betaFeedback

WorksPhysics and Astronomy

Fission fragment distributions and their impact on the r-process nucleosynthesis in neutron star mergers

Abstract

Neutron star (NS) merger ejecta offer a viable site for the production of heavy r-process elements with nuclear mass numbers A >140. The crucial role of fission recycling is responsible for the robustness of this site against many astrophysical uncertainties. Here, we introduce new improvements to our scission-point model, called SPY, to derive the fission fragment distribution for all neutron-rich fissioning nuclei of relevance in r-process calculations. These improvements include a phenomenological modification of the scission distance and a smoothing procedure of the distribution. Such corrections lead to a much better agreement with experimental fission yields. Those yields are also used to estimate the number of neutrons emitted by the excited fragments on the basis of different neutron evaporation models. Our new fission yields are extensively compared to those predicted by the so-called GEF model. The impact of fission on the r-process nucleosynthesis in binary neutron mergers is also reanalyzed. Two scenarios are considered, the first one with low initial electron fraction is subject to intense fission recycling, in contrast to the second one which includes weak interactions on nucleons. The various regions of the nuclear chart responsible for fission recycling during the neutron irradiation as well as after freeze-out are discussed. The contribution fission processes may have to the final abundance distribution is also studied in detail in the light of newly defined quantitative indicators describing the fission recycling, the fission seeds and the fission progenitors. In particular, those allow us to estimate the contribution of fission to the final abundance distribution stemming from specific heavy nuclei. Calculations obtained with SPY and GEF fission fragment distributions are compared for both r-process scenarios.

Introduction

The r-process, or the rapid neutron-capture process, of stellar nucleosynthesis is invoked to explain the production of the stable (and some long-lived radioactive) neutron-rich nuclides heavier than iron that are observed in stars of various metallicities, as well as in the solar system (for a review, see Ref. [1–3]). In recent years, nuclear astrophysicists have developed more and more sophisticated r-process models, trying to explain the solar system composition in a satisfactory way by adding new astrophysical or nuclear physics ingredients. The r-process remains the most complex nucleosynthetic process to model from the astrophysics as well as nuclear-physics points of view. Progress in the modeling of type-II supernovae and γ\gamma-ray bursts has raised a lot of excitement about the so-called neutrino-driven wind environment. However, until now a successful r-process cannot be obtained ab initio without tuning the relevant parameters (neutron excess, entropy, expansion timescale) in a way that is not supported by the most sophisticated existing models [4, 5]. Although these scenarios remain promising, especially in view of their potential to contribute to the galactic enrichment significantly, they remain affected by large uncertainties associated mainly with the still incompletely understood mechanism responsible for the supernova explosion and the persistent difficulties to obtain suitable r-process conditions in self-consistent dynamical explosion and neutron-star (NS) cooling models [5, 6, 7, 8]. In particular, a subclass of core-collapse supernovae, the so-called collapsars corresponding to the fate of rapidly rotating and highly magnetized massive stars and generally considered to be at the origin of observed long gamma-ray bursts, could be a promising r-process site [9]. The production of r-nuclides in these events may be associated with jets predicted to accompany the explosion, or with the accretion disk forming around a newly born central BH [10].

Since early 2000, special attention has been paid to NS mergers as r-process sites following the confirmation by hydrodynamic simulations that a non-negligible amount of matter could be ejected from the system. Newtonian [11, 12, 13, 14, 15], conformally flat general relativistic [16, 17, 18], as well as fully relativistic [19, 20, 21, 22, 23] hydrodynamical simulations of NS-NS and NS-black hole (BH) mergers with microphysical equations of state have demonstrated that typically some 10−3M⊙10^{-3} M_\odot up to more than 0.1M⊙0.1 M_\odot can become gravitationally unbound on roughly dynamical timescales due to shock acceleration and tidal stripping. Also the relic object (a hot, transiently stable hypermassive NS [24] followed by a stable supermassive NS, or a BH-torus system), can lose mass through outflows driven by a variety of mechanisms [18, 25, 26, 27, 28, 29, 30].

Simulations of growing sophistication have confirmed that the ejecta from NS mergers are viable strong r-process sites up to the third abundance peak and the actinides [15, 17, 18, 22, 31]. The r-nuclide enrichment is predicted to originate from both the dynamical (prompt) material expelled during the NS-NS or NS-BH merger phase and from the outflows generated during the post-merger remnant evolution of the relic BH-torus system. The resulting abundance distributions are found to reproduce very well the solar system distribution, as well as various elemental distributions observed in low-metallicity stars [3]. During the dynamical phase of the merging scenario, the number of free neutrons per seed nucleus can reach so high values (typically few hundreds) that heavy fissioning nuclei are produced. Fissioning results in a robust reproduction of the solar-system-like abundance pattern of the rare-earth elements, as observed in metal-poor stars [1, 3, 17, 32]. This supports the possible production of these elements by fission recycling in NS merger ejecta. In addition, the ejected mass of r-process material, combined with the predicted astrophysical event rate (around 10 Myr−110\ \mathrm{Myr}^{-1} in the Milky Way [33]) can account for the majority of r-material in our Galaxy, e.g. [17, 21, 34, 35]. A further piece of evidence that NS mergers are r-nuclide producers indeed comes from the very important 2017 gravitational-wave and electromagnetic observation of the kilonova GW170817 [36, 37, 38, 39, 40, 41, 42].

In this specific NS merger scenario, the neutron richness was found to be so high that heavy fissioning nuclei can be produced. For this reason, in this astrophysical site, fission plays a fundamental role, more particularly by ii) recycling the matter during the neutron irradiation (or if not, by allowing the possible production of super-heavy long-lived nuclei, if any), iiii) shaping the r-abundance distribution in the 110≤A≤170110 \le A \le170 mass region at the end of the neutron irradiation, iiiiii) defining the residual production of some specific heavy stable nuclei, more specifically Pb and Bi, but also the long-lived cosmochronometers Th and U, and iviv) heating the environment through the energy released. More details can be found for instance in Refs. [14, 32, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52].

Despite the recent success of nucleosynthesis studies for NS mergers, the details of r-processing in these events is still affected by a variety of uncertainties, both from the nuclear physics and astrophysics point of view. For this reason, we present here a new effort to further improve the description of fission fragment distributions (FFD) obtained on the basis of the so-called Scission Point Yield (SPY) model [53]. Sec. II is devoted to the description of such improvements. A phenomenological correction to the fission distance and a new smoothing procedure for deriving the fission yields are presented in Sec. II.1. Since the FFD of relevance for the r-process application correspond to the post-neutron-emission one deduced from pre-neutron FFDs, our procedure for estimating the neutron evaporation is detailed in Sec. II.2. As r-process is impacted by neutron captures, sensitivity studies are performed in Sec. II.3 to demonstrate that the neutron emission depends mainly on the primary fission yields. SPY pre-neutron FFDs are compared with experimental data and the well-known GEF (“General description of Fission observables”) calculations [54] in Sec. II.4.

In addition, it has been shown that weak interactions may strongly affect the composition of the dynamical ejecta and thus the efficiency of the r-process [22, 55, 56, 57, 58, 59, 60, 61, 62, 63]. In this case, the initial neutron richness of the ejecta can be significantly reduced and consequently the role of fission, albeit in basically all calculations a significant fraction of neutron-rich ejecta is present. In Sec. III, we re-evaluate the role of fission in NS merger in two distinct scenarios, namely the traditional one neglecting weak interactions on nucleons and an updated version of such a simulation where neutrino interactions are included phenomenologically [56, 62] to reproduce qualitatively the results found in Ref. [63]. In Sec.III.1, we will present the final abundance distributions obtained within both the scenarios as well as the role of the fission recycling and link them to fundamental initial properties of ejected material, namely the initial neutron mass fraction, electron fraction and entropy. The relevance of the fission recycling and its contribution to the final r-abundances is thoroughly detailed in Secs. III.2–III.4 for both scenarios. Finally, conclusions are drawn in Sec. IV.

Fission fragments distributions

SPY model

The SPY model is a static and statistical scission point model [53, 64] where a thermodynamic equilibrium at scission is assumed, hence the evolution between the saddle and the scission points is neglected. The model is based on two pillars, namely the absolute available energy balance at the scission configurations and the statistical description of the available phase space. The available energy balance is performed for all energetically possible fragmentations of a fissioning system at scission as a function of the deformation of both fragments. The available energy is defined as the difference between the potential energy of the fissioning system at scission and the energy of the excited compound nucleus where both nascent fragments are supposed to be at rest. The potential energy of the fissioning system at scission is obtained as the sum of the individual binding energies of the two fragments and the interaction energy between the fragments composed of the Coulomb repulsion and the nuclear attraction. The system at scission is treated as a microcanonical ensemble where all available states are equiprobable. In this framework, the number of available states of a given fragmentation is the product of the state densities of the two isolated fragments. The yield of a fragmentation is the number of available states associated with this fragmentation regardless of the deformation of the fragments.

