arXiv:0801.1828v1 [astro-ph] 11 Jan 2008
Nucleosynthesis in Early Neutrino Driven Winds
R.D. Hoffman∗, J.L. Fisker∗, J. Pruet†, S.E. Woosley∗∗, H.-T. Janka‡and R. Buras‡
∗Lawrence Livermore National Laboratory, PO Box 808, L-414, Livermore, CA 94550 USA
†Lawrence Livermore National Laboratory, PO Box 808, L-059, Livermore, CA 94550 USA
∗∗Department of Astronomy & Astrophysics, UC Santa Cruz, Santa Cruz, CA 95064 USA
‡Max Plank Institute for Astrophysics, Karl-Schwarzschild-Str. 1, 85741 Garching, Germany
Abstract. Two recent issues realted to nucleosynthesis in early proton-rich neutrino winds are investigated. In the first part
we investigate the effect of nuclear physics uncertainties on the synthesis of92Mo and94Mo. Based on recent experimental
results, we find that the proton rich winds of the model investigated here can not be the only source of the solar abundance of
92Mo and94Mo. In the second part we investigate the nucleosynthesis from neutron rich bubbles and show that they do not
contribute to the nucleosynthesis integrated over both neutron and proton-rich bubbles and proton-rich winds.
Keywords: supernovae, nucleosynthesis
PACS: 21.10.Dr, 26.30+k, 26.50+x, 27.60+j
Over the past decadeimprovementsin neutrino-transport
and multi-dimensional computer simulations have lead
to a new understanding of the conditions that lead to nu-
cleosynthesis of the elements aboveiron in core-collapse
supernovae. Immediately following the bounce on the
proto-neutronstar, the shock fully photodisintegrates the
infalling material turning it into electron–position pairs,
neutrons, and protons. As the nascent neutron star con-
tinues to collapse it liberates 1053ergs over the span of
∼ 10 seconds primarily in the form of neutrinos. This
enormous neutrino flux is deposited in the low density
region of photodisintegrated matter inside the gain ra-
dius between the neutron star and the accretion shock of
the still infalling material and heats it to temperatures in
excess of 10 billion K while driving mass away in the
form of a neutrino wind theoretically leading to the ex-
plosionof the supernova. The strongflux of neutrinos
and anti-neutrinos results in a detailed balance between
protons and neutrons that favors the lighter mass pro-
tons depending on the respective neutrino spectra lead-
ing to an electron fraction that is proton-rich (Ye> 0.5)
[2, 3]. These protons and neutrons recombine into alpha
particles that proceed via the α(αn,γ)9Be(α,n)12C-
reactionsfollowedbya series of (α,γ)-reactions orcom-
bined (α,p)(p,γ)-reactions along N = Z into the iron
group, primarily56Ni and60Zn which form the seeds of
the subsequent nucleosynthesis.
From this point the resulting nucleosynthesis in the
neutrino-driven wind essentially depends on the num-
ber of seed nuclei to the number of excess neutrons
or protons that were frozen out and did not turn into
seed nucleii (Ye), the entropy per baryon, the expansion
timescale of the ejecta and the amount of the ejecta.
As the explosion evolves, an ejected mass element in-
herits some combination of these parameters and below
∼ 0.5MeV they remain fairly constant as the matter pro-
ceeds to freeze out.
In this paper, we consider the early times when the
wind still contains a proton excess because the rates for
those for the inverse captures on protons. We consider
two interesting problems which are discussed in the fol-
lowing two sections.
THE PUZZLE OF92Mo
The origin of92Mo is a long standing puzzle of nucle-
osynthesis [for reviews, see 4, 5]. It is thought to origi-
nate in the proton-richwind prior to the r-process in core
collapse supernovae, but historically it has been under-
produced in such models or subject to severe model con-
straints [6, 7].
Recent supernovamodelsshowthat theYe≡∑XiZi/Ai
of the innermost ejecta is greater that the Yeof the most
abundant p-nuclei [1, 3]. This implies the existence of
surplus protons which allow the production of proton-
rich p-nuclei nuclei by the νrp-process . However,
similar to the rp-process in the X-ray burst scenario,
there is an important waiting point at64Ge which backs
up material beyond the t < 1s dynamic timescale of the
innermost ejecta in core collapse .
To solve this problem, it was suggested a new νp-
process in which neutrinos convert some of the surplus
protons into neutrons allowing the waiting points to be
bridged via an (n,p)-reaction . This accelerates the
flow into heavier elements and creates the light p-nuclei
between Zr and Cd when T9= 2.06,ρ5= 2.74,and Ye= 0.561
showng nuclear flows in the A ∼ 90 region. Each isotope is
labled according to its proton separation energy The arrows
indicate the dominant net nuclear flows. All net flows within a
factor of 50 of the largest flow in this figure (84Nb(p,γ)85Tc =
4.5×10−5s−1) are shown. The most important flows affecting
92,94Mo are the proton capture flows on92Ru and93Rh.
