficiently, via the
ααn-reaction, and the remaining alpha-fraction vanishes.
As known from moderately neutron-rich neutrino wind
simulations, none of the entropies encountered here leads
to sizable neutron/seed ratios. For such low entropies
only a strongly neutron-rich initial composition permits
large(r) neutron/seed ratios. This is what can be noticed
when following mass zones from #070 down to #040,
where the lowest Ye’s are encountered. Mass zones #020
- #040 experience a more complicated history, initial ex-
pansion after shock passage, later partial fallback, and
then final ejection and further expansion. Here the neu-
tron/seed ratio after the first charged-particle freeze-out
is the one of importance for the production of heavy el-
ements.
The re-heating leads to partial photo-disintegration of
heavy elements, the production of alphas and the buildup
towards nuclei with mass numbers around A = 90 − 110
during the final expansion. The neutron/seed ratio at
this second charged-particle freeze-out (boost) is rather
a measure for local rearrangements of matter.
The maximum neutron/seed ratio of slightly above
10 obtained over the complete range of ejected mass
zones, does not support conditions to produce the third
r-process peak, in fact only a small production of the
second peak is expected. The effect of the ”boost”, i.e.
second reheating and expansion in mass zones 20-40, re-
arranges/reshapes the abundance distribution via photo-
disintegrations and captures, but does not alter this con-
clusion.
With this background we have a look at the final abun-
dances of representative mass zones, displayed in Fig. 7.
We see in the outer layers remaining alpha-fractions be-
yond 60%. These are regions, which experience entropies
of S = 33 − 50kb/baryon and Ye’s close to 0.5 and see
vanishing neutron/seed ratios. Thus, we expect essen-
tially the production of the Fe-group up to A=50-70.
This can be observed in the last two subfigures of Fig. 7
(zones 80 and 120). Mass zones 60 and 70, which ex-
perience the highest entropies and moderately decreased
Ye’s, can move matter up to and (slightly) beyond A=90
(for mass zone 70 still with a large fraction of matter in
the Fe-group), mostly due to a more neutron-rich (lower
Ye) charged-particle freeze-out and not due to further
neutron processing (see neutron/seed ratio in Fig. 6).
One can also see a variation in the final carbon-fraction,
underlining how effective the bridging of the alpha-to-
carbon bottle neck of reactions is in comparison to sub-
sequent capture reactions to heavier nuclei. At smaller
radii (mass zones #020 - #051 and Ye’s as small as 0.33),
also the A = 130 peak starts to be populated. This is
due to the neutron/seed ratio of up to 10 attained af-
ter charged-particle freeze-out. For the mass zones 20-45
abundances are shown after the first expansion (dashed)
and after reheating/boost (solid), which is characterized
by some photo-disintegration of heavy nuclei, the appear-
ance of4He, and further processing of nuclei beyond the
Fe-group (see also Fig. 8). Summarizing the results of all
mass zones with Ye< 0.5, we notice that none of these
zones produce matter beyond A ∼ 130, nor shows the
130 peak dominating abundances.
Deeper mass zones, ejected via the neutrino wind turn
proton-rich and they experience entropies as high as 85
kb per baryon. These are conditions where we expect
a νp-process (Fr¨ ohlich et al. 2006b; Pruet at al. 2006;
Wanajo 2006). This can be seen in the first three pan-
els of Fig. 7, dashed lines show abundances without the
inclusion of the νp-process, solid lines show the final re-
sults after νp-processing. While in terms of total mass
fraction the production of nuclei beyond the Fe-group is
not too impressive, Fig. 9 displays this more prominently,
where the overproduction ratio over solar is plotted. In
the proton-rich environment anti-neutrino captures on
free protons produce neutrons and permit to overcome
the rp-process waiting point64Ge via an (n,p)-reaction,
winning against a slower β+-decay. In this way nuclei up
to A = 80 − 90 can be produced on the proton-rich side
of stability. The comparison of the black and red bullets
indicates the strength of this process.