All nuclear inputs to the SPY model, i.e. the individual binding energies, proton distributions and proton and neutron single-particle level schemes of each fission fragment as a function of its axial deformation are estimated for about 7000 nuclei from Z=20Z = 20 to Z=100Z = 100 within the framework of the constrained self-consistent Hartree-Fock-Bogoliubov (HFB) formalism using the Skyrme BSk27 effective nucleon-nucleon interaction [65]. The nuclear state density is calculated in the framework of the statistical model of nuclear level densities [53, 66, 67, 68, 69] on the basis of the discrete single-particle level scheme obtained self-consistently within the same HFB calculation.

A scission criteria is needed to characterize the scission configurations, more precisely, to define the relative position of the fragments, hence to estimate the Coulomb and nuclear energies. This criteria can be based on an inter-surface distance, as in the initial version of the SPY model [64] or through the proton density at the scission neck which defines the scission distance as in our new version of the SPY model [53]. Presently, with respect to our last study [53], a phenomenological modification of the scission distance is introduced in order to better reproduce experimental fission yields. The corrected scission distance (dsc,corrd_{\mathrm{sc,corr}}) for a fragmentation (Z1,N1)+(Z2,N2)(Z_1, N_1) + (Z_2, N_2) is defined through the proton density at the scission neck (dscd_{\mathrm{sc}}), as in Ref. [53], though with two additional corrections, one per fragment

dsc,corr=dsc+dcorr(N1)+dcorr(N2).(1)d_{\mathrm{sc,corr}} = d_{\mathrm{sc}} + d_{\mathrm{corr}}(N_1) + d_{\mathrm{corr}}(N_2). \tag*{(1)}

Each correction term is composed by a positive term and a negative one which are chosen to be dependent on the difference between the fragment neutron number (NN) and some specific neutron numbers (NcorrN_{\mathrm{corr}}), i.e.

dcorr(N)=∑Ncorr∈{47,51,81,85,89}0.2 e−(N−Ncorr)22−∑Ncorr∈{64,66,68,70,80}0.2 e−(N−Ncorr)22 fm.(2)\begin{aligned} d_{\mathrm{corr}}(N) ={}& \sum_{N_{\mathrm{corr}}\in\{47,51,81,85,89\}} 0.2\,e^{-\frac{(N-N_{\mathrm{corr}})^2}{2}}\\ &- \sum_{N_{\mathrm{corr}}\in\{64,66,68,70,80\}} 0.2\,e^{-\frac{(N-N_{\mathrm{corr}})^2}{2}}\ \mathrm{fm}. \tag*{(2)} \end{aligned}

Both light and heavy fragments contribute to the distance correction (Figure 1a) with a maximal correction of ±0.4\pm0.4 fm. The available energy is impacted by a few MeV,

Distance correction and available energy for fission fragments

Figure 1. (Color online) (a) Distance correction associated with the light fragment, d(N1)d(N_1) with N1≤NCN/2N_1 \leq N_{\mathrm{CN}}/2 (orange curve), with the heavy fragment, d(N2)d(N_2) with N2≥NCN/2N_2 \geq N_{\mathrm{CN}}/2 (green curve), and with the total one d(N1)+d(N2)d(N_1)+d(N_2) (black dashed curve) for 236^{236}U fission. The specific neutron numbers NcorrN_{\mathrm{corr}} (Eq. (2)) are indicated by gray arrows. (b) Available energy from thermal neutron-induced fission of 235^{235}U calculated with (red curve) or without (green curve) the phenomenological distance correction (Eq. (2)). The blue line corresponds to the difference between corrected and non-corrected energies.

for example by around 3 MeV for the neutron-induced fission of 235^{235}U (Figure 1b). Such an effect is of the same order of magnitude as pairing effects and can be regarded as a way to correct the approximate pairing effects which assumes the pairing gap at zero temperature can be described by a five-points mass difference [70] and remains constant with the nucleus deformation. For nucleosynthesis applications, the distance correction is kept constant irrespective of the fissioning system, even if a specific adjustment could be performed.

This distance correction help to better reproduce the yields of the thermal neutron induced fission of 235^{235}U (Figure 2a and Figure 2d, green and blue curves), more particularly its symmetric component is reduced due to the distance correction around N=64,66,68,70N = 64, 66, 68, 70. As seen in Figure 2c and Figure 2f, this correction has a negligible impact on the FFD for the spontaneous fission of 252^{252}Cf. However, yields of the thermal neutron-induced fission of 239^{239}Pu around A=130A = 130 becomes underestimated (Figure 2b and Figure 2e) due to the same correction as in the U case.

Isotopic and isobaric fission yields of $^{236}\mathrm{U}$, $^{240}\mathrm{Pu}$ and $^{252}\mathrm{Cf}$

Figure 2. (Color online) Isotopic (top panels) and isobaric (bottom panels) fission yields of 236U^{236}\mathrm{U}, 240Pu^{240}\mathrm{Pu} and 252Cf^{252}\mathrm{Cf}. The black lines with dots represent experimental (pre-neutron-emission) fission yields for fission of 236U^{236}\mathrm{U} (Q=6.5Q = 6.5 MeV) (a) [71] and (d) [72], 240Pu^{240}\mathrm{Pu} (Q=6.5Q = 6.5 MeV) (b) [73] and (e) [74] and 252Cf^{252}\mathrm{Cf} (Q=0Q = 0 MeV) (c) [75] and (f) [76]. The green lines (labeled by “raw”) correspond to raw fission yields, the blue lines (labeled by “dcorrd_{\mathrm{corr}}”) to raw yields including distance corrections and the orange lines (labeled by “smooth”) to the final smooth yields after distance corrections.

As described in [53], the SPY FFD or yields distribution present strong staggering patterns that are not found experimentally. Those can be smoothed, as traditionally done, by a double Gaussian function

Ysmooth(Z,N)=∑i=−44∑j=−1515Yraw(Z+i,N+j)Cze−i22σz2Cne−j22σn2(3)Y_{\mathrm{smooth}}(Z,N)=\sum_{i=-4}^{4}\sum_{j=-15}^{15}Y_{\mathrm{raw}}(Z+i,N+j)C_z e^{-\frac{i^2}{2\sigma_z^2}}C_n e^{-\frac{j^2}{2\sigma_n^2}} \tag*{(3)}

where CzC_z and CnC_n are normalization factors and σz=0.55\sigma_z=0.55 and σn=3\sigma_n=3 are adopted. In the present paper the width of the Gaussian function is larger in the neutron direction σn\sigma_n than in the proton direction in order to be simultaneously consistent with experimental width of the isobaric and isotopic fission yields distributions (Figure 2). As there is not coupling between ZZ and NN in the double Gaussian function (Eq. (3)), the isotopic yields are not affected by the smoothing in the neutron direction. Similarly to yields, other fragments observables OO, such as the available energy or the kinetic energy, are also smoothed following the same procedure

Osmooth(Z,N)=∑i=−44∑j=−1515Oraw(Z+i,N+j)Cze−i22σz2Cne−j22σn2.(4)O_{\mathrm{smooth}}(Z,N)=\sum_{i=-4}^{4}\sum_{j=-15}^{15}O_{\mathrm{raw}}(Z+i,N+j)C_z e^{-\frac{i^2}{2\sigma_z^2}}C_n e^{-\frac{j^2}{2\sigma_n^2}}. \tag*{(4)}

Neutron evaporation model

The SPY model allows us to investigate the primary fragments, i.e. fragments formed at scission, hence their de-excitation by neutron evaporation. The excitation energy of fragments has two components. First, each fragment carries a fraction of the available energy at scission under an intrinsic excitation form. Second, since each fragment can be deformed, a deformation energy may contribute to the excitation energy. No assumption about available energy sorting at scission between primary fragments is needed within the SPY framework, contrary to other neutron evaporation models. Indeed, some models require to introduce an energy sorting parameter RT(A)R_T(A) [80, 81] or a thermal equilibrium to deduce the energy sorting on the basis of level density parameters with phenomenological corrections in order to increase the excitation energy of the light fragment [82].

In the SPY model, the energy sorting, denoted by xx (Eq. 2 in Ref. [53]), results from the competition between the state density of the light fragment (ρ1\rho_1) and the heavy one (ρ2\rho_2). The intrinsic excitation energy and deformation energy of each fragment are obtained by averaging over all possible energy sorting and deformations (Eq. 3 in Ref. [53]). In the simplest model, the competition between neutron and gamma evaporation during the de-excitation cascade is not taken into account, so that the fission fragments are sequentially de-excited by neutron evaporation as long as it is energetically possible. The kinetic energy of evaporated neutrons EnE_n can be assumed to follow a Maxwellian distribution

ϕM(En)=2πT3Ene−En/T,(5)\phi_M(E_n)=\frac{2}{\sqrt{\pi T^3}}\sqrt{E_n}e^{-E_n/T}, \tag*{(5)}

where TT is the nuclear temperature before neutron evaporation. TT can be extracted from the statistical model of nuclear level densities including BCS pairing interaction, denoted as TBCST_{\mathrm{BCS}} (see Eqs. 16 and 18 in Ref. [53] and Refs. [66, 67, 68]).