A closeup of Figure 8 in  for the region
These calculationswereindependentlyconfirmedby cal-
culations based on simulations [8, 11].
Still, relative to the solar abundances, both calcula-
tions show underproductionof92Mo (the most abundant
of the p-nuclei) relative to the p-nuclei of Ru and Pd.
There are three possible reasons why92Mo is not co-
produced with the other p-nuclei: 1) The νp-process is
active, but92Mo is primarily synthesized at other sites.
2) The νp-process is not active, so another explanation
is needed. 3) The νp-process is active, but the nuclear
parameters that enter the nucleosynthesis calculation are
incorrect. In this paper, we investigate the third possibil-
The production of the light p-nuceli
Nucleosynthesis results obtain from the sum total of
the reaction flow in all the matter trajectories of the
supernova ejecta. Here we only consider the reaction
flow in “trajectory 6” (see Table 2 of ) based on the
model of  (see  for specific code details and 
for more details). “Trajectory 6” is the trajectory where
neutrino interactions are the most important in making
the p-nuclei between Sr and Pd.
The νp-process starts on the iron groupbut it is halted
at the long-lived64Ge waiting point which is known to
be bridged by an (n,p)-reaction allowing the νp-process
to continue . The flow from64Ge passes through
all even-even Tz= (N −Z)/2 = 0 isotopes until88Ru
is reached . As fig. 1 shows, the pattern is broken
because of the low proton separation energy of90Ru
that prevents immediate proton captures up to92Pd. In-
stead the flow proceeds via90Ru(n,p)90Tc(p,γ)91Ru.
A (p,γ)-reaction would result in the92Rh progenitor
provided it does not get destroyed by another (p,γ)-
reaction. Alternatively, an (n,p)-reaction to91Tc fol-
lowed by a (p,γ)-reaction would result in the92Ru pro-
genitor once again provided it does not get destroyed by
another (p,γ)-reaction. In both cases the reverse reac-
tions from93Pd and93Rh would increase the survival of
the A = 92 progenitors.
Many of the relevant reaction rates, spins, partition
functions, and proton separation are not known experi-
mentally and the theoretical values are subject to con-
siderable uncertainties which may change the flow. For
instance, a 50% yield increase in92Mo was found after a
plausible 1MeV increase in the proton separation energy
We systematically investigated the effect relevant nu-
clear uncertainties on this reaction flow using the model
described in [8, 13]. We find that variation within current
uncertainties  of the91Rh proton separation energy
and the92Rh proton separation energy does not change
the solar abundance ratio of92Mo to94Mo whereas the
ratio is highly sensitive to the proton separation energy
of93Rh. Fig. 2 shows the dependence of the solar ratio
92Mo to94Mo to variations in entropy of “trajectory 6”.
We show that Sp(93Rh) = 1.63 MeV is a solution to a
range of entropy variations between 0.8 and 1.6 of the
nominal value. The figure also shows no solution above
Sp(93Rh) = 1.71 MeV.
Fig. 3 shows the dependence of the solar ratio92Mo
to94Mo to variations in entropy in “trajectory 6” as a
function of Yeand Sp(93Rh). The figure also shows the
solutions where92Mo and94Mo are co-produced within
a factor 4,5 and 7. Isotopes produced with precisely the
solar abundance pattern have equal production factors.
A co-production factor of no more than 7 is typically
regarded as acceptable as the global characteristics of
nucleosynthesis are sensitive to details of the outflow.
The conclusion that the92Mo and94Mo ratio is pre-
dominantly influenced by Sp(93Rh) has been shown
to be robust (Fisker et al., submitted for publication).
However, our calculations predict that Sp(93Rh) = 1.63
MeV whereas recent experimental results suggest that
Sp(93Rh) = 2.0001±0.008 MeV . This leads to the
tentative conclusion that proton rich winds under the
conditions in the model investigated here can not be the
sole source of the solar92Mo and94Mo.
0.6 0.8 1 1.2 1.4 1.6 1.8 2
of changes in entropy, S relative to the entropy of “trajectory
6”, S0= 77, in the outflowing wind for the solar ratio of
The allowed values of SP(93Rh) as a function
THE CONTRIBUTION FROM NEUTRON
Using the same supernova model as above, the contri-
bution to core-collapse nucleosyntheis of the proton-rich
bubblesandproton-richwinds was investigatedin[13, 8]
However, some bubbles also contains neutron-rich mat-
ter that is ejected in coincidence with the proton-rich
bubbles. Here, investigate their contribution to the over-
all nucleosynthesis by considering newly extrated trajec-
tories with 0.47 ≤Ye≤ 0.50.
For Yecloser to 0.5, primarily56,57,58Ni are formed.