4. SURVEY AND INTEGRATED COMPOSITION
After having discussed the individual composition,
ejected from different positions in the exploding model,
we want to give a final survey of the conditions attained
in all mass zones of explosive Si-burning which are af-
fected by the explosion mechanism (here the deconfine-
ment phase transition). The discussion of nucleosynthe-
sis in layers further out is only affected by the energy in
the shock wave, as has been discussed extensively in the
literature (Thielemann et al. 1996), and will not be re-
peated here. Finally, we also discuss the overall features
of the integrated yields (of these inner mass zones, close
to the explosion mechanism). The survey is displayed in
Fig. 10 which features abundances after charged-particle
freeze-out, and thus the setting for the final nucleosyn-
thesis features, if free neutrons or protons are remaining
in sizable fractions.
Essentially all mass zones have total entropies in ex-
cess of S=30 kb per baryon. In the outer mass zones
with a Yeclose to 0.5 this produces high alpha-fractions
plus dominantly56Ni. Moving somewhat further in with
slightly increasing entropies, noticeable amounts of64Ge
are produced as well, with the64Ge/56Ni being a measure
of entropy. Decreasing Ye, i.e. having more neutron-rich
conditions, changes the initially pure alpha-rich charged
particle freeze-out to an alpha-rich freeze-out with sizable
abundances up to A=90. A further decrease in Yecauses
remaining free neutrons, which permit an additional se-
quence of neutron captures. The decrease of Yedown to
0.33, leads to a charged-particlefreeze-out with vanishing
alpha-fractions but a sufficient amount of free neutrons,
which permit later to produce nucleosynthesis ejecta in
the mass A=130 peak. The mass zones which are af-
fected by a reheating boost are characterized by photo-
disintegrations and a second charged-particle freeze-out
with remaining alpha-fractions, Finally the innermost
proton-rich zones, with higher entropies of about S=85
kB per baryon and experiencing a continuous neutrino-
flux, show an alpha-rich and proton-rich freeze-out with
Ye’s up to 0.55. This causes a νp-process after charged-
particle freeze-out , permitted by neutron production via
anti-neutrino captures on free protons. This process will

Page 8

8 Nishimura et al.
-6
0
-5
-4
-3
-2
-1
0
10 50 90 130
#015
mass number, A
50
mass number, A
abundance, log10 XA
abundance, log10 XA
10 90 130
#017
-6
0
-5
-4
-3
-2
-1
0
10 50 90 130
#019
-6
0
-5
-4
-3
-2
-1
#020 #040
-6
0
-5
-4
-3
-2
-1
#045
-6
0
-5
-4
-3
-2
-1
#050#051
-6
0
-5
-4
-3
-2
-1
#060
-6
-5
-4
-3
-2
-1
105090 130
#070
10 5090 130
#080
105090 130
-6
-5
-4
-3
-2
-1
#120
Fig. 7.— Final nucleosynthesis results for selected mass zones showing mass fractions as a function of nuclear mass number A. The mass
zone number is given in each panel. In panels, where we show dashed and solid lines, the solid lines correspond always to the final result. In
zones up to 19 the dashed lines correspond to abundances without including the νp-process, i.e. before proceeding beyond the beta-decay
bottle-neck64Ge. In zones 20-45 they show abundances before the reheating boost.
produce nuclei up to A=90, but on the proton-rich side
of stability, especially Sr, Y, and Zr isotopes.
After this survey of the conditions at charged-particle
freeze-out, combined with the final ejecta composition as
a function of radial mass coordinate as discussed in the
previous section, we want to give an integrated presenta-
tion for these inner mass zones which experience condi-
tions for the possible formation of nuclei beyond the Fe-
group. Fig. 11 (top) shows the composition for this range
in mass numbers in comparison to solar r-abundances.