Following this approximation, the distribution of evaporated neutrons are calculated for the three well-known fissioning systems using both raw and smooth pre-neutron yields and shown in Figure 3. In both cases, the number of emitted neutrons is underestimated for the heaviest fragments A=150−160A=150-160 due to the low available energy, which, in turn, originates from an overestimate of the kinetic energy of these related fragmentations, as mentioned in Sec. IV of Ref. [53]. However, for heavy fragments with A=130−140A=130-140, the emitted neutron number is overestimated which, in this case, is due to an underestimate of the total kinetic energy of the fission fragments. The distributions of evaporated neutron deduced from smooth yields (Figure 3, orange curves) follow roughly the same pattern as the raw ones except for the heaviest fragments region where the number of emitted neutron is larger. This overestimation is due to the smoothing procedure (Eq. (4)) which leads to an increase of the total excitation energy by around 2 MeV, hence sometimes makes the emission of an additional neutron possible depending on its kinetic energy. More generally, the number of neutrons emitted by the light fragment is in better agreement with data, though slightly underestimated, than for the heavy fragment where ν\nu is systematically overestimated. This is linked to the available energy sorting within a fragmentation which is too favorable for the heavy fragment.

Evaporated neutron distributions as a function of the fragment mass number

Figure 3. (Color online) Evaporated neutron distributions as a function of the fragment mass number. The black dotted lines represent experimental evaporated neutron distribution of (a) 236U^{236}\mathrm{U} (Q=6.5Q = 6.5 MeV) [77], (b) 240Pu^{240}\mathrm{Pu} (Q=6.5Q = 6.5 MeV) [78] and (c) 252Cf^{252}\mathrm{Cf} (Q=0Q = 0 MeV) [79]. The orange/green lines (labeled by “smooth”/“raw”) are SPY evaporated neutron distribution using smooth/raw pre-neutron yields.

The overestimation the number of neutron emitted by the heavy fragments leads an overestimation of the total mean number of evaporated neutrons per fission. As displayed in Table I, it is overestimated by 0.77-1.1 neutron when smooth yields are used and by 0.6-1 neutron for raw yields. The overestimation is higher with smooth yields due to a more important contribution of the heavy fragments.

NucleusreactionENDFSPY(raw)SPY(smooth)
235U{}^{235}\mathrm{U}nif2.443.083.20
239Pu{}^{239}\mathrm{Pu}nif2.883.973.96
252Cf{}^{252}\mathrm{Cf}sf3.774.374.39

Table I. Total mean number of evaporated neutrons per fission, νˉtot\bar{\nu}_{\mathrm{tot}}, from the evaluated data library ENDF/B-VII.1 [83] and SPY using the raw or smooth yields (see Figure 3) for the 3 benchmarked nuclei shown in Figure 3 and obtained either by neutron-induced fission (nif) or spontaneous fission (sf).

To study the sensitivity of the predicted number of evaporated neutrons to the various model uncertainties, the impact of the neutron kinetic energy distribution ϕ(En)\phi(E_n), the nuclear temperature and the neutron-gamma competition are analyzed in the next sub-sections for the specific case of 252Cf{}^{252}\mathrm{Cf} spontaneous fission.

Sensitivity studies of the evaporation model

Impact of the kinetic energy distribution of the neutron

There are various models for the kinetic energy distribution of the neutron ϕ(En)\phi(E_{\mathrm{n}}) [84]. The chosen one is related to the Maxwell-Boltzmann distribution describing the distribution of speeds of particles in an idealized gas which yields to the kinetic energy distribution of neutron ϕM\phi_{\mathrm{M}} (Eq. (5)). Another possible description relies on the evaporation model of Weisskopf and Ewing [85, 86] yielding to

ϕW(En)=EnTn2e−En/Tn(6)\phi_{\mathrm{W}}(E_{\mathrm{n}})=\frac{E_{\mathrm{n}}}{T_{\mathrm{n}}^{2}}e^{-E_{\mathrm{n}}/T_{\mathrm{n}}} \tag*{(6)}

for small EnE_{\mathrm{n}} [84] and where TnT_{\mathrm{n}} is the nuclear temperature after the neutron emission. The impact of the neutron kinetic energy distribution on the number of evaporated neutron of 252Cf^{252}\mathrm{Cf} spontaneous fission is negligible, as shown in Figure 4, (orange and light green curves, labeled respectively by “ϕM,TBCS\phi_{\mathrm{M}},T_{\mathrm{BCS}}” and “ϕW,TBCS\phi_{\mathrm{W}},T_{\mathrm{BCS}}”) where the difference is seen to rarely exceed 0.1 neutron (Figure 4b, light green curve, labeled by “ϕW,TBCS\phi_{\mathrm{W}},T_{\mathrm{BCS}}”).

Evaporated neutron distribution and difference for spontaneous fission of $^{252}\mathrm{Cf}$

Figure 4. (Color online) (a) Evaporated neutron distribution of spontaneous fission of 252Cf^{252}\mathrm{Cf} using the SPY raw yields. (b) Difference with respect to the evaporated neutron distribution computed with the Maxwell distribution (Eq. (5)) and temperature from the BCS state density.

Impact of the nucleus temperature

As an alternative to the BCS theory of nuclear level densities, the phenomenological Fermi Gas approximation [87], excluding shell and pairing effects, can also be considered to estimate the nuclear temperature,

TFG=U/a(7)T_{\mathrm{FG}}=\sqrt{U/a} \tag*{(7)}

where UU is the excitation energy and a∼A/8a\sim A/8 is the empirical high-energy limit of the level density parameter [88]. As seen in Figure 4 (cyan and black curves, labeled respectively by “ϕM,TFG\phi_{\mathrm{M}},T_{\mathrm{FG}}” and “ϕW,TFG\phi_{\mathrm{W}},T_{\mathrm{FG}}”), both models for the nuclear temperature (TBCST_{\mathrm{BCS}} or TFGT_{\mathrm{FG}}) give rather similar predictions of the evaporated neutron distribution. For the sake of coherence, TBCST_{\mathrm{BCS}} is adopted in the subsequent study.

Impact of neutron-gamma competition

In the previous calculation of the number of emitted number of neutrons, it was assumed that neutron emission dominates over the electromagnetic de-excitation. To test the impact of such an approximation, TALYS nuclear reaction code [89] is now used to describe the competition between the strong and electromagnetic channels in the de-excitation of the primary fragments (Figure 4, green curve, labeled by “TALYS”). Note that all de-excitation channels are taken into account in the Hauser-Feshbach and pre-equilibrium framework. The difference between the evaporated neutron distribution from TALYS and the one obtained with the simplified Maxwellian approach described above is shown in Figure 4 (green curve, labeled by “TALYS”) and does not exceed 0.3 neutron.

These results indicate that the primary fragments de-excitation can be essentially approximated by a sequence of neutron emissions, followed finally by gamma emission when no more neutron can be evaporated. This simplified description based on energetic considerations seems to be sufficient within an accuracy of about 0.1–0.2 neutron. This conclusion is only valid for low excited primary fragments up to 20–25 MeV. The crucial point to improve the description of the emitted neutron distribution concerns the SPY model and particularly the kinetic energy of the fission fragments which directly affect the available energy and mean deformation of the fragments.

In the following, if not mentioned otherwise, the smooth yields (Eq. (3)) are adopted and neutron evaporation are reduced using the Maxwell distribution prescription (Eq. (5)) with the BCS temperature.

Systematics

Over all fissionable nuclei

Systematic SPY calculations were performed for isotopic chains between Z=70Z=70 and 124124 from the proton to the neutron driplines with an initial excitation energy Q=8Q=8 MeV (Figure 5). With Q=8Q=8 MeV, only few nuclei with Z<80Z<80 can fission. The smoothing procedure (Eq. (3)) may impact locally the peak multiplicity (Figure 5c) when compared those obtained with raw yields (Figure 5a). In particular, peaks can be merged if there are close each other or if they are small enough. However, the location of transition between the various fission modes are seen not to be globally affected by the smoothing procedure and does not distort significantly the peaks location of FFDs and their widths as shown in Figure 2.

Peak multiplicity for the raw pre-neutron isobaric yields with corrected distanceMean prompt neutron multiplicity per fission according to the prescription given in Sec. II.2, deduced from the smooth pre-neutron isobaric yields with corrected distancePeak multiplicity for the smooth pre-neutron isobaric yields with corrected distanceMean available energy release per fission deduced from the smooth pre-neutron isobaric yields with corrected distancePeak multiplicity for the smooth pre-neutron isobaric yields without corrected distanceIsolines of mean available energy release per fission and mean prompt neutron multiplicity per fission

Figure 5. (a) Peak multiplicity for the raw pre-neutron isobaric yields with corrected distance (Eq. (1)). (b) Mean prompt neutron multiplicity per fission according to the prescription given in § 0.2.2, deduced from the smooth pre-neutron isobaric yields with corrected distance (Eq. (1)). (c) Peak multiplicity for the smooth pre-neutron isobaric yields with corrected distance (Eq. (1)). (d) Mean available energy release per fission deduced from the smooth pre-neutron isobaric yields with corrected distance (Eq. (1)). (e) Peak multiplicity for the smooth pre-neutron isobaric yields without corrected distance. (f) Isolines of mean available energy release per fission (orange lines) from Figure 5d and mean prompt neutron multiplicity per fission (green lines) from Figure 5b. FIG. 5: (Color online) Systematics in the (N,Z)(N, Z) plane of major fission observables for some 3000 nuclei with 70≤Z≤12470 \le Z \le124 for an initial excitation energy of Q=8Q = 8 MeV.

