Chemical evolution with rotating massive star yields II. A new assessment of the solar s- and r- process components

arXiv:1911.02545v1 [astro-ph.GA] 6 Nov 2019 MNRAS 000,1–22 (2019) Preprint 7 November 2019 Compiled using MNRAS L A T EX style file v3.0 Chemical evolution with rotating massive star yields II. A new assessment of the solar sand rprocess components N. Prantzos,1⋆C. Abia,2S. Cristallo3,4M. Limongi,5,6A. Chieffi7,8 1Institut d’Astrophysique de Paris, UMR7095 CNRS, Sorbonne Universit´e, 98bis Bd. Arago, 75104 Paris, France 2Departmento de F´ısica Te´orica y del Cosmos, Universidad de Granada, E-18071 Granada, Spain 3Istituto Nazionale di Astrofisica - Osservatorio Astronomico d’Abruzzo, Via Maggini snc, I-64100, Teramo, Italy 4Istituto Nazionale di Fisica Nucleare - Sezione di Perugia, Via Pascoli, I-06123, Perugia, Italy 5Istituto Nazionale di Astrofisica - Osservatorio Astronomico di Roma, Via Frascati 33, I-00040, Monteporzio Catone, Italy 6Kavli Institute for the Physics and Mathematics of the Universe, Todai Institutes for Advanced Study, the University of Tokyo, Kashiwa, Japan 277-8583 (Kavli IPMU, WPI) 7Istituto di Astrofisica e Planetologia Spaziali, INAF, via Fosso del cavaliere 100, 00133 Roma - Italy 8Monash Centre for Astrophysics (MoCA), School of Mathematical Sciences, Monash University, Victoria 3800, Australia Accepted XXX. Received YYY; in original form ZZZ ABSTRACT The decomposition of the Solar system abundances of heavy isotopes into their sand rcomponents plays a key role in our understanding of the corresponding nuclear processes and the physics and evolution of their astrophysical sites. We present a new method for determining the sand rcomponents of the Solar system abundances, fully consistent with our current understanding of stellar nucleosynthesis and galactic chemical evolution. The method is based on a study of the evolution of the solar neighborhood with a state-of-the-art 1-zone model, using recent yields of low and intermediate mass stars as well as of massive rotating stars. We compare our results with previous studies and we provide tables with the isotopic and elemental contributions of the sand r-processes to the Solar system composition. Key words: Galaxy: abundances – Galaxy: evolution – Nucleosynthesis – Sun: abundances – Stars: abundances 1 INTRODUCTION In their compilation and analysis of Solar system isotopic abundances Suess & Urey (1956) were the first to notice that, if heavier than Fe nuclei are formed by successive capture of neutrons, one should expect two abundance peaks for each of the regions near magic neutron numbers: a sharp one at the position of the magic nucleus, from material pilled up there due to the low neutron capture cross-section when neutron captures take place near the β-stability valley; and a smoothed one at a few mass units below, from material made by neutron captures occurring in the neutron-rich side of the stability valley and radioactively decaying after the end of the process. Building on that compilation, Burbidge et al. (1957) worked out the details of the two nucleosynthetic processes, which they called sand r-, respectively1. The former (slow) ⋆E-mail: pran[email protected] 1There are observational indications of intermediate density would occur on timescales long with respect to the lifetimes of radioactive nuclei along the neutron path, i.e. tens to thousands of years, as a result of low neutron densities Nn∼106÷107cm−3. The latter (rapid) would take place on short timescales of the order of 1 s, as a result of high neutron densities Nn>1024 cm−3.Burbidge et al. (1957) also noticed that, along the sprocess path (i.e. the valley of nuclear stability), the product of the neutron capture cross-section σA and the abundance NAof a nucleus with mass number A>70 is a smooth function of A, first declining up to A∼100 and then levelling off up to A=208. They attributed that feature to the operation of the s-process in two different regimes, the former one having ”not enough neutrons available per 56Fe neutron capture processes (i.e. between the sand rprocess), like the iprocess (Cowan & Rose 1977;Dardelet et al. 2014; Hampel et al. 2016), possibly occurring in rapidly accreting white dwarfs (Denissenkov et al. 2017), proton ingestion episodes in low-metallicity low-mass asymptotic giant branch (AGB) stars (Cristallo et al. 2016) or super-AGB stars (Jones et al. 2016). c 2019 The Authors 2Prantzos et al. nucleus to build the nuclei to their saturation abundances”, while the constancy of σANAin the latter is ”strongly suggestive of steady flow being achieved and of all of the nuclei reaching their saturation abundances”. Following the work of Weigert (1966), the environment provided by low and intermediate mass stars (LIMS) on their AGB phase was identified by Schwarzschild & H¨ arm (1967) and Sanders (1967) as a promising site for the operation of the s-process. Today, those stars are thought to produce the bulk of the s-isotopes above A∼90 during their thermal pulses, with neutrons released mainly by the 13C(α,n)16O reaction (see Straniero et al. 1995;Gallino et al. 1998 and references therein). On the other hand, Peters (1968) suggested that in the He-burning cores of massive stars, neutrons released by the 22Ne(α,n)25Mg reaction should also produce s-nuclei. Today, those stars are thought to produce the s-nuclei in the regime of ”few neutrons per 56Fe seed”, i.e. below A∼90 (Couch et al. 1974;Lamb et al. 1977; Busso & Gallino 1985): stellar models - including those of Prantzos et al. (1987) with mass loss - show that despite the large abundance of 22Ne, most of the released neutrons are captured by its progeny 25Mg and other abundant nuclei, leaving few neutrons to be captured by 56Fe (see Prantzos et al. 1990, for details of the ”neutron economy trio”, i.e. the roles of neutron sources, seed and poisons as function of metallicity in the case of massive stars). In contrast, in the thermally pulsing phase of AGBs, the periodic mixing of protons in the He-layer maintains the 13C source to a high abundance level - through 12C(p,γ)13C - and releases sufficient neutrons to reach the ”saturation regime”. Thus, both the mechanism(s) and site(s) of the sprocess are considered to be sufficiently well known (see e.g. K¨ appeler et al. 2011, and references therein). On the other hand, the situation with the site of the rprocess is still unsatisfactory. After more than fifty years of research on its astrophysical origin(s), the identification of a fully convincing site remains still elusive. An exhaustive description and discussion of experimental, observational and theoretical aspects of the r-process, as well as on the sites so far proposed is provided in the recent reviews of Cowan et al. (2019) and Thielemann et al. (2017). However, up to date, no numerical simulation in the proposed scenarios has been able to fully reproduce the observed distribution of the r-process elemental and isotopic abundances in the Solar system. Nowadays the neutron star merging (NSM) scenario is given support by the recent joint detection of electromagnetic and gravitational signal from the γ−ray burst GW170817/GRB170817A (see Pian et al. 2017, and references therein), and, in particular, by the identification of the neutron-capture element Sr in the spectrum of the associated kilonova AT2017gfo (Watson et al. 2019). However, it is not yet completely understood which component of those systems (dynamical, disk, ν-wind) dominates the nucleosynthesis, since any one of them may cover a wide range of chemical distributions, depending on the adopted input parameters (Rosswog 2015;Fern´andez & Metzger 2016;Wu et al. 2016;Perego et al. 2017). An additional important source of uncertainty comes from the nuclear inputs adopted to calculate the r-process nucleosynthesis, the most important ones being nuclear masses, β-decay rates and nuclear fission models (Eichler et al. 2015;Thielemann et al. 2017). Finally, the observed evolution of the r-elements in the Galaxy is hard (albeit not impossible) to conciliate with our current understanding of the occurrence rate of NSMs, regarding both the early (halo) and the late (disk) phases of the Milky Way (Tsujimoto & Shigeyama 2014;Ishimaru et al. 2015; Ojima et al. 2018;Cˆot´e et al. 2018;Hotokezaka et al. 2018; Cˆot´e et al. 2018;Guiglion et al. 2018;Wehmeyer et al. 2019; Siegel et al. 2019;Haynes & Kobayashi 2019;Cˆot´e et al. 2019). The decomposition of the Solar system abundances of heavy elements into their sand rcomponents, played and will continue to play a pivotal role in our understanding of the underlying nuclear processes and the physics and evolution of the corresponding sites. The s-contribution can be more easily determined, since isotopes dominated by the sprocess form close the β−stability valley. Their nuclear properties (β−decay half-times, nuclear cross sections, etc.) are more easily measured, while the astrophysical sites are better understood today. On the other hand, due to the large astrophysics and nuclear physics uncertainties related with the r-process, its contribution to the isotopic solar abundances has been so far deduced by a simple subtraction of the s-process contribution from the observed solar value. In this work, we present a new method for determining the sand r-components of the Solar system abundances. It is based on a global study of the evolution of the solar neighborhood with a state-of-the-art 1-zone model of galactic chemical evolution (GCE), which is presented in detail in Prantzos et al. (2018) - Paper I hereafter - and adopts recent stellar yields of rotating massive stars (from Limongi & Chieffi 2018) and of LIM stars (from Cristallo et al. 2015a). The plan of the paper is as follows: In §2, we review the various methods used so far in order to derive the scomponent of the isotopic abundances of the heavy nuclei, and we discuss their shortcomings. In §3, we present in detail our new method and its assumptions. In §4we present our results. We compare first the isotopic contributions to previous studies (§4.1) , as well as to the measured Solar system abundances taking into account the uncertainties of the latter (§4.3). We discuss the resulting σANAcurve in §4.2 and we derive the r-residuals in §4.4. In §4.5 we derive the elemental sand r-components, and finally in §5we summarized the main results of this study. 