As we do not yet know the frequency of such events, i.e.
which range of the initial mass function of stellar masses
leads to these types of explosions, we show a scaling nor-
malized to the A ∼ 100 mass region (where the dominant
abundances are obtained). The bottom part of Fig. 11
shows the individual contributions to the overall abun-
dances by the different mass zones as presented in Table
1, Fig. 1 and shown in different colors in Fig. 2 and
3 as well as 5 and 6. The high end of the abundance
distribution in the prompt (blue, dashed) as well as the

Page 9

Nucleosynthesis in Supernova Explosions Triggered by a Quark-Hadron Phase Transition9
-2
0
2
4
6
0 20 40 60 80 100 120 140
log10 XA /XA (no boost)
mass number, A
-5
-4
-3
-2
-1
0 20 40 60 80 100 120 140
abundance, log10 XA
delayed
no boost
Fig. 8.— Top: The comparison of results with (solid) and with-
out (dashed) the effect of reheating and second expansion (boost)
integrated over the whole range of mass zones 20-50. Bottom: The
ratios of the two cases, based on the full hydrodynamic evolution
with reheating (boost) and the neglection of reheating around 1.5
s.The reheating leads to photodisintegrations, a strong appearance
of4He and a reshaping of the heavy element distribution.
0
1
2
3
30 40 50 60 70 80 90
overproduction, log10 X/X⊙
mass number, A
νp-proc.
without
Fig. 9.— Resulting nucleosynthesis for the entire mass zones in
the neutrino wind, experiencing a νp-process. The result is shown
for two options: (a) including the anti-neutrino captures which
turn protons into neutrons in the late phase of nucleosynthesis and
permit to overcome the64Ge beta-decay bottle neck via an (n,p)-
reaction (filled red circles), (b) neglecting this effect (open black
circles). It can be seen that the inclusion of anti-neutrino reactions
enhances abundances of nuclei heavier than A = 64 and permits
the production of nuclei up to A = 80 − 90.
delayed (boosted) ejecta (green, solid) is due to the most
neutron-rich conditions with Yeclose to 0.33. The outer-
most ejected mass zones with Yecloser to 0.5 contribute
to the Fe-group and matter up to A = 80 (lower range of
mass numbers of blue dashed line). The neutrino-driven
wind (NDW, red line) with slightly proton-rich ejecta (
Ye = 0.55) produces significant abundances only up to
-2
-1
0
0.50 0.70 0.90 1.10 1.30 1.50
enclosed mass, [10-2 M⊙]
(b) inner (neutron rich)
mass fraction, log10 X
(a) outer (neutron rich)
88Sr
90Zr
4He
60Ni
58Ni
56Ni
-2
-1
0
0.25 0.30 0.35 0.40 0.45 0.50
enclosed mass, [10-2 M⊙]
mass fraction, log10 X
n
4He
84Ge
96Kr
94Kr
94Sr
88Kr
88Sr
-4
-3
-2
-1
0
0.210
enclosed mass, [10-2 M⊙]
0.212 0.2140.216
mass fraction, log10 X
(c) innermost (proton rich)
4He
p
56Ni
57Ni
52Fe
48Cr
Fig. 10.— Overview of dominating conditions, i.e. abundances
of a few key nuclei after charged-particle freeze-out, as a func-
tion of radial Lagrangian mass, with variations from (a) typical
explosive Si-burning products and an alpha-rich freeze-out in the
outer zones to (b) an increasing remaining neutron abundance af-
ter charged-particle freeze-out, permitting a weak r-process, down
to (c) proton-rich neutrino wind ejecta, permitting the onset of an
νp-process.
A = 64, if the same normalization for the relevant mass
zones is used (see, however, also Fig. 9).
It is clearly visible that, first of all, objects like these su-
pernovae, exploding by a mechanism based on equations
of state with a low density quark-hadron phase transi-
tion, do only experience a weak r-process in ejected mass
zones which were neutronized during collapse. There is
no matter produced in the third r-process peak. If nor-
malizing the abundance curve at A = 100, in order to
avoid an overproduction of this mass region, also only
a small contribution (less than 10% is expected to the

Page 10

10 Nishimura et al.
second r-process peak, i.e. A = 130. The major produc-
tion affects the atomic mass range from A = 80 − 115,
curiously also reproducing a minimum at A = 97 − 99.