Four zones can be identified where the peak multiplicity is affected by the distance correction (Eq. (2)), as seen when comparing Figure 5c and Figure 5e obtained with or without this correction. In general terms, a negative/positive distance correction makes fragments closer/farther which induced a decrease/increase of the available energy, hence of the fission probability. The first one is the shift of the symmetric/asymmetric fission transition toward neutron-deficient nuclei from NCN≈140N_{\mathrm{CN}} \approx140 to NCN≈133N_{\mathrm{CN}} \approx133. The symmetric mode is disfavored by the negative distance correction for Nfrag=64,66,68,70N_{\mathrm{frag}} = 64, 66, 68, 70 leading to the appearance of the asymmetric mode while for Nfrag=47,51,81,85,89N_{\mathrm{frag}} = 47, 51, 81, 85, 89 the positive distance correction favors the asymmetric mode. In the second zone, located at 155≲NCN≲165155 \lesssim N_{\mathrm{CN}} \lesssim165, a second asymmetric mode emerges, less asymmetric than the one without distance correction. It is induced by a negative distance correction for the light fragment of the most asymmetric fission mode, i.e. Nfrag=64,66,68,70N_{\mathrm{frag}} = 64, 66, 68, 70 and by a positive distance correction for the heavy fragment of the new slightly asymmetric fission mode (Nfrag=85,89N_{\mathrm{frag}} = 85, 89). The peak multiplicity increases from 2 to 4 when the distance corrections are important. For more neutron-rich nuclei, in the third zone 165≲NCN≲180165 \lesssim N_{\mathrm{CN}} \lesssim180, a symmetric mode emerges for high distance corrections. It is induced by a negative distance correction due to the light fragment of the asymmetric fission mode (Nfrag=64,66,68,70N_{\mathrm{frag}} = 64, 66, 68, 70) which disfavors the asymmetric mode. The symmetric mode is further accentuated by the positive distance correction for Nfrag=81,85,89N_{\mathrm{frag}} = 81, 85, 89. Finally, for nuclei in the fourth zone 180≲NCN≲200180 \lesssim N_{\mathrm{CN}} \lesssim200, the positive distance correction for Nfrag=85,89N_{\mathrm{frag}} = 85, 89 induces a slightly asymmetric fission mode on top of the symmetric one. Although some fission yields distributions are impacted by the phenomenological correction to the scission distance, globally, SPY predictions for neutron-rich nuclei remains rather robust with respect to such a correction, as seen in Figure 5c–Figure 5e.

The mean number of neutrons emitted per fission νˉtot\bar{\nu}_{\mathrm{tot}} ranges from 2 up to 37 neutrons depending on the fissioning system (Figure 5b). It depends mostly on the neutron richness of the fissioning nucleus. Its evolution with proton number is not trivial despite the strong variation of the mean available energy of fragmentations (Figure 5d). For a given fissioning nucleus, an increase of the excitation energy (QQ) enhances the intrinsic excitation energy of fission fragments, hence more neutrons are emitted. However, this relation between available energy and emitted neutrons is not sufficient to explain the relation between emitted neutrons and mean available energy of fragmentations. Along a line of constant available energy (Figure 5f) the number of evaporated neutron is multiplied by at least five between proton-rich and neutron-rich fissioning nuclei. Similarly, along a constant line of emitted neutron number (Figure 5f), the available energy increases by a factor up to hundred between light and heavy nuclei. These features are due to the neutron binding energy of fission fragments, i.e. the more neutron-rich the fissioning nucleus, the more-neutron rich the fission fragments and the lower their neutron binding energy, hence the more neutrons can be evaporated. Even if the fission of a heavy nucleus releases more available energy compared to a light one, the number of evaporated neutrons can be similar because the neutron binding energies of fragments from heavy-nucleus fission is higher (less neutron-rich) than the fragments from the lighter one.

SPY vs GEF

SPY FFDs with corrected distance are compared in Figure 6 to experimental data or with the evaluation from the ENDF/B-VII database [83], if no experimental data is available. Evaluated data (Figure 6, gray curves) correspond to post-neutron FFDs while experimental data can be pre-neutron or post-neutron emission (Figure 6, black curves). SPY FFDs (Figure 6, green curves) are also compared to GEF (version 2019/1.2) thermal neutron-induced fission yields [54, 103]. The excitation energies of experimental or evaluated data may not be exactly the same as those considered in SPY or GEF calculations but a variation of the excitation energy by a few MeV only impacts moderately the shape of the FFD.

Isobaric fission yields from SPY and GEF models compared with experimental yields

Figure 6. (a) Evaluated isotopic yields (formed by Coulomb excitation) from Ref. [90] converted into isobaric yields using the UCD (Unchanged Charge Density) hypothesis. For 230Th{}^{230}\mathrm{Th} post-neutron yields from ENDF/B-VII database [83].

(b) Yields of 232–233U{}^{232\text{--}233}\mathrm{U} are from [90] (converted into isobaric), 234U{}^{234}\mathrm{U} from [91], 235–237U{}^{235\text{--}237}\mathrm{U} (formed by 0.5 MeV neutron) from [83], 236U{}^{236}\mathrm{U} from [72], 238U{}^{238}\mathrm{U} and 239Np{}^{239}\mathrm{Np} and 244Cm{}^{244}\mathrm{Cm} from [92], 239U{}^{239}\mathrm{U} [76], 238Np{}^{238}\mathrm{Np} from[93], 240Pu{}^{240}\mathrm{Pu} from [94], 242–243Pu{}^{242\text{--}243}\mathrm{Pu} and 255Es{}^{255}\mathrm{Es} (formed by thermal neutron) from [83], 243Am{}^{243}\mathrm{Am} from [95], 244Am{}^{244}\mathrm{Am} from [96], 246Cm{}^{246}\mathrm{Cm} and 253Es{}^{253}\mathrm{Es} and 254–256Fm{}^{254\text{--}256}\mathrm{Fm} (spontaneous fission) from [83], 248Cm{}^{248}\mathrm{Cm} from [97], 250Cf{}^{250}\mathrm{Cf} from [98], 252Cf{}^{252}\mathrm{Cf} from [76], 256Cf{}^{256}\mathrm{Cf} and 258Fm{}^{258}\mathrm{Fm} from [99], 244–246–248Fm{}^{244\text{--}246\text{--}248}\mathrm{Fm} from [100], 259Lr{}^{259}\mathrm{Lr} from [101], 262Rf{}^{262}\mathrm{Rf} from [102].

FIG. 6: (Color online) Isobaric fission yields from SPY model (green curves) compared to those calculated by the GEF model (blue curves) [54] and experimental pre-neutron (black curves) or post-neutron (gray curves) yields. For each FFD, the x axis (mass number) ranges from A=70A = 70 to A=160A = 160 by step of 10 and the y axis (FFD) from 0 to 15% by step of 2%. The background color refers to the peak multiplicity.

:::

In the Po to U region (Figure 6a), the transition between symmetric fission for neutron-deficient nuclei to asymmetric fission for neutron-rich nuclei is rather different between both SPY and GEF predictions. SPY FFDs are in good agreement with experimental data [90] for Th and Pa isotopes where the transition is well reproduced, in contrast to what is obtained by the GEF model. The transition from symmetric to asymmetric fission predicted by GEF occurs for more neutron-deficient isotopes compared to experimental data.

For Th isotopes, GEF transition occurs between 216Th126{}^{216}\mathrm{Th}_{126} and 226Th136{}^{226}\mathrm{Th}_{136} for which no symmetric peak is found anymore. From an experimental point of view, the transition seems to start at 222Th132{}^{222}\mathrm{Th}_{132} where the symmetric peak broadens while beyond 229Th139{}^{229}\mathrm{Th}_{139} the FFD is completely asymmetric. The symmetric to asymmetric transition of Th predicted by SPY is slightly more complex (Figure 5e and Figure 6a); there are two peaks in addition to the symmetric component for 222–226Th132–136{}^{222\text{--}226}\mathrm{Th}_{132\text{--}136} which gives a peak multiplicity of two and even four for 226Th136{}^{226}\mathrm{Th}_{136}. For more neutron-rich Th, this substructure disappears into one symmetric peak which become smaller for more neutron-rich isotopes and completely vanishes for 235Th145{}^{235}\mathrm{Th}_{145} (Figure 5e). This kind of transition from symmetric to asymmetric through a substructure at the top of the vanishing symmetric peak is also visible on the SPY Pa FFDs.

The FFD broadening in the Rn to Ac isotopic chains is well reproduced by SPY even if the FFDs of neutron-deficient isotopes tend to be narrower than the experimental ones. A strong odd-even staggering in GEF FFD is observed for even isotopes of Po and Rn which makes FFD plots completely blue filled (Figure 6a).