The flow from these nuclei leads to64Ge. Unlike the νp-
process , there is not a sufficient amount of protons
left at this time for neutrinos provide sufficient numbers
of neutrons to capture on64Ge and thus move beyond
this waiting point. As a result, heavier isotopes are not
co-produced with the62Ni and64Zn isotopes. In partic-
ular, there is no overproduction of the light p-nuclei for
Ye≤ 0.5. For Yecloser to 0.47, primarily58,59,60Ni are
1.5 1.6 1.7 1.8 1.9 2 2.1
Co-production factor for 92Mo = 4
Co-production factor for 92Mo = 5
Co-production factor for 92Mo = 7
Co-production factor for 94Mo = 4
Co-production factor for 94Mo = 5
Co-production factor for 94Mo = 7
SP(93Rh) where the92Mo/94Mo ratio in the outgoing wind
matches the solar ratio. Error bars indicate the extent of similar
lines for ratios of 1.54 and 1.59. Also shown are the solutions
where92Mo and94Mo are coproduced within a factor 4, 5,
and 7. A solution is found for a co-production factor of 5 with
Ye=0.555 and SP(93Rh) = 1.72 (see main text for details).
The solid line shows the solution for Ye and
formed. This means that the64Ge waiting point is cir-
cumvented which leads to overproduction of74Se,78Kr,
and92Mo whichis co-producedwith64Zn.Withdecreas-
ing Ye,92Mo production falls off and the overproduction
of N=50 nuclei ensues.
The figure shows the integrated production factors for
all studied neutron-rich bubble trajectories. The most
produced isotopes in the neutron-richparts of the bubble
relative to solar abundances are62Ni and64Zn which
originate in bubbles with Ye closer to 0.5. These are co-
producedalongwith74Se and78Kr whichoriginatein the
bubbles with Ye closer to 0.47. The neutron-richbubbles
add74Se,78Kr, and92Mo to the bubble-outflow, but
this contribution is much smaller than the contribution
from the proton-rich winds when neutrino interactions
are included.The neutron-richbubbles also add62Ni and
64Zn to the total outflow but only in comparableamounts
tothe windoutflowsandthe proton-richbubbleoutflows.
ries of the convective bubble ejecta. The most abundant iso-
tope for a given element is shown with an asterisk. Diamonds
indicate that the isotope was made primarily as a radioactive
Production factors of the neutron-rich trajecto-
Our results show that the overproduction factors of the
neutron-rich bubbles folded with the mass-ejecta does
not contribute significantly to the nucleosynthesis of the
light p-nuceli compared to the nucleosynthesis of the
This work was performed under the auspices of the U.S.
Department of Energy by Lawrence Livermore National
part under Contract DE-AC52-07NA27344. It was also
in Garching was supported by the Deutsche Forschungs-
gemeinschaft through the Transregional Collaborative
Research Centers SFB/TR 27 “Neutrinos and Beyond”
and SFB/TR 7 “GravitationalWave Astronomy”,and the
Cluster of Excellence EXC 153 “Origin and Structure
of the Universe”. The SN simulations were performed
on the national supercomputer NEC SX-8 at the High
Performance Computing Center Stuttgart (HLRS) under
grant number SuperN/12758.
1. Y.-Z. Qian, and S. E. Woosley, Astrophys. J. 471, 331–351
M. Liebendörfer, A. Mezzacappa, O. E. B. Messer,
G. Martínez-Pinedo, W. R. Hix, and F.-K. Thielemann,
Nucl. Phys. A719, 144 (2003).
C. Fröhlich, P. Hauser, M. Liebendörfer, G. Martínez-
Pinedo, F.-K. Thielemann, E. Bravo, N. T. Zinner, W. R.
Hix, K. Langanke, A. Mezzacappa, and K. Nomoto,
Astrophys. J. 637, 415–426 (2006).
D. L. Lambert, Astron. Astrophys. Rev. 3, 201–256 (1992).
B. S. Meyer, Ann. Rev. Astron. Astrophys. 32, 153–190
G. M. Fuller, and B. S. Meyer, Astrophys. J. 453, 792–809
R. D. Hoffman, S. E. Woosley, G. M. Muller, and B. S.
Meyer, Astrophys. J. 460, 478–488 (1996).
J. Pruet, R. D. Hoffman, S. E. Woosley, H.-T. Janka, and
R. Buras, Astrophys. J. 644, 1028–1039 (2006).
R. K. Wallace, and S. E. Woosley, Astrophys. J. Suppl. 45,
10. C. Fröhlich, G. Martínez-Pinedo, M. Liebendörfer, F.-K.
Thielemann, E. Bravo, W. R. Hix, K. Langanke, and N. T.
Zinner, Phys. Rev. Lett. 96, 142502 (2006).
11. H.-T. Janka, R. Buras, and M. Rampp, Nucl. Phys. A718,
12. M. Rampp, and H.-T. Janka, Astron. Astrophys. 396,
13. J. Pruet, S. E. Woosley, R. Buras, H.-T. Janka, R. Buras,
and R. D. Hoffman, Astrophys. J. 623, 325–336 (2005).
14. G. Audi, A. H. Wapstra, and C. Thibault, Nucl. Phys. 729,
15. V. Elomaa, R. Ferrer, C. Weber et al., in preparation.