Thus, the type of events discussed here, contribute to the
whole mass region beyond the Fe-group up to A = 115
in a significant way, accompanied by minor contributions
to the second r-process peak at A = 130.
One can argue, that there might exist some uncertainty
in weak interactions (electron captures and neutrino cap-
tures) which determine Ye. For this reason we also re-
peated the present nucleosynthesis calculations with vari-
ations in the initial Ye for the mass zones experiencing
prompt and delayed explosions, according to the recipe
Ye, cor= 0.5 + (Ye− 0.5) ×
?
1 +pcor
100
?
,
where Ye, cor’s are corrected ones and pcor denotes the
percentage of uncertainty in deviations of Ye from the
symmetric value 0.5 and enlarges these deviation from
0.5. The nucleosynthesis results are also presented in
Fig. 11 and show the options of obtaining a full r-process.
However, we expect that any uncertainties beyond 10%
are unrealistic for the explosion model we adopt in the
current work. Thought we assume 10% differences would
not change the discussion of the results given above, for
simulations beyond 10% of uncertainties we need dif-
ferent progenitor, explosion models or additional input
physics i.e. multi dimensional hydrodynamics and differ-
ent EOS’s.
As is obvious from the discussion above, that only a
”weak” r-process can be supported by the nucleosyn-
thesis conditions found in the explosion mechanism dis-
cussed and presented here, one might wonder whether
such conditions support abundance features found in
”weak r-process” low metallicity stars as observed by
Honda et al. (2006). For this reason we also show such a
comparison in Fig. 12. What can be seen is that these ob-
servations also show sizable r-process features above the
A=130 peak, although weaker than in solar r-element
abundances. If such abundance distributions are the re-
sult of a single nucleosynthesis pollution, also the ”weak”
r-process found in the present paper cannot explain such
features. One could argue, however, that such observed
abundance features are a combination of at least two pol-
lutions, one (low level) solar r-contribution plus another
”weak” r-contribution extending only up to A = 130.
5. DISCUSSION AND SUMMARY
Supernova nucleosynthesis is well understood for
the outer ejected mass zones, which can be well
approximated bya shock wave with
energy passing through the layers of the progen-
itor(Woosley & Weaver
1996;Woosley et al. 2002;
Woosley & Heger 2007; Thielemann et al. 2011). What
remains uncertain is the composition of the innermost
ejecta, directly linked to the explosion mechanism, i.e.
the collapse and explosion phase. In the present paper
we analyzed these mass zones of core-collapse supernovae
explosions triggered by a quark-hadron phase transition
during the early post-bounce phase (Sagert et al. 2009;
Fischer et al. 2011). A number of aspects are important
for understanding these results.
ejecta are strongly affected by the neutrino wind.
appropriate
1995; Thielemann et al.
Nomoto et al.2006;
The very innermost
-7
-6
-5
-4
-3
-2
40 80 120 160 200
abundance, log10 YA
standard
pcor = 10
pcor = 30
pcor = 40
solar
-9
-8
-7
-6
-5
-4
-3
4080 120 160200
abundance, log10 YA
mass number, A
NDW
delayed
prompt
Fig. 11.— Integrated abundance distributions as a function of
atomic mass number A for all ejecta in comparison to solar r-
abundances (top) (normalized to the A=100 region) and separated
by ejection process(bottom). We also indicate the effect of uncer-
tainties in the Yedetermination, given in terms of percentage Pcor.
-5
-4
-3
-2
-1
0
30 40 50 60 70 80
abundance, log10 YZ
atomic number, Z
standard
pcor = 10
pcor = 30
pcor = 40
HD122563
Fig. 12.— Integrated abundance distributions as a function of
atomic charge number Z in comparison to abundance features in
”weak r-process” low metallicity stars as observed by Honda et al.