2 DETERMINATION OF SAND RABUNDANCES The ”classical” (or ”canonical”) s-process model was originally proposed by Burbidge et al. (1957) and developed by Clayton & Rassbach (1967). In this model two main assumptions are made: a) the s-process temperature is constant, allowing one to adopt well determined neutroncapture cross sections; b) nuclei on the s-process path are either stable (τβ>> τn) or sufficiently short-lived that the neutron capture chain continues with the daughter nucleus (τβ<< τn). This second assumption, however, is not valid at the s-process branchings (τβ∼τn), which requires a special treatment (see e.g. Kappeler et al. 1989). In addition, the classical model assumes that some stellar material composed by iron nuclei only is exposed to the superposition of 3 exponential distributions of the time-integrated neutron MNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 3 exposure, defined as τo=Rt oNnvTdt (where vTis the thermal neutron velocity at the temperature T). The 3 exponential distributions are usually referred to as the ”weak” component (responsible of the production of the 70 ⩽A⩽90 snuclei), the ”main”component (for the 90 ⩽A⩽204 isotopes) and the ”strong component” (for A>204). For long-enough exposures, the equations governing the evolution of the snuclei abundances result in equilibrium between the production and destruction terms, leading to a constant product, σANA, of neutron cross section and s-process abundance. Although this condition is not completely reached, the product σANAshows a very smooth dependence on mass number (see, e.g., Clayton 1968). Therefore, the product σANAfor a given isotope is fully determined by the cross section, after the parameters τoand the number of neutrons captured per 56Fe seed nucleus are fixed. The goal of the classical approach is to fix the empirical σANAvalues for the s-only isotopes, i.e. nuclei that are shielded against the r-process by the corresponding stable isobar with charge Z−1or Z−2 (see §3.1 for a discussion about our selection of s-only isotopes). Once the Solar system s-only distribution is fitted, the s-contribution for the rest of the ”mixed” isotopes (with both a sand r-contribution) are automatically obtained. Finally, the r-contribution is derived just subtracting this s-contribution Ns,Afrom the measured total Solar system abundance NA. This classical method has been used frequently in the literature, providing satisfactory results as the measurement of neutron cross sections have been improving during the years (see e.g. Kappeler et al. 1989;Sneden et al. 2008;K¨ appeler et al. 2011). However, the classical model is affected not only by observational and nuclear input data uncertainties, but also by the assumption that the s-process operates at a fixed constant temperature and neutron and electron density, and by the hypothesis that the irradiation can be considered as exponential one. To test the influence of these assumptions, Goriely (1999) (see also Arnould et al. 2007) developed the so-called ”multi-event”s-process, which constitutes a step forward in the canonical method. The multi-event approach assumes a superposition of a number of canonical events taken place in different thermodynamic conditions, namely: a temperature range 1.5⩽T(K)/108⩽4, neutron densities 7.5⩽log Nn(cm−3)⩽10 and a unique electron density Ne=1027 cm−3. Each canonical event is characterized by a given neutron irradiation on the 56Fe seed nuclei during a given time at a constant temperature and neutron density. These conditions try to mimic the astrophysical conditions characterizing the site of the s-process, although it is well known that temperature and neutron density are not constant during the s-process (see e.g. K¨ appeler et al. 2011, and references therein). The s-only nuclei abundance distribution obtained with that method is remarkably close to the solar observed one, because of the minimization procedure adopted in the selection of the aforementioned parameters. However, it presents non-negligible deviations from the classical method in the regions A⩽90 and A⩾204, mainly because the resulting neutron exposures in the multi-event model clearly deviate from exponentials. Within the multievent model it was possible to evaluate the major uncertainties (both nuclear and due to abundance measurements) affecting the prediction of the s-(r-)abundance distribution. Goriely (1999) concluded that the uncertainties in the observed meteoritic abundances and the relevant (n,γ)rates have a significant impact on the predicted s-component of the solar abundance and, consequently, on the derived rabundances, especially concerning the s-dominated nuclei (see also Nishimura et al. 2017 and Cescutti et al. 2018). Stellar models of LIM stars during the AGB phase and of massive stars during hydrostatic core He-burning and shell C-burning (the two widely recognized sites of the s-process), have shown that the interplay of the different thermal conditions for the 13C and 22Ne neutron sources is hardly represented by a single set of effective parameters constant in time (Busso et al. 1999;Straniero et al. 2006; Limongi & Chieffi 2018), such as those used in the classical (or the multi-event) approach. In an effort to overcome this shortcoming, the results of the ”stellar” model have been used to estimate the contributions of the sand r-process to the Solar system abundances. This method is based on post-processing nucleosynthesis calculation performed in the framework of ”realistic” stellar models. The first attempt to apply this method was made by Gallino et al. (1998) and Arlandini et al. (1999), and more recently by Bisterzo et al. (2010) with updated nuclear input. These authors showed that the solar s-process main component can be reasonably reproduced by a post-processing calculation from a particular choice (mass and extension) of the 13C pocket (the main neutron source in AGB stars) by averaging the results of stellar AGB models (Gallino et al. 1998) between 1.5 and 3 M⊙with [Fe/H]∼ −0.3. This model is particularly successful in reproducing the s-only nuclei solar abundances and showed general improvements with respect to the classical method, especially in the mass region A<88 (Arlandini et al. 1999). In fact, all these nuclei (mainly produced by the weak s-component) are synthesized in much smaller quantities. This difference is caused by the very high neutron exposures reached in the stellar model, which favor the production of heavier elements. In particular, at the s-termination path, 208Pb is produced four times more than in the classical approach. Nevertheless, the stellar model used to derive the physical inputs of post-process calculations are affected by several theoretical uncertainties. One of the less constrained physical mechanisms is the one leading to the formation of the 13C pocket, which forms at the base of the convective envelope after each TDU episode. Different processes have been proposed as responsible of the formation of such a pocket: convective overshoot (Herwig et al. 1997), gravity waves (Denissenkov & Tout 2003;Battino et al. 2016), opacity induced overshoot Cristallo et al. (2009) and mixing induced by magnetic mixing (Trippella et al. 2016). Other critical quantities are the mass fraction dredged-up after each thermal instability (third dredge up, TDU) during the AGB phase, and the mass-loss rate. Actually, the two processes are degenerate, since the number (and the efficiency) of TDUs is determined by the mass of the H-exhausted core and of the H-rich convective envelope, which in turn depend on the adopted mass-loss rate. However, AGB stellar models show that an asymptotic s-process distribution is reached after a limited number of pulses, so the mass-loss uncertainty mainly affects the total yield of the s-processed material, and not so much the shape of the resulting distribution (see e.g. Fig. 12 in Cristallo et al. 2015b). In the ”stellar” model, the r-residuals are calculated subtracting the arithmetic average of the 1.5 and 3 M⊙models at [Fe/H]∼ −0.3 MNRAS 000,1–22 (2019) 4Prantzos et al. (Z∼1/2Z⊙)2best reproducing the main s-component to the observed solar abundances. In Arlandini et al. (1999) the sand r-components obtained by the stellar model method are compared to the classical one for nuclei A>88, together with the corresponding uncertainty determined from the cross sections and solar abundances. Uncertainties in the sand r-residuals coming from the stellar model