For actinides from U to Rf (Figure 6b), FFDs from SPY and GEF model are rather similar and in good agreement with experimental data, particularly for U and Es isotopes. Compared to experimental data for Am and Cm isotopes, the symmetric part of the yields distributions is underestimated by both models. The peaks are narrower with SPY compared to GEF model. However, for neutron-rich nuclei (like 258Fm{}^{258}\mathrm{Fm}), SPY predicts a purely asymmetric fission, contrary to GEF and evaluated data. This asymmetric splitting predicted by SPY model is due to the doubly magic nuclei 132Sn{}^{132}\mathrm{Sn} which is disfavored. Indeed, even if 132Sn{}^{132}\mathrm{Sn} has more available energy than other fragmentations, the available states for these fragmentations is significantly lower and consequently their yields is lower than other fragmentations leading to an asymmetric fission. 132Sn{}^{132}\mathrm{Sn} affects fissioning systems in the region around 258Fm{}^{258}\mathrm{Fm} (Figure 5e) where a doubly asymmetric fission is found by SPY model whereas GEF model predicts an asymmetric mode. For the same reasons, the symmetric fission of 259Lr{}^{259}\mathrm{Lr} and 262Rf{}^{262}\mathrm{Rf} are not reproduced with the SPY model contrary to the GEF model.

Impact of fission on the r-process nucleosynthesis in NS mergers

The NS-NS merger simulation was performed with a general relativistic Smoothed Particle Hydrodynamics scheme [16, 31, 104] representing the fluid by a set of particles with constant rest mass, the hydrodynamical properties of which were evolved according to Lagrangian hydrodynamics. We only consider dynamical (prompt) ejecta expelled during the binary NS-NS merger in the present paper since this ejecta component provides generally the most favorable conditions for fission recycling due to the low initial electron fraction. A symmetric 1.365M⊙−1.365M⊙1.365M_{\odot} - 1.365M_{\odot} binary system compatible with the total mass detected in the GW170817 event [36] has been modelled on the basis of the so-called SFHO equation of state [105]. R-process calculations are performed on 500 ejected “particles” (i.e. mass elements) representative of the dynamical ejecta. The r-process nucleosynthesis is calculated with a reaction network including all 5000 species from protons up to Z=110Z = 110 lying between the valley of β\beta-stability and the neutron drip line. All charged-particle fusion reactions on light and medium-mass elements that play a role when the nuclear statistical equilibrium freezes out are included in addition to radiative neutron captures and photodisintegrations. The reaction rates on light species are taken from the NET-GEN library, which includes all the latest compilations of experimentally determined reaction rates [106]. Experimentally unknown reactions are estimated with the TALYS code [89, 107] on the basis of the Skyrme HFB nuclear mass model, HFB-21 [108]. On top of these reactions, β\beta-decays as well as β\beta-delayed neutron emission probabilities are also included, the corresponding rates being taken from the relativistic mean-field model of Ref. [109]. Fission processes, including neutron-induced, spontaneous and β\beta-delayed fission, are carefully introduced in the network together with the corresponding FFD. Fission barriers obtained with the BSk14 Skyrme interaction are adopted here to estimate fission probabilities [110, 111]. The new smooth FFDs from SPY model, as described in the previous section, as well as GEF predictions for comparison, are used in the present study. Note that all fission products with a yield larger than typically 10−5%10^{-5}\% are linked to the parent fissioning nucleus in the network calculation, i.e. about 500 fission fragments are included for each fissioning nucleus. The neutron-rich fission fragments located outside the network, i.e. across the neutron drip line, are assumed to instantaneously emit neutrons down to the neutron drip line. The total mass fraction as well as number of nucleons is conserved at all time and the emitted neutrons are recaptured consistently by all the existing species. Finally, α\alpha-decays are taken into account for all heavy species, the rates being extracted from Ref. [112].

Despite the recent success of nucleosynthesis studies for NS mergers, the details of r-processing in these events is still affected by a variety of uncertainties. In particular, the exact impact of neutrino is not yet understood in all details, the main reason of which is the not yet manageable computational complexity associated with full neutrino transport in a generically three-dimensional (3D), highly asymmetric environment with nearly relativistic fluid velocities and rapid changes in time. It was long assumed that neutrino interactions could not, at least not drastically, affect the initial neutron richness of the ejecta. However, recent NS-NS merger simulations [22, 55, 57, 58, 59, 60, 61, 63] including the effect of weak interactions on free nucleons demonstrate that neutrino reactions can significantly affect the neutron-to-proton ratio in the merger ejecta, with direct consequences in particular on the amount of synthesized low-mass (A<140A < 140) r-nuclides produced in the dynamical ejecta along with the heavier species. To broadly study the impact of weak interactions of free nucleons on the r-process nucleosynthesis, a parametric approach in terms of prescribed neutrino luminosities and mean energies, guided by hydrodynamical simulations, was followed in Ref. [56]. A similar approach is also considered here to test the role of fission during the r-process nucleosynthesis in mainly two scenarios. The first one (hereafter scenario I) neglects all weak interactions on free nucleons and assumes the initial electron fraction of the ejected material is consequently not modified at the beginning of the expansion. In this case, the mean initial electron fraction at the time of the network calculation (i.e. when the temperature has dropped below 10 GK) amounts to ⟨Ye⟩=0.03\langle Y_e \rangle= 0.03. This case is likely to be relevant for NS-BH mergers or highly asymmetric NS-NS mergers, where the lower mass component develops an extended tidal tail, from which considerable amounts of cold, unshocked matter can be centrifugally ejected before neutrino exposure of these ejecta may play an important role [18, 21, 23]. In the second case (scenario II), the equilibrium between electrons, positrons and neutrinos is assumed to hold down to the fiducial density of ρeq=1012 g cm−3\rho_{\mathrm{eq}} = 10^{12}\ \mathrm{g\,cm^{-3}}, and from that density on, electron, positron and electron neutrino and antineutrino captures on nucleons

νe+n⇌p+e−νˉe+p⇌n+e+(8)\nu_e + n \rightleftharpoons p + e^{-} \\ \bar{\nu}_e + p \rightleftharpoons n + e^{+} \tag*{(8)}

are systematically included assuming the (anti)neutrino luminosities and mean energies remain constant in time, as detailed in Ref. [56], and compatible with the results obtained within the Improved Leakage-Equilibration-Absorption Scheme (ILEAS) method [63] (see in particular their Fig. 14). To do so, we adopt the values of Lνe=0.3×1053L_{\nu_e} = 0.3 \times10^{53} erg/s; Lνˉe=1053L_{\bar{\nu}_e} = 10^{53} erg/s; ⟨Eνe⟩=8\langle E_{\nu_e} \rangle= 8 MeV; ⟨Eνˉe⟩=12\langle E_{\bar{\nu}_e} \rangle= 12 MeV. The resulting mean electron fraction at the time of the network calculation reaches ⟨Ye⟩=0.23\langle Y_e \rangle= 0.23 in relatively good agreement with the 0.26 value obtained in Ref. [63].

Without weak interaction, the initial electron fraction, Ye0Y_e^0, of the various mass elements is low and characteristic of the NS inner crust, i.e. less than 0.06; consequently the number of free neutrons per seed nuclei can reach a few hundreds. With such a neutron richness, heavy fissioning nuclei can be efficiently produced during the r-process. Weak interactions on nucleons (Eq. 8) increase the initial electron fraction in the ejecta to values ranging in this scenario II from 0.06 to 0.38 and make the production of heavy fissioning nuclei and fission recycling during the r-process less favorable, though not impossible.

On the basis of the BSk14 fission barriers adopted here, the production of super-heavy neutron-rich nuclei is possible up to Z=110Z = 110 if the neutron irradiation is large enough to overcome the N=184N = 184 shell closure. Beyond Z=110Z = 110, the production of elements is unlikely because these elements fission spontaneously (Fig. 7a) with a very short lifetime (Fig. 7b). This is the main reason why the reaction network has been limited to elements Z≤110Z \le110. Nuclei reaching this limit (Z=110Z = 110 and A≥320A \ge320) define a “fission roof” area (A≥320A \ge320) (Fig. 7d).

To produce such super-heavy neutron-rich nuclei, the matter has to flow through the fissioning region formed by (i) β\beta-delayed fission of 94286Pu192{}^{286}_{94}\mathrm{Pu}_{192}, (ii) spontaneous fission of 95286Am191{}^{286}_{95}\mathrm{Am}_{191}, (iii) β\beta-delayed fission of 97288–289Bk191–192{}^{288\text{--}289}_{97}\mathrm{Bk}_{191\text{--}192}, and (iv) spontaneous fission of elements with Z≥98Z \ge98 (Fig. 7a). 95286Am191{}^{286}_{95}\mathrm{Am}_{191} and Z>98Z > 98 elements have lifetime shorter than 10−410^{-4} s, which makes neutron capture by these nuclei unlikely. The β\beta-delayed fission lifetime of 94286Pu192{}^{286}_{94}\mathrm{Pu}_{192} and 97288–289Bk191–192{}^{288\text{--}289}_{97}\mathrm{Bk}_{191\text{--}192} ranges from 10−310^{-3} to 10−210^{-2} s. The neutron-induced fission of 98287–291Cf189–193{}^{287\text{--}291}_{98}\mathrm{Cf}_{189\text{--}193} and 95285Am190{}^{285}_{95}\mathrm{Am}_{190} isotopes also happens to be efficient and contributes to fission recycling (Fig. 7c). The flow through Pu isotopes is negligible due to the fast β\beta-delayed neutron emission of 94278Pu184{}^{278}_{94}\mathrm{Pu}_{184} with a lifetime τβkn=6.6×10−4\tau_{\beta\mathrm{kn}} = 6.6 \times10^{-4} s and 94279Pu185{}^{279}_{94}\mathrm{Pu}_{185} with τβkn=4.7×10−3\tau_{\beta\mathrm{kn}} = 4.7 \times10^{-3} s (Fig. 7b). In addition, the neutron separation energy of 279Pu{}^{279}\mathrm{Pu} equals to 0.28 MeV which makes the neutron capture by 278Pu{}^{278}\mathrm{Pu} unfavorable. Consequently, matter mainly flows through the Cm isotopic chain to produce super-heavy neutron-rich elements. This region forms a “fission bottleneck” (A<292A < 292 and N>184N > 184) where 96287Cm191{}^{287}_{96}\mathrm{Cm}_{191} represents the “crossing point” (Fig. 7d). However, the production of super-heavy neutron-rich nuclei is no more possible for temperatures higher than 2 GK (i.e. kBT=0.17k_{\mathrm{B}}T = 0.17 MeV) because the neutron-induced fissions become more favorable than the radiative neutron captures of 96285–288Cm189–192{}^{285\text{--}288}_{96}\mathrm{Cm}_{189\text{--}192} and 97285Bk188{}^{285}_{97}\mathrm{Bk}_{188} (Fig. 7c) which close the “fission bottleneck”.