(2006). We also indicate the effect of uncertainties in the Ye deter-
mination, given in terms of percentage Pcor.

Page 11

Nucleosynthesis in Supernova Explosions Triggered by a Quark-Hadron Phase Transition11
Recent investigations noticed that this neutrino wind
turns matter proton-rich, producing specific Fe-group
isotopes and in the subsequent νp-process nuclei with
masses up to A = 80 − 90 (Liebend¨ orfer et al. 2003;
Pruet at al. 2005; Fr¨ ohlich et al. 2006a,b; Pruet at al.
2006; Wanajo 2006). Even in the longterm evolution
proton-rich conditionsprevail
H¨ udepohl et al. 2010). Thus, there seems to exist no
chance to produce r-process matter in these innermost
regions, despite many interesting parameter studies for
neutrino wind ejecta in terms of entropy S, electron
fraction Ye and expansion timescale τ (Hoffman et al.
1997; Meyer & Brown 1997; Freiburghaus et al. 1999a;
Farouqi et al. 2010) or hydrodynamic studies,
tially with parameter variations (Arcones et al. 2007;
Kuroda et al. 2008; Panov & Janka 2009; Roberts et al.
2010; Arcones & Montes 2011).
In order to obtain r-process conditions, a better chance
to eject neutron-rich matter is provided, when neutron-
rich matter stems from the initial collapse and com-
pression, where electron captures made it neutron-rich,
early in an explosion, before neutrino interactions have
the chance to turn it proton-rich in the neutrino wind.
Electron-capture supernovae, which explode without a
long phase of accretion onto the proto-neutron star, ap-
parently provide such conditions (Wanajo et al. 2011).
However, the Yeobtained under such conditions do not
support a full r-process. This kind of outcome also char-
acterizes the conditions we find in the prompt and quasi-
prompt ejecta of the present study, which did not expe-
rience a strong neutrino wind. However, the Ye-values
attained are not smaller than 0.33. Whether uncertain-
ties in the input or explosion physics can change this
down to values close to 0.23, necessary for obtaining the
third r-process peak, remains questionable.
Core-collapse supernovae exploding via the quark-
hadron phase transition (the focus of the present study)
or electron capture supernovae (Wanajo et al. 2011) both
lead to a rather prompt ejection of prior compressed
and neutronized mass zones. Given our results described
above, we conclude that these objects can contribute to
a weak r-process, consistent with observations of low
metallicity stars (Honda et al. 2006) and with LEPP
abundances (Travaglio et al. 2004). Such conditions ap-
parently do not occur in regular core-collapse super-
novae. However, strong r-process conditions, which also
produce the third r-process peak and the actinides, have
in simulations only materialized in neutron star merg-
ers (Freiburghaus et al. 1999b; Goriely et al. 2011), fast
rotating core collapse supernovae with strong magnetic
fields and jet ejecta (Cameron 2003; Nishimura et al.
2006; Fujimoto et al. 2008) or accretion disks around
(Fischer et al. 2010;
par-
black holes (Surman et al. 2008; Wanajo & Janka 2011).
Neutron star mergers have been shown to be powerful
sources of r-process matter, in fact ejecting a factor of
100 to 1000 more r-process material than required on av-
erage from core collapse supernovae, if those would have
to explain solar r-process abundances. This would actu-
ally support the large scatter of Eu/Fe in comparison to
e.g. O, Mg, Si, S, Ca/Fe, where the latter are clearly pro-
duced in supernovae. The only problem is that it might
be hard to explain the early appearance of r-process mat-
ter for metallicities at and below [Fe/H] = −3. Neu-
tron star mergers will appear after the first supernovae
have already produced Fe and studies by Argast et al.