itself are, however, difficult to estimate. The massive star contribution to the solar sonly composition has been explored with non-rotating stellar models in e.g. Prantzos et al. (1990); Raiteri et al. (1993) and more recently, with rotating massive stars in Pignatari et al. (2008); Frischknecht et al. (2016); Choplin et al. (2017, 2018); Limongi & Chieffi (2018). Such models have their own uncertainties (mass loss, mixing, nuclear etc.). The role of rotation, in particular, is poorly explored and understood at present. The main reason is that the rotation driven instabilities are included in a parametric way, and this means that the efficiency with which fresh protons are ingested in the He-burning zone is not based on first principles but it is determined by two free parameters that must be calibrated. The calibration adopted in the models adopted in this paper is discussed in detail in Limongi & Chieffi (2018). Moreover, since the proton ingestion scales directly with the initial rotational velocity (and hence the neutron flux as well), the adopted initial distribution of rotational velocities (IDROV) plays a pivotal role: already in Paper I we have shown that at least the average rotational velocity of the stars must be limited to <50 km/s at metallicities [Fe/H]>−1, in order to avoid an overproduction of heavy nuclei, mainly in the Ba peak. But there are also other subtle indirect factors that may change the yields predicted by rotating models: in order to bring protons in an He active environment, at least part of the H rich mantle must be present while He is burning. A substantial change in the mass loss rate (e.g. due to the inclusion of a dust driven component to the mass loss rate or to the overcome of the Eddington luminosity) may affect the range of masses that retain a substantial fraction of the H rich mantle while the stars are in the central He-burning phase. Uncertainties of stellar models is one of the reasons why the validity of the stellar method has been questioned (see e.g. Arnould et al. 2007). Another one is that this method does not consider the solar s-(r-)process abundance distribution in an astrophysical framework, i.e. as the result of all the previous generations of stars which polluted the interstellar medium prior to the formation of the Solar system. In particular, these generations of stars covered a large range of metallicities and not a unique value (or even a limited range of values) of [Fe/H] as it is assumed in the classical and stellar methods. For instance, it is well known that at low metallicities a large neutron/seed ratio is obtained, leading to the production of the heaviest s-nuclei, while at high metallicities the opposite happens (see e.g. Travaglio et al. 2004, and references therein). The Solar system s-(r-) process abundances have to be understood in the framework of a galactic chemical evolu2We adopt here the usual notation [X/H]=log (X/H)⋆−log (X/H)⊙, where (X/H)⋆is the abundance by number of the element X in the corresponding object. tion (GCE) model. This is certainly a difficult task that requires a good understanding of the star formation history in the Galaxy, of stellar evolution, and of the interplay between stars and the interstellar gas, among other things. We are still far from fully understanding these issues. Therefore, this third method is based on a necessarily schematic description of the situation considering the chemical evolution of our Galaxy, accounting for the fact that the site(s) of the r-process have not been clearly identified yet. Attempts to obtain the sand rcomponents of the solar composition from a GCE model were pioneered by Travaglio et al. (2004), later updated by Serminato et al. (2009) and more recently by Bisterzo et al. (2014,2017). These authors employed a GCE code adopting s-process yields from AGB stellar models by Gallino et al. (1998) in a range of masses and metallicities (see these papers for details). Regarding the r-process yields, and for elements from Ba to Pb, they estimated the contribution to the Solar system by subtracting the s-residuals from the solar abundances. Then, they scale the r-process yields to the yield of a primary element (in a similar way we do here, see Eq. 4) mainly produced in core collapse supernovae, which they assumed to occur in the mass range 8−10 M⊙. They derived the weak s-process contribution from Raiteri et al. (1993). On the other hand, for the lighter elements, in particular for Sr-Y-Zr, they deduced the r-residuals and, thus, the r-process yields, from the abundance pattern found in CS 22892-052 (Sneden et al. 2002), by assuming that the abundance signatures of this star is of pure r-process origin (i.e., any contamination by other possible stellar sources is hidden by the r-process abundances). The Bisterzo et al. (2014) model resulted in good agreement with the Solar s-only isotopic abundances between 134,136Ba and 204Pb, also showing that the solar abundance of 208Pb is well reproduced by metal-poor AGB stars, without requiring the existence of a ”strong” component in the s-process as is done in the classical method. Below the magic number N=82, however, they found a significant discrepancy between the abundance distribution obtained with their GCE model and the Solar system values. It turned out that their GCE model underproduces the solar s-process component of the abundances of Sr, Y and Zr by ∼20%−30% and also the s-only isotopes from 96Mo up to 130Xe. This result prompted Travaglio et al. (2004) to postulate the existence of another source of neutron-capture nucleosynthesis named the light element primary process (LEPP). They argued that this process is different from the s-process in AGB stars and also different from the weak s-process component occurring in massive stars. The recent updates of this study by Bisterzo et al. (2014,2017), reach the same conclusion3. In particular, these two studies ascribe a fraction ranging from 8% to 18% of the solar Sr, Y and Zr abundances to this LEPP, and suggest that lighter elements from Cu to Kr could be also affected. On the other hand, they obtained a 3Bisterzo et al. (2014,2017) mainly focus on the impact of the different 13C pocket choices in AGB stars and weak s-process yields from massive stars, on the s-process residuals at the epoch of the Solar System formation. In Bisterzo et al. (2017) yields from massive stars are included considering the impact of rotation in a limited range of masses and metallicities according to the models by Frischknecht et al. (2016). MNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 5 r-process fraction at the Solar system ranging from 8% (Y) to 50% (Ru). The need of a LEPP has been recently questioned by Cristallo et al. (2015b) and later by Trippella et al. (2016) on the basis of a simple GCE model using updated s-process yields from AGB stars (Cristallo et al. 2011) and AGB stellar models only, respectively. These studies show that a fraction of the order of that ascribed to the LEPP in the predicted solar abundances of Sr, Y and Zr can be easily obtained, for instance, by just a moderate change in the star formation rate prescription in a GCE model, still fulfilling the main observational constrains in the solar neighbourhood. The same effect can be found by modifying AGB stellar yields as due to nuclear uncertainties, or the choice of the mass and profile of the 13C pocket. Introducing such a changes in the GCE models (i.e. stellar yields) one can easily account for the missing fractions of the solar abundance of these elements within the observational uncertainties. In addition, in Paper I we have very recently shown that the LEPP is not necessary when metallicity-dependent s-process yields from rotating massive stars (i.e. the ”weak” s-process) are considered in a GCE model. The stellar yields adopted in that paper are from an extended grid of stellar masses, metallicities and rotation velocities from Limongi & Chieffi (2018)4; for the first time in GCE studies, the IDROV was introduced through an empirically determined function of metallicity and rotation velocity. In this study, we use the GCE model of Paper I to derive the sand r-process contributions to the solar isotopic abundances in the full mass range from 69Ga to 235U through a new method. 