At the end of the neutron irradiation, heavy neutron-rich elements β\beta-decay up to the region of efficient fission. Elements accumulated during the neutron irradiation along the N=184N = 184 neutron shell closure β\beta-decay up to reach a region of efficient fission. This can be split in two zones : the “decay roof” (A<278A < 278) and “decay roof 1” (A≥278A \ge278 and N≤184N \le184) which correspond to the region reached by nuclei Z<94Z < 94 and Z≥94Z \ge94, respectively. Elements with 292≤A<320292 \le A < 320 accumulated after the “fission bottleneck” β\beta-decay up to reach another region of efficient fission named “decay roof 2” (292≤A<320292 \le A < 320).

In the following subsection (Sec. III.1), we present the final abundances of the ejected material obtained in both scenarios as well as their sensitivity to the initial conditions of the trajectories. The fission contribution to the r-process and its impact on the final abundances will be

studied in both scenarios in Sec. III.2. The contributions of the various fissioning regions to the fission recycling is studied in Sec. III.3, while, in Sec. III.4, the impact of the FFDs on the final abundances are studied with a specific comparison made when adopting SPY or GEF FFDs.

Final abundances

The final abundance distributions of the ejected material are shown in Figure 8 for both scenarios I and II. Calculations are performed adopting either the SPY or GEF predictions of the FFDs. While in scenario I, only nuclei with A≳120A \gtrsim120 are essentially produced, weak interactions on nucleons also allow the production of lighter species with 80≲A≲12080 \lesssim A \lesssim120 in scenario II, but also decrease the production of the Pb peak nuclei as well as the long-lived actinides. In this case, the choice of the FFD model (SPY or GEF) is seen to have a negligible impact on the final abundance distribution. The same conclusions hold if we consider our original version of the SPY FFD [53]. Without weak interactions, with the low initial electron fractions, the number of neutron per seed is rather high (a few hundreds), so that the production of the heaviest nuclei is efficient. In this case, nuclear fission plays a fundamental role during both the neutron irradiation and at freeze-out [32] and the FFD model directly affects the abundance distribution, particularly around the second peak A=130A = 130 where SPY FFDs give higher abundances than GEF (see Sec. III.4). This nucleosynthesis has been described in detail in Refs. [17, 18, 31, 34, 56, 62]. Note that the corrections introduced in the present study (Sec. III.1) in comparison with our 2019 version [53] have a small impact on the final abundance distribution. They tend to slightly increase the second r-abundance peak at A∼130A \sim130 and decrease its low-AA tail. Such differences are significantly smaller than the one observed between the SPY and GEF predictions.

Final composition in terms of molar fraction of matter ejected in scenarios I and II

Figure 8. (Color online) Final composition (in terms of molar fraction) of matter ejected in scenarios I and II and obtained either with SPY or GEF FFDs. For comparison, the solar r-abundance distribution [113] (blue dotted circles) is shown and arbitrarily normalized.

It is well accepted [114] that the efficiency of the r-process nucleosynthesis can be related to three main physical properties of the ejected material, namely the electron fraction YeY_{\mathrm{e}}, the entropy per baryon SS and the expansion timescale τ\tau [115]. In scenario I, trajectories are characterized by a low initial electron fraction Ye0Y_{\mathrm{e}}^{0} ranging between 0.02 and 0.06 at the beginning of the nucleosynthesis, i.e. when the temperature has dropped below 10 GK and the density below the drip density (ρdrip∼4×1011 g/cm3\rho_{\mathrm{drip}}\sim4\times10^{11}\ \mathrm{g}/\mathrm{cm}^{3}) (Figure 9g). Initial entropies S0S^{0} reach values up to 170 kBk_{\mathrm{B}}/nucleon, though most of them remain below 50 kBk_{\mathrm{B}}/nucleon. As already mentioned, in scenario II, the initial electron fraction covers a much larger range between 0.06 and 0.38 due to the weak interactions (Eq. (8)).

Final abundances, trajectory distribution, and group properties

Figure 9. (Color online) (a-e) Final abundances of the trajectories computed within scenario II classified into 5 groups: “1st peak”, “between 1st and 2nd peak”, “2nd peak”, “rare earth & no fiss.” and “fiss.”. The green curves with dots are the final abundances averaged over all trajectories computed within scenario II (a-e). The gray curves with dots correspond to the mean final abundances within each group. (f) Final abundances of the trajectories computed within scenario I (black curves) and the final abundances averaged over all trajectories (gray curve with dots). (g) Distribution of the trajectories in the (S0,Ye0)(S^{0},Y_{\mathrm{e}}^{0}) plane and iso-Xn0X_{\mathrm{n}}^{0} curves (green curves). (h) Mean value of the initial neutron mass fraction Xn0X_{\mathrm{n}}^{0} (green curve) and mean mass ⟨A⟩A≥50\langle A\rangle_{A\ge50} (gray curve) of each group and the associated ejecta mass fraction of groups for the scenario II (color bars).

Both the initial electron fraction and entropy affect the initial mass fraction of free neutrons Xn0X_{\mathrm{n}}^{0} (Figure 9g, green curves) which is consequently a parameter of first relevance in the analysis of the efficiency of the r-process nucleosynthesis. In scenario I, Xn0X_{\mathrm{n}}^{0} ranges between 0.85 and 0.98, as seen in Figure 9g, while in scenario II it varies between 0.02 and 0.82. The lower Ye0Y_{\mathrm{e}}^{0}, the larger Xn0X_{\mathrm{n}}^{0} by definition, but also the higher the initial entropy, the larger Xn0X_{\mathrm{n}}^{0} due to the release of free neutrons through photodissociation. Therefore, Xn0X_{\mathrm{n}}^{0} engulfs information on both the initial electron fraction and entropy. The sudden variation of Xn0X_{\mathrm{n}}^{0} observed at S0=8S^{0}=8–10 kBk_{\mathrm{B}}/nucleon, independently of Ye0Y_{\mathrm{e}}^{0}, is due to the initial abundances of trans-alpha elements (A>4A>4): ∑A>4Y0(A)\sum_{A>4}Y^{0}(A). The matter is not completely dissociated for trajectories with a low initial entropy S0<10 kBS^{0}<10\ k_{\mathrm{B}}/nucleon, which makes the initial abundance of trans-alpha elements (with a mean mass ⟨A⟩A>40\langle A\rangle_{A>4}^{0} ranging from 50 to 85) non-negligible and reduces the initial neutron mass fraction. For S0>10 kBS^{0}>10\ k_{\mathrm{B}}/nucleon, the initial abundances of trans-alpha elements is negligible, i.e. the ejected nuclear matter is completely dissociated into protons, neutrons and alphas. The low entropy effect is more important for trajectories in scenario II where for 18% of the ejected mass, ∑A>4Y0(A)>0.001\sum_{A>4}Y^{0}(A)>0.001, whereas in scenario I, only 5% of the ejecta has ∑A>4Y0(A)>0.001\sum_{A>4}Y^{0}(A)>0.001. Trans-alpha elements can boost the r-process depending on the neutron irradiation. Initial and final abundance distributions are more or less similar for trajectories with a low initial neutron mass fraction (Xn0<0.3X^{0}_{n}<0.3) and a low initial entropy (S0<10 kB/nucleonS^{0}<10\ k_{B}/\mathrm{nucleon}); in both cases, the r-process is weak due to the low neutron irradiation.