(2004) showed that one expects r-process products only
for metallicities [Fe/H] = −3 ∼ −2. Some recent studies,
which include the fact that our Galaxy is possibly the
result from smaller merging subsystems (with different
star formation rates) have been expected to show a way
out of this dilemma. If this cannot solved, we need an-
other strong r-process source already at low metallicities,
and possibly jets from rotating core collapses with strong
magnetic fields could be the solution (Nishimura et al.
2006; Fujimoto et al. 2007).
ACKNOWLEDGMENTS
N.N. acknowledges to M. Hashimoto and S. Fujimoto
for early development of the nuclear reaction network
which are extended and used in this work and also
thanks to A. Arcones to critical and useful comments.
The project was funded by the Swiss National Science
Foundation grant. no. PP00P2-124879/1 and 200020-
122287, the Helmholtz Research School for Quark Mat-
ter Studies, and the Helmholtz International Center
(HIC) for FAIR. T.F. is supported by HIC for FAIR
project No. 62800075 and G.M.P is partly supported by
the Sonderforschungsbereich 634, the ExtreMe Matter
Institute EMMI and HIC for FAIR. C.F. acknowledges
support from the DOE Topical Collaboration ”Neutri-
nos and Nucleosynthesis in Hot and Dense Matter” under
contract DE-FG02-10ER41677. M.L. and M.H. acknowl-
edges support from the High Performance and High Pro-
ductivity Computing (HP2C) project. M.H. is supported
by the Swiss National Science Foundation (SNF) under
project number no. 200020-132816/1 and is also grateful
for participating in the ENSAR/THEXO project. T.R. is
supported by the European Commission within the FP7
ENSAR/THEXO project. The authors are additionally
supported by CompStar, a research networking program
of the European Science Foundation and EuroGENESIS,
a collaborative research program of the ESF. F.-K.T. is
an Alexander von Humboldt awardee.
REFERENCES
Alford, M., Braby, M., Paris, M. & Reddy, R. 2005, ApJ, 629,
969
Argast, D., Samland, M., Thielemann, F.-K. & Qian, Y.-Z. 2004,
A&A, 416, 997
Audi, G. & Wapstra, A. H. 1995, Nucl. Phys. A, 595, 409
Aoki, Y., Endrodi, G., Fodor, Z., Katz, S. D. & Szabo, K. K.
2006, Nature, 443, 675
Arcones, A., Janka, H.-Th. & Scheck, L. 2007, A&A, 467, 1227
Arcones, A., & Montes, F. 2011, ApJ, 731, 5
Bauswein, A., Janka, H.-Th., Oechslin, R., Pagliara, G.,
Sagert, I., Schaffner-Bielich, J., Hohle, M.M., Neuh¨ auser 2009,
Phys. Rev. Lett., 103, 011101
Bethe, H. A. & Wilson, J. R. 1985, ApJ, 295, 14
Bogolyubov, P. N. 1968, Annales Poincare Phys. Theor., 8, 163
Cameron, A. G. W. 2003, ApJ, 587, 327
Chodos, A., Jaffe, R. L., Johnson, K., Thorn, C. B. &
Weisskopf, V. F. 1974, Phys. Rev. D, 9, 3471
Dasgupta, B., Fischer, T., Horiuchi, S., Liebend¨ orfer, M.,
Mirizzi, A., Sagert, I. & Schaffner-Bielich, J. 2010,
Phys. Rev. D, 81, 103005

Page 12

12 Nishimura et al.
Demorest, P. B., Pennucci, T., Ransom, S. M., Roberts, M. S. E.
& Hessels, J. W. T. 2010, Nature, 467, 1081
Detar, C. E. & Donoghue, J. F. 1983, Annual Review of Nuclear
and Particle Science, 33, 235
Farouqi, K., Kratz, K.-L., Pfeiffer, B., Rauscher, T., Thielemann,
F.-K., & Truran, J. W. 2010, ApJ, 712, 1359
Fischer, T., Whitehouse, S. C., Mezzacappa, A.,
Thielemann, F.-K. & Liebend¨ orfer, M. 2010, A&A, 517, A80
Fischer, T., Sagert, I., Pagliara, G., Hempel, M.,
Schaffner-Bielich, J., Rauscher, T., Thielemann F.-K.,
K”appeli, R., Mart´ ınez-Pinedo, G. & Liebend¨ orfer, M. 2011,
Astrophys. J. Suppl., 194, 28
Freiburghaus, C., Rembges, J.-F., Rauscher, T., Kolbe, E.,
Thielemann, F.-K., Kratz, K.-L., Pfeiffer, B. & Cowan, J. J.