3 THE METHOD 3.1 s-only and r-only isotopes The classification of nuclei belonging to the s-only group is not a trivial task. By definition, an s-only nucleus owes its entire abundance to the slow neutron capture process. As a consequence, we tentatively identify as s-only any nucleus with atomic number Z for which a stable isobar with atomic number Z-1 (or Z-2) exists: that isobar shields the nucleus from any r-process contribution. However, such a condition is necessary, but not sufficient to define an s-only isotope. In fact, there are isotopes lying on the proton-rich side of the β-stability valley, that are shielded from the r-process, but 4As stated in Paper I and Limongi & Chieffi (2018), the nuclear network for massive stars includes 335 isotopes in total, from H to209Bi, and is suited to properly follow all the stable and explosive nuclear burning stages of massive stars. The portion of the network from H to 98Mo takes into account all the possible links among the various nuclear species due to weak and strong interactions. For heavier nuclei, we consider only (n,γ) and β-decays. Since we are mainly interested in following in detail the flux of neutrons through all the magic number bottlenecks and since in the neutron capture chain the slowest reactions are the ones involving magic nuclei, between 98Mo and 209Bi we explicitly follow and include in the nuclear network, all the stable and unstable isotopes around the magic numbers corresponding to N=82 and N=126 and assume all the other intermediate isotopes at local equilibrium. Table 1. List of 30 s-only isotopes adopted in this work Z Isotope Possible contribution 32 70Ge 34 76Se 36 80Kr from p-process 36 82Kr 38 86Sr from p-process 38 87Sr from 87Rb 42 96Mo 44 100Ru 46 104Pd 48 110Cd 50 116Sn 52 122Te 52 123Te 52 124Te 54 128Xe 54 130Xe 56 134Ba 56 136Ba 60 142Nd 62 148Sm 62 150Sm 64 154Gd 66 160Dy 70 170Yb 71 176Lu radiogenic to 176Hf 72 176Hf radiogenic from 176Lu 76 186Os 78 192Pt 80 198Hg 82 204Pb may receive a non negligible contribution from the p-process (see Travaglio et al. 2015). Moreover, there are isotopes with unstable isobars with (Z-1), whose lifetimes are comparable to the age of the Universe: in that case, therefore, a delayed r-process contribution cannot be excluded (e.g. for the couples 87Sr-87Rb and 187Os-187Re). Finally, there are a few isotopes, with stable (Z-1) isobars, which may receive an important contribution from the neutrino process in core collapse supernovae (e.g. 113In and 115Sn; see Fujimoto et al. 2007). As a matter of fact, in the past different lists of s-only isotopes circulated in the literature. We list in Table 1the s-only isotopes considered in this study, including those that may receive a small contribution from other processes (see Travaglio et al. 2015). The definition of r-only isotopes in even more ambiguous. In principle, at odds with s-only nuclei (shielded by the r-process from stable isobars), there is no nucleus fully shielded by the s-process. In fact, all nuclei on the neutronrich side of the β-stability valley can receive a contribution (perhaps very small, but not null) from the s-process, depending on the activation of various branchings. For instance, net yields from AGB stars by Cristallo et al. (2015a) for isotopes marked as r-only in previous compilations (e.g. Goriely 1999 and Sneden et al. 2008) are all positive (from some % to significant fractions, depending on the isotope), apart from 130Te. In this study we shall not pre-define ”ronly” nuclei, but we shall explore with our method the contribution of our stellar yields to the abundances of all heavy isotopes. MNRAS 000,1–22 (2019) 6Prantzos et al. 3.2 Assumptions The method adopted in this study is based on a couple of key assumptions. Assumption 1: Our current understanding of stellar nucleosynthesis and galactic chemical evolution allows us to reproduce the pre-solar isotopic abundances to a precision of (a) a factor of ∼2for elements with charge 2 <Z<30 (between Li and Zn) but (b) to a factor of ∼20-30 % (or less) for the s-component of heavier elements. Statement (a) above is based on the fact that all calculations done up to now with ”state-of-the-art” stellar yields and models of the chemical evolution of the solar neighborhood show indeed a dispersion of a factor ∼2 around the solar value in the region up to the Fe-peak. This is true e.g. for the models of Timmes et al. (1995), who adopted yields of Woosley & Weaver (1995), Goswami & Prantzos (2000) with yields of Woosley & Weaver (1995), Kubryk et al. (2015) with yields of Nomoto et al. (2013) and Paper I with yields of Limongi & Chieffi (2018). Even if in each case the adopted models and yields differ considerably, the outcome is the same: a dispersion by a factor of ∼2 is always found, implying that uncertainties in the various parameters of the problem (regarding both stellar and galactic physics) remain important in the past two decades or so. Statement (b) is based on a limited sample of GCE models, namely those of Travaglio et al. (2004), Cristallo et al. (2015b) and Bisterzo et al. (2017) - see previous section - as well as our own model presented in Paper I. In those by Travaglio et al. (2004) and Bisterzo et al. (2017), the model values of most heavy pure s-nuclei barely exceeds the corresponding solar value and there is a systematic deficiency of ∼20-30% as one moves to lighter s-nuclei. This deficiency was interpreted as evidence for the need of another heavy isotope component, the so-called LEPP (see previous section). However, Paper I showed that rotating massive stars may produce through the weak s-process that ”missing”component, with no need for a new process. In that study, it is found that most pure s-nuclei are co-produced within ∼1020% from their pres-solar values, with only a few of them displaying higher values (up to 40% at most). We think that it is illusory at the present stage of our knowledge to reproduce the pre-solar pure s-composition to a higher accuracy. We believe however that it is possible to use this result and try to infer the solar sand r-components of all mixed (s+r) nuclei, as presented in Sec. 3.3. Assumption 2:The r-process is of ”primary” nature and, in particular, it mimics the behaviour of the ”alpha” process which produces α-elements like e.g. 16O. This assumption is based on the observational fact that pure relements, like Eu, display an α-like behaviour, i.e. the ratio [Eu/Fe] remains ∼constant at a value of ∼0.3-0.5 dex during the late halo evolution and then declines smoothly to its solar value at [Fe/H]∼0. This means that, in contrast to the s-process, which is basically of ”secondary” nature (i.e. the s-yields of both LIM stars and massive stars depend on the abundance of iron-seed nuclei), the r-yields are independent of the initial metallicity of their source. The ratio of those yields to the yields of α-isotopes should then be constant with metallicity. These inferences allow one to adopt r-process yields ”scaled” to the stellar model yields used in the GCE model. 3.3 Method Our ”bootstrap” method proceeds as follows: Step 0: We run a GCE model as in Paper I but using exclusively the s-component for all elements with Z>30. For that purpose we remove from the adopted yields the r-component (i.e. existing in the initial composition of the stars, through their scaled solar composition). In practice, we calculate a new set of yields as: yi(M,Z)=yi,0(M,Z)−fr,i,0Xi,⊙Z Me j(M,Z)(1) where: •yi,0(M,Z)are the original yields of isotope ifrom stars of mass Mand metallicity Z. •fr,i,0is the solar r-fraction of nucleus i, as provided i.e. in Sneden et al. (2008) or Goriely (1999). •Me j(M,Z)is the total mass ejected by the star of mass Mand metallicity Z. The results of the model at the time of Solar system formation (i.e. 4.56 Gyr before the end of the simulation) are stored as Wi,0=Xi,0/Xi,⊙(2) i.e., they are normalized to the corresponding Solar system isotopic abundances adopted from Lodders et al. (2009). These normalized abundances appear in the top panel of Fig. 1for all nuclei with charge Z>30. Among the s-only nuclei (red dots), most are reproduced within a factor of 20% solar abundances (see also Paper I 5), except Kr, Ba and Gd which differ from their solar values by 20-40 %.Taking into account the uncertainties in nuclear, stellar and galactic physics involved in the calculation, which lead to a larger dispersion for the lighter nuclei (up to 100 % , factor of ∼2, see Fig. 11 in Paper I), we think that this agreement is quite satisfactory. In particular, regarding the nuclear uncertainties, we note that our results are obtained with nucleosynthesis calculations using the set of neutron capture cross sections described in Straniero et al. (2006). Since then, two new cross sections became available, i.e. those of 176Lu (Wisshak et al. 2006a) and 176Hf (Wisshak et al. 2006b). Both cross sections are larger than those adopted to calculate our models, so that we expect a decrease for both isotopes (see Table 1), thus providing a better agreement with observations. We expect a similar behavior for 134Ba (and possibly 136Ba): both neutron capture cross sections will be measured in the next years at the n TOF facility (Guerrero et al. 2013). Moreover, we further stress that the abundance of 134Ba strongly depends on the activation of the branching at 134Cs, whose neutron capture cross section and temperature-dependent β-decay lifetime are rather uncertain. By varying theoretical nuclear inputs within uncertainties in a single model, we can obtain a decrease of about 15% and 12% for 134Ba and 5Notice that with respect to Paper I we have slightly reduced here the proportion of fast rotating massive stars (at 300 kms−1) in our mixture, in order to avoid an overproduction of the lighter s-only nuclei like 70Ge and 76Se; this reduction affects correspondingly the results of 80,82Kr (compare e.g. to Fig. 11 in Paper I) but no other nuclei, either lighter or heavier ones), since they are essentially produced by LIM stars. MNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 7 Figure 1. Top: Model results W0after Step 0, without r-component (see text); horizontal dashed lines indicate levels of ±10% and a factor of 2 deviation from solar; 2nd from top: Adopted initial r-fractions from Sneden et al. (2008); 3rd from top: Results W1after Step 1, with r-component introduced from Sneden et al. (2008) ; horizontal dashed lines indicate levels of ±10% deviation from solar; 4th from top: Our r-component after Step 1 is obtained as r1=1−W0/W1(where the corresponding s-component is obtained first as s1=W0/W1) and is introduced in the next iteration; Bottom: Same as the 3d panel, after the final (N=17 here) iteration of our ”bootstrap” method; horizontal dashed lines indicate levels of ±10% deviation from solar. Note that the scale in the Y-axis changes in the different panels.The names of the elements with sonly isotopes are indicated in the bottom panel. 