The relevance of the initial neutron mass fraction Xn0X^{0}_{n} in the analysis of the efficiency of the r-process can be clearly seen in Figure 9. The various trajectories can be classified according to their final abundance distributions (Figure 9a–Figure 9f). The trajectories computed within scenario I are classified in the group labeled “no ν\nu” while those computed within scenario II are divided into five groups: the group labeled “fission” contains trajectories with the highest fission recycling (Xfiss,totcum>0.01X^{\mathrm{cum}}_{\mathrm{fiss,tot}}>0.01, more details about the definition of this fission recycling indicator will be given in Sec. III.2). Among the trajectories with ∑125≤A≤135Yf(A)>0.01\sum_{125\leq A\leq135}Y^{f}(A)>0.01, if ∑190≤A≤200Yf(A)<0.001\sum_{190\leq A\leq200}Y^{f}(A)<0.001 then they are classified in the group “2nd peak” else in the group “rare earth & no fission”. The trajectories with ∑75≤A≤85Yf(A)>0.1\sum_{75\leq A\leq85}Y^{f}(A)>0.1 are classified in the group “1st peak”. Remaining trajectories correspond to the group “between 1st and 2nd peak”. The two first groups “1st peak” and “between 1st and 2nd peak” represent each 7% and 6%, respectively, of the ejected mass within scenario II (see Figure 9h). The other groups “2nd peak”, “rare earth & no fission” and “fission” represent 41%, 30% and 16%, respectively. The mean initial neutron mass fraction increases with the group (Figure 9h, green curve), heavy elements production is possible only with a high initial neutron mass fraction. Similarly, the mean mass of elements heavier than 50 (⟨A⟩A>50\langle A\rangle_{A>50}) also increases with the group (Figure 9h, gray curve) but reaches a maximum around 170 when fission occurs in groups “fission” and “no ν\nu” corresponding to the saturation regime induced by the fission recycling (see Sec. III.2). In the (S0,Ye0S^{0},Y^{0}_{e}) plane (Figure 9g), the trajectories of the various groups are distributed according the iso-Xn0X^{0}_{n} lines which clearly demonstrate the relevance of the initial neutron mass fraction Xn0X^{0}_{n} in the analysis of the r-process efficiency.

Fission recycling

Fission recycling for a given trajectory can be quantified by the so-called total cumulative fissioning mass fraction

Xfiss,totcum=∑Z,A=r fiss∫dXCdt(Z,A,t,r)dt=∑Z,A∫X(Z,A,t)[Nn⟨σv⟩n,fZ,A+λsfZ,A+λβdfZ,A]dt=∑Z,AXfisscum(Z,A)(9)\begin{aligned} X^{\mathrm{cum}}_{\mathrm{fiss,tot}}=\sum_{Z,A=r\ \mathrm{fiss}}\int\frac{dX_{C}}{dt}(Z,A,t,r)dt \\ &=\sum_{Z,A}\int X(Z,A,t)\left[N_{n}\langle\sigma v\rangle_{n,f}^{Z,A}+\lambda_{\mathrm{sf}}^{Z,A}+\lambda_{\mathrm{\beta df}}^{Z,A}\right]dt \\ &=\sum_{Z,A}X^{\mathrm{cum}}_{\mathrm{fiss}}(Z,A) \tag*{(9)} \end{aligned}

where dXCdt(Z,A,t,r)\frac{dX_{C}}{dt}(Z,A,t,r) corresponds to the reaction flux, expressed in terms of the mass fraction XX for a given nucleus (Z,A)(Z,A), at a given time tt and for a given fission reaction rr. Xfiss,totcumX^{\mathrm{cum}}_{\mathrm{fiss,tot}} consequently corresponds, at a given time, to the cumulative mass fraction of fissioning nuclei that are destroyed by fission reactions and gives a quantitative indicator of the amount of mass recycled by fission processes. Fission can take place by spontaneous (at a rate λsf\lambda_{\mathrm{sf}}) or β\beta-delayed (at a rate λβdf\lambda_{\mathrm{\beta df}}) fission or by neutron-induced fission (at a rate ⟨σv⟩n,f\langle\sigma v\rangle_{n,f}, where ⟨σv⟩n,f\langle\sigma v\rangle_{n,f} is the corresponding astrophysical rate and NnN_{n} the neutron density). Consequently, there are three factors contributing to the fission recycling: (i) the quantity of fissionable nuclei, (ii) the fission rates and (iii) the amount of time over which fission reactions may occur.

The total fissioning cumulative mass fraction Xfiss,totcumX^{\mathrm{cum}}_{\mathrm{fiss,tot}} increases with initial neutron mass fraction Xn0X^{0}_{n} (Figure 10, black curve with dots). The fission recycling can therefore be explained in term of initial neutron mass fraction Xn0X^{0}_{n}, i.e. the more neutron-rich the initial ejecta, the larger the production of heavy elements, hence the more efficient the fission.

Initial neutron mass fraction and ejecta mass distribution depending on fission recycling

Figure 10. (Color online) Initial neutron mass fraction Xn0X^{0}_{n} (black curve with dots) and the mass distribution of the ejecta (blue curve) depending on the fission recycling Xfiss,totcumX^{\mathrm{cum}}_{\mathrm{fiss,tot}}. The dashed line separates abundances computed within scenario I (Xfiss,totcum>1X^{\mathrm{cum}}_{\mathrm{fiss,tot}}>1) to the ones computed within scenario II (Xfiss,totcum<1X^{\mathrm{cum}}_{\mathrm{fiss,tot}}<1).

Scenario I

Within scenario I, Xfiss,totcum>1X^{\mathrm{cum}}_{\mathrm{fiss,tot}}>1 for all ejected trajectories due to the high initial neutron mass fraction (Xn0>0.85X^{0}_{n}>0.85). In Figure 11a, the final abundances of the 500 trajectories have been sorted and binned as a function of Xfiss,totcumX^{\mathrm{cum}}_{\mathrm{fiss,tot}} which is found to range between 1.19 and 1.82 with a mean value over all trajectories of

Final abundances and fissioning cumulative mass fraction distributions

Figure 11. (a) For scenario I where the mean value of the total fissioning cumulative mass fraction ⟨Xfiss,totcum⟩traj=1.45\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} = 1.45. (b) For scenario II where the mean value of the total fissioning cumulative mass fraction ⟨Xfiss,totcum⟩traj=0.03\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} = 0.03. (Color online) (α\alpha) Final abundances obtained with SPY FFDs, averaged for each bin of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} colored by their Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} value and compared to the average final abundances (green solid curve with dots). (β\beta) Same final abundances in the (A,Xfiss,totcum)(A, X_{\mathrm{fiss,tot}}^{\mathrm{cum}}) plan colored by their abundances. The mean mass ⟨A⟩A≥50\langle A\rangle_{A\geq50} for each Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} bins are represented by red crosses.

⟨Xfiss,totcum⟩traj=1.45\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} = 1.45. The final abundances distribution for a given trajectory can be directly linked to its total fissioning cumulative mass fraction Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}}. An overall shift of final abundances towards heavier elements can be expected with increasing fission recycling but the mean A≥50A \geq50 mass, ⟨A⟩A≥50\langle A\rangle_{A\geq50}, does not increase with Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} (Figure 11a$\beta$, red crosses), and stays rather constant between 166 and 172. This value remains constant because the final abundances are dominated by the second (A=130A = 130) and third (A=195A = 195) peaks.

The final abundance of light elements (A≤4A \leq4), mainly α\alpha-particles, is significant with 0.1<∑A≤4Yf(A)<0.840.1 < \sum_{A\leq4}Y^{\mathrm{f}}(A) < 0.84 depending on the trajectory. It increases with the initial entropy S0S^{0}, inducing an overall downward shift of the final abundances of heavier elements (A>4A > 4). As Xn0X_{\mathrm{n}}^{0} is linked to Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} (Figure 10), the initial entropy S0S^{0} of trajectories within a given bin of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} covers almost the whole range of possible S0S^{0} (since Xn0X_{\mathrm{n}}^{0} isolines evolve in the (S0S^{0}, Ye0Y_{\mathrm{e}}^{0}) plane, as seen in Figure 9g). This S0S^{0} admixture of trajectories within a bin of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} smooths the overall shift of the final abundances of heavy elements (A>4A > 4). However, fission recycling has a significant impact : the final A<195A < 195 abundances decrease with Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}}, whereas for elements A>195A > 195 they increase (Figure 11a$\alpha$). With increasing values of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}}, the abundances in the mass region 140<A<170140 < A < 170 drop by an order of magnitude, but around A=202A = 202 they increase from 4×10−44 \times10^{-4} up to 2×10−32 \times10^{-3}, while for A<80A < 80 elements, they drop below 10−510^{-5} (Figure 11a).

Neglecting the effect related to the variation of light elements abundance, a higher initial neutron mass fraction Xn0X_{\mathrm{n}}^{0} induces more fission recycling which implies a decrease of the final A<80A < 80 abundances. A high neutron irradiation leads to a significant production of super-heavy elements (A>300A > 300) whose fission fragments combined to new neutron captures shift the third peak to higher masses.

Part of the nuclei that accumulated during neutron irradiation along the neutron drip line (including the neutron shell closure N=184N = 184) with 208<A≤278208 < A \leq278 are the progenitors of the trans-lead elements. Among these accumulated nuclei, those with 260≤A≤278260 \leq A \leq278 (79≤Z≤9479 \leq Z \leq94) predominantly β\beta-decay up to reach the “decay roof 0” where they fission. Those with 208<A<232208 < A < 232 β\beta- and α\alpha-decay up to the third abundance peak, while the remaining intermediate nuclei with 232≤A≤260232 \leq A \leq260 β\beta- and α\alpha-decay up to the third abundance peak or β\beta-decay and reach the “decay roof 0” where they fission or decay into the long-lived 232Th{}^{232}\mathrm{Th} and 235−238U{}^{235-238}\mathrm{U} nuclei. Final A>204A > 204 abundances decrease with Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} because their progenitors are the nuclei accumulated along the neutron shell closure N=184N = 184 during the neutron irradiation. This accumulation is reduced for high neutron irradiations which enable to overcome the neutron shell closure N=184N = 184, hence to increase the fission recycling.