1999a, ApJ, 516, 381
Freiburghaus, C., Rosswog, S., Thielemann, F.-K. 1999b, ApJ,
525, L121
Freire, P. C. C. & Wex, N. 2010, print (astro-ph/1006.0642)
Fr¨ ohlich, C. et al. 2006a, ApJ, 637, 415
Fr¨ ohlich, C., Mart´ ınez-Pinedo, G., Liebend¨ orfer, M.,
Thielemann, F.-K., Bravo, E., Hix, W.-R., Langanke, K. &
Zinner, N. T. 2006b, Phys. Rev. Lett., 96, 142502
Fujimoto, S., Hashimoto, M., Kotake, K., & Yamada, S. 2007,
ApJ, 656, 382
Fujimoto, S., Nishimura, N. & Hashimoto, M. 2008, ApJ, 680,
1350
Goriely, S., Bauswein, A., & Janka, H.-T. 2011, arXiv:1107.0899
Hempel, M, Pagliara, G., Schhaffner-Bielich, J. 2009
Phys. Rev. D, 80, 125014
Hoffman, R. D., Woosley, S.E., Qian, Y.-Z. 1997 ApJ, 482, 951
Honda, S., Aoki, W., Ishimaru, Y., Wanajo, S. 2006, ApJ, 643,
1180
H¨ udepohl, L., M¨ uller, B., Janka, H.-T., Marek, A., & Raffelt,
G. G. 2010 Physical Review Letters, 104, 251101
Kodama, T. & Takahashi, K. 1975, Nucl. Phys. A, 239, 489
Kuroda, T., Wanajo, S., Nomoto, K. 2008, ApJ, 672, 1068
Lattimer, J M. & Swesty, D. F. 1991, Nuclear Physics A, 535,
331
Liebend¨ orfer, M., Mezzacappa, A. & Thielemann, F.-K., 2001a,
Phys. Rev. D, 63, 104003
Liebend¨ orfer, M., Rosswog, S. and Thielemann, F.-K., Messer,
O. E. B., Hix, W. R. & Bruenn, S. W. 2001b, Phys. Rev. D,
63, 103004
Liebend¨ orfer, M., Rosswog, S. & Thielemann, F.-K., 2002,
ApJS, 141, 229
Liebend¨ orfer M., Mezzacappa, A., Messer, O. E. B.,
Mart´ ınez-Pinedo, G., Hix, W. R. & Thielemann, F.-K. 2003,
Nuclear Physics A, 719, 144
Liebend¨ orfer, M., Messer, O. E. B., Mezzacappa, A., Bruenn,
S. W., Cardall, C. Y. & Thielemann, F.-K. 2004, ApJS, 150,
263
Meyer, B.S. & Brown, J. 1997, ApJ, 112, 199
Mezzacappa, A. & Bruenn, S. W. 1993a, ApJ, 405, 637
Mezzacappa, A. & Bruenn, S. W. 1993b, ApJ, 405, 669
Mezzacappa, A. & Bruenn, S. W. 1993c, ApJ, 410, 740
M¨ oller, P., Nix, J. R., Myers, W. D., & Swiatecki, W. J. 1995,
Atomic Data and Nuclear Data Tables, 59, 185
Nishimura, S., Kotake, K., Hashimoto, M., Yamada, S.,
Nishimura, N., Fujimoto, S., & Sato, K. 2006, ApJ, 642, 410
Nomoto, K., Tominaga, N., Umeda, H., Kobayashi, C., Maeda, K.