136Ba, respectively (see also Cristallo et al. 2015b;Goriely 1999). All the above concern s-only isotopes which are overproduced with respect to their pre-solar system values in Fig. 1. As for the s-only isotopes that are under-produced with respect to the solar distribution, we stress that the neutron capture cross section of 82Kr is quite uncertain at typical s-process temperatures (∼25% at 8 keV; KADONIS database6). Moreover, it has to be stressed that the solar Kr and Xe abundances are not directly measured in the Sun, but they ” are based on theoretical values from neutron-capture element systematics” (Lodders 2003). On the other hand, the synthesis of 154Gd is strongly affected by the branching at 154Eu. Its neutron capture cross section has never been measured and its temperature-dependent β-decay lifetime is uncertain by a factor of three (Goriely 1999). Note that for 6https://exp-astro.de/kadonis1.0/ the decay, no hints on its trend between 5×107K and laboratory temperature is provided in Takahashi & Yokoi (1987). As already done for barium isotopes, if we just vary theoretical nuclear inputs within uncertainties, we can obtain an increase of about 25% for 154Gd7. Finally, it should be emphasized that we did not make any attempt to adjust the parameters of the GCE model (distribution of stellar rotational velocities, initial mass function or star formation and infall rates) as to optimize the s-only distribution; as discussed in Paper I, our GCE model is tuned in order to reproduce as well as possible local parameters like the current gas fraction, the metallicity distribution and age-metallicity relation and the abundances of major elements like O and Fe at Solar system formation. 7Note that the 154Gd neutron capture cross section has been recently measured at the n TOF facility (Massimi et al., in preparation) MNRAS 000,1–22 (2019) 8Prantzos et al. Despite that, we find that the parameter g=exp          1 nS nS X Z,A ln2Ncal(Z,A) N⊙(Z,A)          1/2 (3) where the sum runs over the nS=30 s-only nuclei (of charge Z and mass A, see Table 1) is g=1.18, i.e. it is not much higher than the value of 1.10 obtained in Goriely (1999). This author optimized the few parameters of his multi-event model as to minimize g, while we did not attempt such an optimization here. (with a classical analysis Goriely (1999) found g=1.44). Although our results are obtained with a different method and data (nuclear cross sections, stellar conditions, solar abundances), we believe that our result regarding the s-only distribution is quite reasonable and constitutes a good starting point for our GCE method. We discuss a little more our distribution of the s-component of our GCE model in Sec. 4, where we present the resulting σANAdistribution. We emphasize here that, in contrast to the GCE method of Travaglio et al. (2004); Bisterzo et al. (2014,2017) we do not proceed directly after the first run to the evaluation of the solar s-component by subtracting our results from the solar composition. This might lead to the need of a LEPP to justify the underproduction of several s-only nuclei, as the aforementioned GCE studies did. We proceed in a different way, allowing us to keep the ”s-only”property of the isotopes of Table 1 and at the same time evaluate self-consistently the s-fraction of the mixed (s+r) isotopes. For that, we need to introduce a priori their r-fractions, as described below. Step 1: We run a model by using now the original stellar yields yi,0(M,Z)and introducing this time the r-component of each isotope as in Paper I, namely by assuming that it is co-produced with a typical product of massive stars like 16O, i.e. the new yield for massive stars (M>10 M⊙) is yi,1(M,Z)=yi,0(M,Z)+fr,i,0y16O(M,Z)Xi,⊙/X16O,⊙(4) where the last term represents the rcomponent of the yield and fr,i,0is an ”educated guess”for the solar rfraction of isotope i; we start by adopting the rfractions of Sneden et al. (2008) but our results are independent of that choise (see below). The underlying physical assumption of Eq. 4is that 16O and the r-component have the same source, namely massive stars and this implicit assumption allows one to reproduce naturally the observed alpha-like behaviour of elements that are mostly of r-origin, like e.g. Eu. The method can be used in essentially the same way in the case that the main source of r-process turns out to be a rare class of massive stars, like collapsars (see e.g. Siegel et al. 2019, and references therein). In that case a stochastic treatment should be made, e.g. as applied for neutron star mergers in Ojima et al. (2018). If neutron star mergers are assumed to be the site of the rprocess, a different prescription should be used, involving the rate of occurrence of that site (through a delayed time distribution, as for SNIa, e.g. Cˆot´e et al. 2018) and the mass ejected in the form of isotope i, normalized as to get a solar abundance for the pure r-isotopes of Th and U. Notice that in this run we treat all nuclei except the s-only ones of Table 1 as mixed s+r: those classified as pure rin Sneden et al. (2008) or Goriely (1999) are also treated as such. They are simply given an initial r-fraction fr,i,0=1, which may change after Step 1. The result of the new run is also plotted in Fig. 1for all isotopes with charge Z>30 as overabundances Wi,1=Xi,1/Xi,⊙(5) where Xi,1=Xs,i,1+Xr,i,1(6) with Xs,i,1=Xs,i,0(the s-component remains the same) and Xr,i,1/Xi,⊙=C fr,i,0(7) is the r-component (proportional to the r-fraction fr,i)with the constant Cbeing the IMF average of the r-component term in Eq. 4and adjusted as to obtain at Solar system formation the exact solar abundances of pure r-isotopes, like Th, which we use here as benchmarks8. The value of Cdepends on the adopted ingredients of the GCE model (IMF, SF and infall rates, stellar yields) and it is ∼1.12 in our case. One notices that: •s-only isotopes are produced exactly at the same level as in step 0, since their yields have not been modified. •r-only isotopes with the meaning discussed in §3.1 are produced exactly at their pre-solar abundances -because of the adopted normalization in Eq. 4and 7except a few of them which have received a non-zero contribution from the s-process in step 0 (see green symbols in top panel of Fig. 1) and are now slightly overproduced. The most prominent of them are 76Ge (by ∼25%), 82Se (∼10%), 96Zr (∼30%) and 142Ce (∼15%), as shown in the 3rd panel of Fig. 1. •isotopes of mixed (s+r) origin are nicely co-produced w.r.t. their pre-solar abundances, to better than 10% in general, although in some regions (A∼205,180,138,< 95) they are overproduced by ∼20% and the overproduction reaches 40% for the lightest ones. Obviously, by comparing the results of runs 0 and 1 (top and third from top panels) one may obtain the s-fraction of each mixed isotope as fs,i,1=Wi,0/Wi,1=Wi,0/(Wi,0+C fr,i,0)(8) and the corresponding r-fraction as fr,i,1=1−fs,i,1=1−Wi,0/(Wi,0+C fr,i,0)(9) This procedure was adopted in Paper I, albeit not for the pure r-isotopes for which we assumed a final r-fraction equal to the initial one fr=1. However, at this level the method was obviously not self-consistent: the resulting rresiduals, obtained with Eq. 9were not the same as those used to run the model with the r-component in Eq. 4. This is obvious in the 4th panel of Fig. 1, in particular regarding the r-fractions of 76Ge (which is now ∼80% instead of 100% initially) and 82Se (now ∼90% instead of 100%). Step 3: In this study, seeking for self-consistency, we proceed by injecting the obtained r-fractions of step 1 and Eq. 9into the yields of Eq. 4and running a new model. The results of the new model are identical with those of previous calculations regarding all isotopes below Z=30 and the pure s-ones, but they fit slightly better the pre-solar distribution of mixed s+r isotopes. 8The radioactive decay of Th and U isotopes is properly taken into account in our GCE model. MNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 9 Table 2. Limits of recursive functions W,fsand frfor k→ ∞ Function C>W0C⩽W0 Wk=W0+C fr,k−1C W0 fs,k=W0 W0+C fr,k−1 W0 C1 fr,k= 1 - W0 W0+C fr,k−11 - W0 C0 We evaluate the quality of the fit to the solar composition through a simple χ2test and we repeat running the models injecting each time the new r-fraction obtained through Eq. 9into the yields of mixed isotopes. The fit improves slower and slower as the number of iterations increases, until the improvement becomes negligible (less than 1 part in a thousand) and we stop. This happens in general after 10-20 iterations, depending on the initial r-fractions adopted9. From the mathematical point of view, it can be easily shown that the quantities W(Eq. 5), fs(Eq. 8), and fr(Eq. 9), expressed as recursive functions, converge to the values indicated in Table 2, depending on whether the constant Cis greater or smaller than the initial overabundance W010, i.e. the scomponent. In other terms, our results for the sand rfractions depend uniquely on a) the adopted stellar yields of s-isotopes (which determine, along with the chemical evolution model, the term W0), and b) the goodness of the fit to the pure risotopes of Th and U (which determine through Eq. 4and 7the constant C), but they are independent of the choice of the initial values of fr,0. The reason why this iterative method improves - albeit slightly - the overall fit is due to the fact that the sum of the sand rfractions for a mixed nucleus is always fs+fr=1. If the new s-fraction is found (Eq. 8) to be smaller than the original one, then the new r-fraction is automatically found to be larger than the original one to compensate, and vice versa. In the bottom panel of Fig. 1 we display the results of the final run. The agreement with pre-solar abundances is now considerably improved for the mixed (s+r) nuclei, which are reproduced to better than a few % in most cases. We consider this a satisfactory result and we believe that it is the best one may hope to get from current models of stellar nucleosynthesis and galactic chemical evolution. We also repeated the procedure by adopting the initial r-residuals of Goriely (1999) and we obtained quantitatively similar results for all mixed (s+r) isotopes, except for the few cases which are classified as pure sor rby Goriely (1999) but not by Sneden et al. (2008); these are cases where the minor residual has a very small contribution to the isotopic abundance, typically less than a few %, which may be smaller than the uncertainties defined by the method of Goriely (1999). 9The number of iterations required to reach a given level of convergence increases with decreasing fr,0; for a level of 10−2we find that 20-30 iterations are sufficient. 10 This can be trivially obtained by putting fr=fr,0in Eq. 9 4 RESULTS AND DISCUSSION Our results concerning the sand rfractions of all the heavy isotopes are presented in Table 3, along with those of Goriely (1999) and Sneden et al. (2008) as well as those of Bisterzo et al. (2014); notice that for the latter we provide only the s-contribution (see below). For an easier comparison with those studies, the data are also presented in Figs. 2 and 3. Although it is impossible (and rather meaningless) to perform a one-to-one comparison for each isotope, we notice some important features. 4.1 The sand rfractions We start by displaying in Fig. 2the results for heavy nuclei that have been classified as s-only (top) or r-only (bottom) in each of the studies of Goriely (1999), Sneden et al. (2008) and the present one. We emphasize that in our study the nuclei considered as s-only in the beginning (Model 0) are also found to be s-only during the whole procedure and in the final model, since the adopted yields for those species are always the same (exactly as in the case of nuclei lighter than Z=31). This does not mean that their final abundances match perfectly well the corresponding Solar system abundances. But we consider that the obtained deviations from the solar abundances are a natural feature of the adopted GCE method, reflecting the current limitations of 1-zone models of GCE (coming mainly from stellar yields). Our method (dividing M0 by M1) allows us to attribute fs=1 to those nuclei that we pre-defined as sonly. This is not the case with the other studies using GCE models (Travaglio et al. 2004;Bisterzo et al. 2014), which try to reproduce perfectly the solar abundances of s-only isotopes, something we think is illusory at present. This is why we postpone the discussion of our differences with those studies to Fig. 3. We notice that the sand r-fractions displayed in Fig. 2are evaluated in different ways for the three studies. Sneden et al. (2008) provide the absolute numbers for both components sand r- (Nsand Nr) for each heavy isotope, in a scale where NSi ≡106; in that case one has obviously: fs=Ns/(Ns+Nr) and fr=Nr/(Ns+Nr). Goriely (1999) provides only the Nrvalues, again in a scale NSi ≡106, but he does not provide the corresponding Nsvalues; we obtain here the corresponding Nsvalues by subtracting Nrfrom the total isotopic abundances NTwhere we use the pre-solar ones of Lodders et al. (2009), which were not available in 1999. This obviously introduces some systematic differences with the actual values found by Goriely (1999), hopefully small ones. We do not take into account a couple of nuclei considered as s-only in Goriely (1999), which we consider instead as p-nuclei, like 152Gd and 164Er. Fig. 2illustrates the difficulties to determine unambiguously whether an isotope is produced exclusively by one or the other of the two neutron capture processes. While Goriely (1999) finds 36 s-only isotopes, we and Sneden et al. (2008) find only 30. For two of the six discrepant cases, 94Zr and 208Pb, we and Sneden et al. (2008) find quite high sfractions of more than 90%, i.e. almost pure s-nuclei. For two others (72Ge and 78Se) we both find a dominant scontribution of 55-65 %, which leaves room however, for a large r-contribution. Finally, there are two cases (86Kr MNRAS 000,1–22 (2019) 16 Prantzos et al. Figure 8. Contribution of the s-, rand p-process to the isotopic and elemental pre-solar composition according to this work (from Tables 3and 4). The contribution of each process is proportional to the colored area of the corresponding box. (Goriely 1999;Sneden et al. 2008;Bisterzo et al. 2014) in Sec. 3.1. We find good overall agreement between the various studies, but also some discrepancies. The most important are found in the region of the ”weak s-process”, namely below A=90. In that region, Sneden et al. (2008) find that 76Ge and 82Se are r-only nuclei (in agreement with Goriely 1999), while we find a substantial contribution from the s-process in rotating massive stars (25% and 11%, respectively). Our σNs,Acurve (first time derived from a GCE model) displays the classical feature of σNs,A∼const between magic neutron numbers, but shows an interesting difference with the classical study of Sneden et al. (2008) (Sec. 4.2): nuclei lying near branching points, like 148Nd, 170Er and 192Os, receive a fairly small s-contribution in Sneden et al. (2008) but a considerably larger one (factors 2-4) in our case; our results are, in general, in better agreement with those of Goriely (1999) and Bisterzo et al. (2014), probably because their studies explore a larger range of (and/or more realistic) physical conditions in stellar interiors than the classical study of Sneden et al. (2008). Comparison of our model distribution with measured solar abundances (§4.3) shows an excellent agreement, especially when uncertainties in the measured abundances are considered. The most important deviations concern the sonly nuclei, because there is no possibility to modify the result by an adjustable r-component, as we do for all other mixed (s+r or pure r-) nuclei. Since the final abundances are ”driven” by the s-component of Model 0, it is clear that the most important deviations are expected in the regions where the s-process dominates, namely the three s-peaks (see Fig. 3). This is encouraging, since it implies that a better treatment of the s-process in both LIM stars and rotating massive stars will allow one to reduce further those deviations from the solar composition by applying our method. We notice here that the study of Goriely (1999) finds that the largest uncertainties in the r-component (displayed in Fig. 6, top panel) concern precisely those regions. In §4.5 we present our results for the s-, rand pcomponents of heavy elements in the Sun, both in graphical (Fig. 7and 8) and tabular (Table 4) forms. In summary, we propose a new method for evaluating the sand r-components of the chemical composition of the Sun and during the Milky Way history, in a way fully consistent with our current understanding of stellar nucleosynthesis and galactic chemical evolution. The accuracy of our results obviously depends on the current uncertainties in those fields, as well as on the uncertainties in the measured solar isotopic composition. ACKNOWLEDGEMENTS This article is based upon work partially supported from the “ChETEC” COST Action (CA16117) of COST (EuMNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 17 ropean Cooperation in Science and Technology). C.A. acknowledges in part to the Spanish grants AYA2015-63588-P and PGC2018-095317-B-C21 within the European Founds for Regional Development (FEDER). MNRAS 000,1–22 (2019) 18 Prantzos et al. Table 3: Contribution (by mass fraction) to the Solar system isotopic composition by the sand r-process, as obtained in Sneden et al. (2008), Goriely (1999), Bisterzo et al. (2014) (only for the s-process) and this work. Solar system abundances N⊙from Lodders et al. (2009) are given per 106Si atoms. Isotope Sne+2008 Gor1999 Bis2014 This Work Z A N⊙srsrssr31 69Ga 2.20E+01 0.490 0.510 0.719 0.281 - 0.653 0.347 31 71Ga 1.46E+01 0.689 0.311 0.866 0.134 - 0.832 0.168 32 70Ge 2.43E+01 1.000 0.000 1.000 0.000 - 1.000 0.000 32 72Ge 3.17E+01 0.567 0.433 1.000 0.000 - 0.678 0.322 32 73Ge 8.80E+00 0.384 0.616 0.283 0.717 - 0.548 0.452 32 74Ge 4.12E+01 0.366 0.634 0.522 0.478 - 0.499 0.501 32 76Ge 8.50E+00 0.000 1.000 0.000 1.000 - 0.247 0.753 33 75As 6.10E+00 0.215 0.785 0.380 0.620 - 0.581 0.419 34 76Se 6.32E+00 1.000 0.000 1.000 0.000 - 1.000 0.000 34 77Se 5.15E+00 0.356 0.644 0.270 0.730 - 0.605 0.395 34 78Se 1.60E+01 0.507 0.493 1.000 0.000 - 0.547 0.453 34 80Se 3.35E+01 0.235 0.765 0.161 0.839 - 0.187 0.813 34 82Se 5.89E+00 0.000 1.000 0.000 1.000 - 0.111 0.889 35 79Br 