2006, Nuclear Physics A, 777, 424
Panov, I. V., & Janka, H.-T. 2009, A&A, 494, 829
Pruet, J., Woosley, S. E., Buras, R., Janka, H.-T. & Hoffman,
R. D. 2005, ApJ, 623, 325
Pruet, J., Hoffman, R. D., Woosley, S. E., Janka, H.-T. & Buras,
R. 2006, ApJ, 644, 1028
Qian, Y.-Z. & Woosley, S. E. 1996 ApJ, 471, 331
L.F. Roberts, S.E. Woosley, R.D. Hoffman 2010, ApJ, 722, 954
Sagert, I., Fischer, T., Hempel, M., Pagliara, G.,
Schaffner-Bielich, J., Mezzacappa, A., Thielemann, F.-K. &
Liebend¨ orfer, M. 2009, Phys. Rev. Lett., 102, 081101
Sagert, I., Fischer, T., Hempel, M., Pagliara, G.,
Schaffner-Bielich, J., Thielemann, F.-K. & Liebend¨ orfer, M.
2010, Journal of Physics G Nuclear Physics, 37, 094064
Sagert, I., Schaffner-Bielich, J. & Sturm, C. 2010, Work in
progress
Schertler, K., Greiner, C., Schaffner-Bielich, J. & Thomas, M. H.
2000, Nuclear Physics A, 677, 463
Shen, H., Toki, H., Oyamatsu, K. & Sumiyoshi, K. 1998,
Nuclear Physics A, 637, 435
Staudt, A. & Klapdor-Kleingrothaus, H. 1992, Nucl. Phys. A,
549, 254
Surman, R., McLaughlin, G. C., Ruffert, M., Janka, H.-T., & Hix,
W. R. 2008, ApJ, 679, L117
Takahashi, K., Witti, J. & Janka, H.-T. 1994, A&A, 286, 857
Thielemann, F.-K., Hashimoto, M. & Nomoto, K. 1990, ApJ,
349, 222
Thielemann, F.-K., Nomoto, K. & Hashimoto, M. 1996, ApJ,
460, 408
Thielemann, F.-K., Brachwitz, F., H¨ oflich, P.,
Mart´ ınez-Pinedo, G. & Nomoto, K. 2004, New Astronomy
Review, 48, 605
Thielemann, F.-K., Hirschi, R., Liebend¨ orfer, M. & Duehk, R.
2011, Lecture Notes in Physics, 812, 153
Thompson, C. & Duncan, R. C. 1993, ApJ, 408, 194
Thompson, T. A.,Burrows, A. & Meyer, B. S. 2001, ApJ, 562,
887
Thompson, T A., Burrows, A. & Pinto, P A. 2003, ApJ, 592, 434
Timmes, F. X. & Arnett, D. 1999, ApJS, 125, 277
Travaglio, C., Gallino, R., Arnone, E., Cowan, J.J., Jordan, F.,
Sneden, C. 2004, ApJ, 601, 864
Wanajo, S., Kajino, T., Mathews, G.J. & Otsuki, K. 2001, ApJ,
554, 578
Wanajo, S. 2006, ApJ, 647, 1323
Wanajo, S., & Janka, H.-T. 2011, arXiv:1106.6142
Wanajo, S., Janka, H.-T., M¨ uller, B. 2011, ApJ, 726, L15
Weissenborn, S., Sagert, I., Pagliara, G., Hempel, M. &
Schaffner-Bielich, J. 2011, print (astro-ph/1102.2869)
Woosley, S. E. Wilson, J. R., Mathews, G. J., Hoffman, R. D.,
Meyer, B. S. 1994, ApJ, 433, 229
Woosley, S. E. & Weaver, T. E. 1995, ApJS, 101, 181
Woosley, S. E., Heger, A. & Weaver, T. E. 2002, Reviews of
Modern Physics, 74, 1015
Woosley, S. E. & Heger, A. 2007, Physics Reports, 442, 269