5.43E+00 0.100 0.900 0.114 0.886 - 0.278 0.722 35 81Br 5.28E+00 0.094 0.906 0.229 0.771 - 0.252 0.748 36 80Kr 1.30E+00 1.000 0.000 1.000 0.000 0.133 1.000 0.000 36 82Kr 6.51E+00 1.000 0.000 1.000 0.000 0.274 1.000 0.000 36 83Kr 6.45E+00 0.347 0.653 0.321 0.679 0.095 0.291 0.709 36 84Kr 3.18E+01 0.370 0.630 0.257 0.743 0.117 0.222 0.778 36 86Kr 9.61E+00 0.911 0.089 1.000 0.000 0.152 0.677 0.323 37 85Rb 5.12E+00 0.198 0.802 0.440 0.560 0.153 0.326 0.674 37 87Rb 2.11E+00 0.957 0.043 0.861 0.139 0.249 0.960 0.040 38 86Sr 2.30E+00 1.000 0.000 1.000 0.000 0.611 1.000 0.000 38 87Sr 1.60E+00 1.000 0.000 1.000 0.000 0.582 1.000 0.000 38 88Sr 1.92E+01 0.869 0.131 0.787 0.213 0.711 0.903 0.097 39 89Y 4.63E+00 0.719 0.281 0.760 0.240 0.719 0.778 0.222 40 90Zr 5.55E+00 0.821 0.179 0.529 0.471 0.603 0.807 0.193 40 91Zr 1.21E+00 0.967 0.033 0.826 0.174 0.712 0.957 0.043 40 92Zr 1.85E+00 0.705 0.295 0.966 0.034 0.682 0.745 0.255 40 94Zr 1.87E+00 0.908 0.092 1.000 0.000 0.836 0.925 0.075 40 96Zr 3.02E-01 0.000 1.000 1.000 0.000 0.387 0.291 0.709 41 93Nb 7.80E-01 0.676 0.324 0.873 0.127 0.560 0.651 0.349 42 95Mo 4.04E-01 0.470 0.530 0.653 0.347 0.454 0.479 0.521 42 96Mo 4.25E-01 1.000 0.000 1.000 0.000 0.782 1.000 0.000 42 97Mo 2.45E-01 0.642 0.358 0.670 0.330 0.433 0.563 0.437 42 98Mo 6.22E-01 0.847 0.153 0.881 0.119 0.575 0.804 0.196 42 100Mo 2.50E-01 0.000 1.000 0.096 0.904 0.023 0.038 0.962 44 99Ru 2.27E-01 0.300 0.700 0.238 0.762 0.210 0.312 0.688 44 100Ru 2.24E-01 1.000 0.000 1.000 0.000 0.801 1.000 0.000 44 101Ru 3.04E-01 0.158 0.842 0.122 0.878 0.128 0.141 0.859 44 102Ru 5.62E-01 0.444 0.556 0.440 0.560 0.433 0.476 0.524 44 104Ru 3.32E-01 0.000 1.000 0.000 1.000 0.014 0.023 0.977 45 103Rh 3.70E-01 0.160 0.840 0.197 0.803 0.118 0.122 0.878 46 104Pd 1.51E-01 1.000 0.000 1.000 0.000 0.839 1.000 0.000 46 105Pd 3.03E-01 0.129 0.871 0.123 0.877 0.107 0.129 0.871 46 106Pd 3.71E-01 0.491 0.509 0.539 0.461 0.398 0.510 0.490 46 108Pd 3.59E-01 0.609 0.391 0.669 0.331 0.505 0.625 0.375 46 110Pd 1.59E-01 0.000 1.000 0.019 0.981 0.016 0.044 0.956 47 107Ag 2.54E-01 0.195 0.805 0.169 0.831 0.001 0.134 0.866 47 109Ag 2.36E-01 0.231 0.769 0.271 0.729 0.226 0.286 0.714 48 110Cd 1.97E-01 1.000 0.000 1.000 0.000 0.776 1.000 0.000 48 111Cd 2.01E-01 0.203 0.797 0.244 0.756 0.249 0.321 0.679 48 112Cd 3.80E-01 0.503 0.497 0.537 0.463 0.499 0.575 0.425 48 113Cd 1.92E-01 0.302 0.698 0.354 0.646 0.285 0.365 0.635 48 114Cd 4.52E-01 0.672 0.328 0.619 0.381 0.591 0.693 0.307 48 116Cd 1.18E-01 0.000 1.000 0.191 0.809 0.088 0.087 0.913 49 115In 1.70E-01 0.320 0.680 0.347 0.653 0.298 0.391 0.609 50 116Sn 5.24E-01 1.000 0.000 1.000 0.000 0.683 1.000 0.000 50 117Sn 2.77E-01 0.533 0.467 0.458 0.542 0.384 0.501 0.499 50 118Sn 8.73E-01 0.837 0.163 0.721 0.279 0.554 0.728 0.272 50 119Sn 3.09E-01 0.705 0.295 0.405 0.595 0.469 0.618 0.382 MNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 19 Continuation of Table 3 Isotope Sne+2008 Gor1999 Bis2014 This Work Z A N⊙srsrssr50 120Sn 1.18E+00 0.934 0.066 0.818 0.182 0.627 0.850 0.150 50 122Sn 1.67E-01 0.000 1.000 0.090 0.910 0.364 0.101 0.899 50 124Sn 2.09E-01 0.000 1.000 0.000 1.000 - 0.008 0.992 51 121Sb 1.79E-01 0.294 0.706 0.533 0.467 0.309 0.393 0.607 51 123Sb 1.34E-01 0.000 1.000 0.157 0.843 0.050 0.051 0.949 52 122Te 1.22E-01 1.000 0.000 1.000 0.000 0.736 1.000 0.000 52 123Te 4.30E-02 1.000 0.000 1.000 0.000 0.741 1.000 0.000 52 124Te 2.26E-01 1.000 0.000 1.000 0.000 0.781 1.000 0.000 52 125Te 3.35E-01 0.248 0.752 0.236 0.764 0.174 0.269 0.731 52 126Te 8.89E-01 0.462 0.538 0.447 0.553 0.363 0.474 0.526 52 128Te 1.49E+00 0.000 1.000 0.011 0.989 0.033 0.026 0.974 52 130Te 1.58E+00 0.000 1.000 0.003 0.997 - 0.004 0.996 53 127I 1.10E+00 0.055 0.945 0.229 0.771 0.038 0.032 0.968 54 128Xe 1.22E-01 1.000 0.000 1.000 0.000 0.763 1.000 0.000 54 129Xe 1.50E+00 0.051 0.949 0.280 0.720 0.028 0.030 0.970 54 130Xe 2.39E-01 1.000 0.000 1.000 0.000 0.851 1.000 0.000 54 131Xe 1.19E+00 0.084 0.916 0.309 0.691 0.065 0.055 0.945 54 132Xe 1.44E+00 0.384 0.616 0.546 0.454 0.268 0.353 0.647 54 134Xe 5.27E-01 0.000 1.000 0.269 0.731 0.041 0.028 0.972 54 136Xe 4.29E-01 0.000 1.000 0.231 0.769 - 0.001 0.999 55 133Cs 3.71E-01 0.151 0.849 0.167 0.833 0.135 0.157 0.843 56 134Ba 1.08E-01 1.000 0.000 1.000 0.000 1.011 1.000 0.000 56 135Ba 2.95E-01 0.186 0.814 0.159 0.841 0.285 0.249 0.751 56 136Ba 3.51E-01 1.000 0.000 1.000 0.000 1.013 1.000 0.000 56 137Ba 5.02E-01 0.568 0.432 0.661 0.339 0.632 0.647 0.353 56 138Ba 3.21E+00 0.940 0.060 0.933 0.067 0.918 0.959 0.041 57 139La 4.57E-01 0.754 0.246 0.656 0.344 0.755 0.800 0.200 58 140Ce 1.04E+00 0.909 0.091 0.846 0.154 0.920 0.935 0.065 58 142Ce 1.31E-01 0.000 1.000 0.496 0.504 0.195 0.126 0.874 59 141Pr 1.72E-01 0.491 0.509 0.360 0.640 0.499 0.535 0.465 60 142Nd 2.31E-01 1.000 0.000 1.000 0.000 0.983 1.000 0.000 60 143Nd 1.03E-01 0.363 0.637 0.315 0.685 0.329 0.377 0.623 60 144Nd 2.03E-01 0.528 0.472 0.508 0.492 0.522 0.565 0.435 60 145Nd 7.50E-02 0.290 0.710 0.280 0.720 0.262 0.297 0.703 60 146Nd 1.47E-01 0.632 0.368 0.637 0.363 0.660 0.699 0.301 60 148Nd 4.90E-02 0.083 0.917 0.141 0.859 0.173 0.156 0.844 60 150Nd 4.80E-02 0.000 1.000 0.000 1.000 - 0.003 0.997 62 147Sm 4.10E-02 0.088 0.912 0.185 0.815 0.265 0.234 0.766 62 148Sm 3.00E-02 1.000 0.000 1.000 0.000 1.034 1.000 0.000 62 149Sm 3.70E-02 0.139 0.861 0.127 0.873 0.129 0.121 0.879 62 150Sm 2.00E-02 1.000 0.000 1.000 0.000 1.000 1.000 0.000 62 152Sm 7.10E-02 0.254 0.746 0.196 0.804 0.228 0.268 0.732 62 154Sm 6.00E-02 0.000 1.000 0.008 0.992 0.025 0.049 0.951 63 151Eu 4.71E-02 0.000 1.000 0.040 0.960 0.059 0.047 0.953 63 153Eu 5.14E-02 0.040 0.960 0.037 0.963 0.061 0.050 0.950 64 154Gd 7.80E-03 1.000 0.000 1.000 0.000 0.891 1.000 0.000 64 155Gd 5.33E-02 0.062 0.938 0.122 0.878 0.060 0.057 0.943 64 156Gd 7.36E-02 0.214 0.786 0.213 0.787 0.179 0.206 0.794 64 157Gd 5.63E-02 0.132 0.868 0.163 0.837 0.111 0.097 0.903 64 158Gd 8.94E-02 0.318 0.682 0.313 0.687 0.275 0.308 0.692 64 160Gd 7.87E-02 0.000 1.000 0.058 0.942 0.007 0.022 0.978 65 159Tb 6.34E-02 0.063 0.938 0.052 0.948 0.080 0.072 0.928 66 160Dy 9.40E-03 1.000 0.000 1.000 0.000 0.901 1.000 0.000 66 161Dy 7.62E-02 0.051 0.949 0.028 0.972 0.053 0.046 0.954 66 162Dy 1.03E-01 0.137 0.863 0.125 0.875 0.160 0.159 0.841 66 163Dy 1.01E-01 0.021 0.979 0.033 0.967 0.044 0.046 0.954 66 164Dy 1.14E-01 0.165 0.835 0.097 0.903 0.239 0.243 0.757 67 165Ho 9.10E-02 0.067 0.933 0.078 0.922 0.083 0.074 0.926 68 166Er 8.80E-02 0.143 0.857 0.144 0.856 0.167 0.134 0.866 68 167Er 6.00E-02 0.086 0.914 0.090 0.910 0.094 0.088 0.912 68 168Er 7.10E-02 0.299 0.701 0.287 0.713 0.314 0.367 0.633 68 170Er 3.90E-02 0.026 0.974 0.054 0.946 0.126 0.126 0.874 69 169Tm 4.06E-02 0.162 0.838 0.163 0.837 0.091 0.128 0.872 70 170Yb 7.60E-03 1.000 0.000 1.000 0.000 0.958 1.000 0.000 70 171Yb 3.61E-02 0.121 0.879 0.177 0.823 0.227 0.226 0.774 MNRAS 000,1–22 (2019) 20 Prantzos et al. Continuation of Table 3 Isotope Sne+2008 Gor1999 Bis2014 This Work Z A N⊙srsrssr70 172Yb 5.56E-02 0.333 0.667 0.315 0.685 0.439 0.454 0.546 70 173Yb 4.13E-02 0.205 0.795 0.235 0.765 0.278 0.305 0.695 70 174Yb 8.21E-02 0.519 0.481 0.524 0.476 0.609 0.635 0.365 70 176Yb 3.33E-02 0.000 1.000 0.123 0.877 0.072 0.115 0.885 71 175Lu 3.70E-02 0.162 0.838 0.176 0.824 0.181 0.176 0.824 71 176Lu 1.10E-03 1.000 0.000 1.000 0.000 1.055 1.000 0.000 72 176Hf 8.10E-03 1.000 0.000 1.000 0.000 1.001 1.000 0.000 72 177Hf 3.16E-02 0.172 0.828 0.247 0.753 0.173 0.175 0.825 72 178Hf 4.25E-02 0.488 0.512 0.548 0.452 0.575 0.618 0.382 72 179Hf 2.12E-02 0.318 0.682 0.349 0.651 0.412 0.451 0.549 72 180Hf 5.47E-02 0.636 0.364 0.735 0.265 0.894 0.809 0.191 73 181Ta 2.10E-02 0.409 0.591 0.495 0.505 0.466 0.503 0.497 74 182W 3.63E-02 0.667 0.333 0.625 0.375 0.638 0.675 0.325 74 183W 1.96E-02 0.650 0.350 0.668 0.332 0.622 0.701 0.299 74 184W 4.20E-02 0.690 0.310 0.748 0.252 0.788 0.790 0.210 74 186W 3.90E-02 0.162 0.838 0.372 0.628 0.424 0.279 0.721 75 185Re 2.07E-02 0.222 0.778 0.271 0.729 0.270 0.341 0.659 75 187Re 3.74E-02 0.029 0.971 0.150 0.850 0.094 0.048 0.952 76 186Os 1.08E-02 1.000 0.000 1.000 0.000 1.035 1.000 0.000 76 187Os 8.60E-03 0.990 0.010 1.000 0.000 0.374 0.751 0.249 76 188Os 9.04E-02 0.168 0.832 0.217 0.783 0.282 0.209 0.791 76 189Os 1.10E-01 0.035 0.965 0.064 0.936 0.048 0.041 0.959 76 190Os 1.79E-01 0.111 0.889 0.151 0.849 0.146 0.132 0.868 76 192Os 2.78E-01 0.003 0.997 0.018 0.982 0.033 0.018 0.982 77 191Ir 2.50E-01 0.020 0.980 0.084 0.916 0.019 0.015 0.985 77 193Ir 4.21E-01 0.007 0.993 0.078 0.922 0.014 0.009 0.991 78 192Pt 1.00E-02 1.000 0.000 1.000 0.000 0.812 1.000 0.000 78 194Pt 4.20E-01 0.044 0.956 0.000 1.000 0.049 0.072 0.928 78 195Pt 4.31E-01 0.013 0.987 0.000 1.000 0.020 0.032 0.968 78 196Pt 3.22E-01 0.101 0.899 0.062 0.938 0.123 0.133 0.867 78 198Pt 9.10E-02 0.000 1.000 0.000 1.000 0.001 0.027 0.973 79 197Au 1.95E-01 0.054 0.946 0.021 0.979 0.061 0.058 0.942 80 198Hg 4.60E-02 1.000 0.000 1.000 0.000 0.828 1.000 0.000 80 199Hg 7.70E-02 0.271 0.729 0.342 0.658 0.216 0.255 0.745 80 200Hg 1.06E-01 0.630 0.370 0.685 0.315 0.519 0.575 0.425 80 201Hg 6.00E-02 0.426 0.574 0.558 0.442 0.399 0.428 0.572 80 202Hg 1.37E-01 0.752 0.248 0.812 0.188 0.704 0.742 0.258 80 204Hg 3.10E-02 0.000 1.000 0.142 0.858 0.082 0.038 0.962 81 203Tl 5.40E-02 0.778 0.222 0.939 0.061 0.807 0.823 0.177 81 205Tl 1.29E-01 0.594 0.406 0.615 0.385 0.667 0.735 0.265 82 204Pb 6.60E-02 1.000 0.000 1.000 0.000 0.870 1.000 0.000 82 206Pb 6.14E-01 0.594 0.406 0.679 0.321 0.729 0.688 0.312 82 207Pb 6.80E-01 0.528 0.472 0.791 0.209 0.702 0.706 0.294 82 208Pb 1.95E+00 0.931 0.069 1.000 0.000 0.977 0.922 0.078 83 209Bi 1.38E-01 0.354 0.646 0.637 0.363 0.204 0.216 0.784 90 232Th 4.40E-02 0.000 1.000 0.000 1.000 - 0.000 1.000 92 235U 5.80E-03 0.000 1.000 0.000 1.000 - 0.000 1.000 92 238U 1.80E-02 0.000 1.000 0.000 1.000 - 0.000 1.000 End of Table MNRAS 000,1–22 (2019) A new assessment of solar sand rcomponents 21 REFERENCES Arlandini C., K¨ appeler F., Wisshak K., Gallino R., Lugaro M., Busso M., Straniero O., 1999, ApJ,525, 886 Arnould M., Goriely S., 2003, Phys. Rep.,384, 1 Arnould M., Goriely S., Takahashi K., 2007, Phys. Rep.,450, 97 Battino U., et al., 2016, ApJ,827, 30 Bisterzo S., Gallino R., Straniero O., Cristallo S., K¨ appeler F., 2010, MNRAS,404, 1529 Bisterzo S., Travaglio C., Gallino R., Wiescher M., K¨ appeler F., 2014, ApJ,787, 10 Bisterzo S., Travaglio C., Wiescher M., K¨ appeler F., Gallino R., 2017, ApJ,835, 97 Burbidge E. M., Burbidge G. R., Fowler W. A., Hoyle F., 1957, Reviews of Modern Physics,29, 547 Busso M., Gallino R., 1985, A&A, 151, 205 Busso M., Gallino R., Wasserburg G. J., 1999, ARA&A,37, 239 Cescutti G., Hirschi R., Nishimura N., Hartogh J. W. d., Rauscher T., Murphy A. S. J., Cristallo S., 2018, MNRAS,478, 4101 Choplin A., Maeder A., Meynet G., Chiappini C., 2016, A&A, 593, A36 Choplin A., Hirschi R., Meynet G., Ekstr¨ om S., 2017, A&A, 607, L3 Choplin A., Hirschi R., Meynet G., Ekstr¨ om S., Chiappini C., Laird A., 2018, A&A,618, A133 Clayton D. D., 1968, Principles of stellar evolution and nucleosynthesis Clayton D. D., Rassbach M. E., 1967, ApJ,148, 69 Cˆot´e B., et al., 2018, ApJ,855, 99 Cˆot´e B., et al., 2019, ApJ,875, 106 Couch R. G., Schmiedekamp A. B., Arnett W. D., 1974, ApJ, 190, 95 Cowan J. J., Rose W. K., 1977, ApJ,212, 149 Cowan J. J., Sneden C., Lawler J. E., Aprahamian A., Wiescher M., Langanke K., Mart´ınez-Pinedo G., Thielemann F.-K., 2019, arXiv e-prints, p. arXiv:1901.01410 Cristallo S., Straniero O., Gallino R., Piersanti L., Dom´ınguez I., Lederer M. T., 2009, ApJ,696, 797 Cristallo S., et al., 2011, ApJS,197, 17 Cristallo S., Straniero O., Piersanti L., Gobrecht D., 2015a, ApJS, 219, 40 Cristallo S., Abia C., Straniero O., Piersanti L., 2015b, ApJ, 801, 53 Cristallo S., Karinkuzhi D., Goswami A., Piersanti L., Gobrecht D., 2016, ApJ,833, 181 Dardelet L., et al., 2014, in XIII Nuclei in the Cosmos (NIC XIII). p. 145 Denissenkov P. A., Tout C. A., 2003, MNRAS,340, 722 Denissenkov P. A., Herwig F., Battino U., Ritter C., Pignatari M., Jones S., Paxton B., 2017, ApJ,834, L10 Dillmann I., Plag R., Heil M., K¨ appeler F., Rauscher T., 2008, arXiv e-prints, Eichler M., et al., 2015, ApJ,808, 30 Fern´andez R., Metzger B. D., 2016, Annual Review of Nuclear and Particle Science,66, 23 Frischknecht U., et al., 2016, MNRAS,456, 1803 Fr¨ ohlich C., Mart´ınez-Pinedo G., Liebend¨ orfer M., Thielemann F.-K., Bravo E., Hix W. R., Langanke K., Zinner N. T., 2006, Physical Review Letters,96, 142502 Fujimoto S.-i., Hashimoto M.-a., Kotake K., Yamada S., 2007, ApJ,656, 382 Gallino R., Arlandini C., Busso M., Lugaro M., Travaglio C., Straniero O., Chieffi A., Limongi M., 1998, ApJ,497, 388 Goriely S., 1999, A&A, 342, 881 Goswami A., Prantzos N., 2000, A&A, 359, 191 Guerrero C., et al., 2013, European Physical Journal A,49, 27 Guiglion G., de Laverny P., Recio-Blanco A., Prantzos N., 2018, A&A,619, A143 Table 4. Solar elementary abundances from Lodders et al. (2009) and fractions produced by the s-, rand pprocesses according to this study. Elm. Z log ǫ(X)1Mass fraction srpGa 31 3.150 7.00E-08 0.730 0.270 0.000 Ge 32 3.645 2.28E-07 0.636 0.364 0.000 As 33 2.372 1.25E-08 0.581 0.419 0.000 Se 34 3.416 1.46E-07 0.379 0.612 0.008 Br 35 2.616 2.35E-08 0.265 0.735 0.000 Kr 36 3.334 1.28E-07 0.387 0.610 0.003 Rb 37 2.446 1.69E-08 0.510 0.490 0.000 Sr 38 2.953 5.58E-08 0.912 0.083 0.005 Y 39 2.252 1.13E-08 0.778 0.222 0.000 Zr 40 2.619 2.70E-08 0.817 0.183 0.000 Nb 41 1.479 1.99E-09 0.651 0.349 0.000 Mo 42 1.993 6.71E-09 0.497 0.275 0.228 Ru 44 1.837 4.94E-09 0.338 0.591 0.071 Rh 45 1.155 1.04E-09 0.122 0.878 0.000 Pd 46 1.719 3.96E-09 0.448 0.542 0.010 Ag 47 1.277 1.45E-09 0.209 0.791 0.000 Cd 48 1.784 4.85E-09 0.548 0.432 0.021 In 49 0.837 5.60E-10 0.374 0.582 0.044 Sn 50 2.144 1.17E-08 0.680 0.301 0.019 Sb 51 1.082 1.05E-09 0.247 0.753 0.000 Te 52 2.258 1.64E-08 0.192 0.807 0.001 I 53 1.628 3.83E-09 0.032 0.968 0.000 Xe 54 2.324 1.96E-08 0.182 0.816 0.002 Cs 55 1.156 1.35E-09 0.157 0.843 0.000 Ba 56 2.237 1.68E-08 0.888 0.109 0.002 La 57 1.247 1.74E-09 0.799 0.200 0.001 Ce 58 1.658 4.53E-09 0.848 0.148 0.004 Pr 59 0.822 6.64E-10 0.535 0.465 0.000 Nd 60 1.519 3.38E-09 0.615 0.385 0.000 Sm 62 1.013 1.10E-09 0.325 0.647 0.029 Eu 63 0.580 4.10E-10 0.049 0.951 0.000 Gd 64 1.143 1.55E-09 0.163 0.835 0.002 Tb 65 0.389 2.76E-10 0.072 0.928 0.000 Dy 66 1.193 1.80E-09 0.151 0.847 0.001 Ho 67 0.546 4.11E-10 0.074 0.926 0.000 Er 68 1.006 1.20E-09 0.184 0.799 0.017 Tm 69 0.195 1.88E-10 0.128 0.872 0.000 Yb 70 0.995 1.22E-09 0.429 0.570 0.001 Lu 71 0.168 1.83E-10 0.204 0.796 0.000 Hf 72 0.786 7.74E-10 0.605 0.393 0.002 Ta 73 -0.091 1.04E-10 0.503 0.497 0.000 W 74 0.724 6.91E-10 0.601 0.397 0.001 Re 75 0.351 2.97E-10 0.154 0.846 0.000 Os 76 1.417 3.53E-09 0.103 0.897 0.000 Ir 77 1.413 3.53E-09 0.011 0.989 0.000 Pt 78 1.692 6.81E-09 0.078 0.922 0.000 Au 79 0.877 1.05E-09 0.058 0.942 0.000 Hg 80 1.248 2.52E-09 0.548 0.450 0.002 Tl 81 0.849 1.03E-09 0.760 0.240 0.000 Pb 82 2.106 1.88E-08 0.831 0.169 0.000 Bi 83 0.727 7.91E-10 0.216 0.784 0.000 Th 90 0.230 2.80E-10 0.000 1.000 0.000 U 92 -0.037 1.55E-10 0.000 1.000 0.000 Note:(1) log ǫ(X)≡log (X/H) +12. MNRAS 000,1–22 (2019) 22 Prantzos et al. Hampel M., Stancliffe R. J., Lugaro M., Meyer B. S., 2016, ApJ, 831, 171 Haynes C. J., Kobayashi C., 2019, MNRAS,483, 5123 Herwig F., Bloecker T., Schoenberner D., El Eid M., 1997, A&A, 324, L81 Hotokezaka K., Beniamini P., Piran T., 2018, International Journal of Modern Physics D,27, 1842005 Ishimaru Y., Wanajo S., Prantzos N., 2015, ApJ,804, L35 Jones S., Ritter C., Herwig F., Fryer C., Pignatari M., Bertolli M. G., Paxton B., 2016, MNRAS,455, 3848 Kappeler F., Beer H., Wisshak K., 1989, Reports on Progress in Physics,52, 945 K¨ appeler F., Gallino R., Bisterzo S., Aoki W., 2011, Reviews of Modern Physics,83, 157 Kubryk M., Prantzos N., Athanassoula E., 2015, A&A,580, A126 Lamb S. A., Howard W. M., Truran J. W., Iben I. J., 1977, ApJ, 217, 213 Limongi M., Chieffi A., 2018, ApJS,237, 13 Lodders K., 2003, ApJ,591, 1220 Lodders K., Palme H., Gail H.-P., 2009, Landolt B¨ ornstein, Nishimura N., Hirschi R., Rauscher T., St. J. Murphy A., Cescutti G., 2017, MNRAS,469, 1752 Nomoto K., Kobayashi C., Tominaga N., 2013, ARA&A,51, 457 Ojima T., Ishimaru Y., Wanajo S., Prantzos N., Fran¸cois P., 2018, ApJ,865, 87 Perego A., Radice D., Bernuzzi S., 2017, ApJ,850, L37 Peters J. G., 1968, ApJ,154, 225 Pian E., et al., 2017, Nature,551, 67 Pignatari M., Gallino R., Meynet G., Hirschi R., Herwig F., Wiescher M., 2008, ApJ,687, L95 Prantzos N., Arnould M., Arcoragi J.-P., 1987, ApJ,315, 209 Prantzos N., Hashimoto M., Nomoto K., 1990, A&A, 234, 211 Prantzos N., Abia C., Limongi M., Chieffi A., Cristallo S., 2018, MNRAS,476, 3432 Raiteri C. M., Gallino R., Busso M., Neuberger D., Kaeppeler F., 1993, ApJ,419, 207 Rosswog S., 2015, International Journal of Modern Physics D, 24, 1530012 Sanders R. H., 1967, ApJ,150, 971 Schatz H., et al., 1998, Phys. Rep.,294 Schwarzschild M., H¨ arm R., 1967, ApJ,150, 961 Serminato A., Gallino R., Travaglio C., Bisterzo S., Straniero O., 2009, Publ. Astron. Soc. Australia,26, 153 Siegel D. M., Barnes J., Metzger B. D., 2019, Nature,569, 241 Sneden C., Cowan J. J., Lawler J. E., Burles S., Beers T. C., Fuller G. M., 2002, ApJ,566, L25 Sneden C., Cowan J. J., Gallino R., 2008, ARA&A,46, 241 Straniero O., Gallino R., Busso M., Chiefei A., Raiteri C. M., Limongi M., Salaris M., 1995, ApJ,440, L85 Straniero O., Gallino R., Cristallo S., 2006, Nuclear Physics A, 777, 311 Suess H. E., Urey H. C., 1956, Reviews of Modern Physics,28, 53 Takahashi K., Yokoi K., 1987, Atomic Data and Nuclear Data Tables,36, 375 Thielemann F. K., Eichler M., Panov I. V., Wehmeyer B., 2017, Annual Review of Nuclear and Particle Science,67, 253 Timmes F. X., Woosley S. E., Weaver T. A., 1995, ApJS,98, 617 Travaglio C., Gallino R., Arnone E., Cowan J., Jordan F., Sneden C., 2004, ApJ,601, 864 Travaglio C., Gallino R., Rauscher T., R¨ opke F. K., Hillebrandt W., 2015, ApJ,799, 54 Travaglio C., Rauscher T., Heger A., Pignatari M., West C., 2018, ApJ,854, 18 Trippella O., Busso M., Palmerini S., Maiorca E., Nucci M. C., 2016, ApJ,818, 125 Tsujimoto T., Shigeyama T., 2014, A&A,565, L5 Wanajo S., Janka H.-T., M¨ uller B., 2011, ApJ,726, L15 Watson D., et al., 2019, Nature,574, 497 Wehmeyer B., Fr¨ ohlich C., Cˆot´e B., Pignatari M., Thielemann F. K., 2019, MNRAS,487, 1745 Weigert A., 1966, Zeitschrift fur Astrophysik, 64, 395 Wisshak K., Voss F., K¨ appeler F., Kazakov L., 2006a, Phys. Rev. C,73, 015807 Wisshak K., Voss F., K¨ appeler F., Kazakov L., Beˇcv´aˇr F., Krtiˇcka M., Gallino R., Pignatari M., 2006b, Phys. Rev. C,73, 045807 Woosley S. E., Weaver T. A., 1995, ApJS,101, 181 Wu M.-R., Fern´andez R., Mart´ınez-Pinedo G., Metzger B. D., 2016, MNRAS,463, 2323 Yokoi K., Takahashi K., Arnould M., 1983, A&A, 117, 65 This paper has been typeset from a T EX/L A T EX file prepared by the author. MNRAS 000,1–22 (2019)