Scenario II

With the low initial neutron mass fraction Xn0X_{\mathrm{n}}^{0} found in scenario II, about 55% of the ejected mass is not subject to fission recycling (Xfiss,totcum<10−7X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 10^{-7}) (Figure 10, blue curve) and the maximum value of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} is 0.6. The fission recycling is reduced by a factor 50 with a mean value of the total fissioning cumulative mass fraction ⟨Xfiss,totcum⟩traj\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} that now amounts to 0.03 to be compared with 1.45 for scenario I. The present scenario is well suited to study low fission recycling effects on the final abundances. For this reason, trajectories are sorted in log⁡10\log_{10} bins of their total fissioning cumulative mass fraction, as shown in Figure 11b. The bins, except for the first one, represent each about 5% to 10% of the ejected mass (Figure 10).

The final abundance of the light A≤4A \leq4 elements in bins are more or less similar ranging from 40% to 55% with no global trend which induces a smaller shift of the final abundances due to trans-alpha elements in comparison with scenario I.

Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} is linked to the initial neutron mass fraction Xn0X_{\mathrm{n}}^{0}, as observed in Figure 10 (black curve with dots) where Xn0X_{\mathrm{n}}^{0} is seen to increase linearly with log⁡10(Xfiss,totcum)\log_{10}\left(X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\right) from 0.5 to 0.75. In other words, Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} increases exponentially with Xn0X_{\mathrm{n}}^{0}. An increase with Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}}, hence with Xn0X_{\mathrm{n}}^{0}, of the mean mass ⟨A⟩A≥50\langle A\rangle_{A\geq50} from 105 up to 168 is observed in Figure 11b$\beta$ (red crosses), i.e. the higher Xn0X_{\mathrm{n}}^{0}, the heavier the final nuclei produced.

For trajectories with a low initial neutron mass fraction, Xn0≈0.5X_{\mathrm{n}}^{0} \approx0.5 and Xfiss,totcum<10−7X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 10^{-7}, A>130A > 130 nuclei, including lanthanides, are not significantly produced in comparison with the other trajectories (Figure 11b) and production beyond A>180A > 180 is negligible. Fission recycling does not take place in this case because there are not enough neutrons to overcome significantly the N=82N = 82 neutron shell closure. For higher initial neutron mass fraction Xn0≈0.6X_{\mathrm{n}}^{0} \approx0.6 (10−7<Xfiss,totcum<10−510^{-7} < X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 10^{-5}), nuclei in the rare earth and third peak regions start to be produced, but those with A>200A > 200 are still highly underproduced and fission recycling remains negligible.

Abundances of third peak elements increase in the range 10−5<Xfiss,totcum<10−110^{-5} < X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 10^{-1} (0.65<Xn0<0.710.65 < X_{\mathrm{n}}^{0} < 0.71) as does the mean mass ⟨A⟩A≥50\langle A\rangle_{A\geq50} which increases linearly with log⁡10(Xfiss,totcum)\log_{10}\left(X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\right) up to 168. In this Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} range, abundances of rare earth nuclei (145<A<180145 < A < 180) increase to reach a maximum at 10−4<Xfiss,totcum<10−210^{-4} < X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 10^{-2} before decreasing for Xfiss,totcum>10−2X_{\mathrm{fiss,tot}}^{\mathrm{cum}} > 10^{-2}. The N=126N = 126 and N=184N = 184 neutron shell closures start to be overcome for Xfiss,totcum>10−4X_{\mathrm{fiss,tot}}^{\mathrm{cum}} > 10^{-4} leading to the synthesis of A>204A > 204 nuclei in a non-negligible amount. Fission recycling starts to play a role for trajectories with Xfiss,totcum>10−4X_{\mathrm{fiss,tot}}^{\mathrm{cum}} > 10^{-4} (or equivalently with Xn0>0.65X_{\mathrm{n}}^{0} > 0.65), which represent 29% of the ejected mass in scenario II.

In the last bin of total fissioning cumulative mass fraction, i.e. 0.1<Xfiss,totcum<0.610.1 < X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 0.61 and an average initial neutron mass fraction Xn0=0.76X_{\mathrm{n}}^{0} = 0.76, the third r-abundance peak becomes higher than the second one and A>204A > 204 nuclei dominate the composition. The mean mass ⟨A⟩A≥50\langle A\rangle_{A\geq50} does not increase significantly, a saturation regime is reached with ⟨A⟩A≥50≈170\langle A\rangle_{A\geq50} \approx170 as observed for scenario I (see Sec. III.2.1); the r-process dynamics for such trajectories is similar to the one taking place in scenario I.

To complete the discussion of Sec. III.2.1, fission recycling plays a dominant role when Xfiss,totcum>0.1X_{\mathrm{fiss,tot}}^{\mathrm{cum}} > 0.1 corresponding to Xn0>0.75X_{\mathrm{n}}^{0} > 0.75 which concerns 9.5% of the ejected mass within scenario II. A saturation regime is reached ⟨A⟩A≥50≈170\langle A\rangle_{A\geq50} \approx170, when a high initial neutron mass fraction Xn0X_{\mathrm{n}}^{0} induces more fission processes feeding back the 80 < AA < 200 mass region.

Fission seeds

We now turn to the mean contribution of fissioning nuclei, or fission seeds, to the fission recycling and to the final abundance distribution. The contribution of a given fissioning nucleus (Z,A)(Z,A) to the fission recycling, xFS(Z,A)x_{\mathrm{FS}}(Z,A), can be defined, for a given set of trajectories, as the fraction of the total fissioning cumulative mass fraction ⟨Xfiss,totcum⟩traj\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} due to this specific fissioning nucleus (Z,A)(Z,A), i.e.

xFS(Z,A)=⟨Xfisscum(Z,A)⟩traj⟨Xfiss,totcum⟩traj.(10)x_{\mathrm{FS}}(Z,A)=\frac{\left\langle X_{\mathrm{fiss}}^{\mathrm{cum}}(Z,A)\right\rangle_{\mathrm{traj}}}{\left\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\right\rangle_{\mathrm{traj}}}. \tag*{(10)}

In the specific NS merger cases studied here, ⟨Xfiss,totcum⟩traj\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} reaches rather similar values of 1.45 and 1.41 in scenario I when using SPY or GEF FFDs, respectively and 0.03 for both FFDs within scenario II. This similitude is due to the fact that SPY and GEF FFDs are not fundamentally different (see Figure 15) and the properties of the fission seeds are not related to their FFDs (see Figure 7). Consequently, only results with SPY FFDs are presented here.

Representation in the $(N,Z)$ plan of decay modes, lifetimes, fission temperatures, and fission seed areas

Figure 7. (Color online) Representation in the (N,Z)(N,Z) plan the decay mode with the highest decay rate (a) between β\beta decay followed by 0, 1, 2 or 3 neutron emission (βkn\beta kn), β\beta-delayed fission (βdf\beta df), spontaneous fission (sf) or α\alpha decay; (b) the corresponding lowest lifetime τmin⁡=1/λmax⁡\tau_{\min}=1/\lambda_{\max} among the various decay mode given in (a). (c) Temperature at which fission and radiative neutron capture are equally favorable, above this temperature the neutron induced fission dominates the neutron capture. (d) Nuclei for which neutron-induced fission is dominant over the (n,γ)(n,\gamma) channel for T<0.2T<0.2 GK or for which the main decay mode is spontaneous fission or β\beta-delayed fission are depicted by the gray area (region of efficient fission. Blue hatched areas are the fission seeds areas : the “fission bottleneck” (A<292A<292 and N>184N>184) hatched by blue circles, the “fission roof” (A≥320A\geq320) hatched by blue vertical+horizontal lines, the “decay roof 0” (A<278A<278) hatched by blue vertical lines, the “decay roof 1” (A≥278A\geq278 and N≤184N\leq184) hatched by blue horizontal lines and the “decay roof 2” (292≤A<320292\leq A<320 and N>184N>184) hatched by blue diagonal lines. The double black lines show the Z=82Z=82, N=126N=126 and N=184N=184 shell closures. See text for more details.

Post-neutron fission fragment distributions around the neutron shell closure

(a) Around the neutron shell closure N=184N = 184, these nuclei are responsible for 64/86% of the fission recycling in scenario I/II.

Post-neutron fission fragment distributions around the fission roof

(b) Around the “fission roof”, these nuclei are responsible for 30/5% of the fission recycling in scenario I/II.

Figure 15. (Color online) Post-neutron FFDs from SPY model computed at a constant excitation energy Q=8Q = 8 MeV (green curves) and from GEF model for thermal neutron-induced fission (blue curves). For each box, the x axis ranges from A=80A = 80 to 210 by step of 10 and the y axis from 0 to 15% by step of 2%. Colored background correspond to the contribution of the compound nucleus xCNx_{\mathrm{CN}} in scenario I/II. Only contributions xCN>0.001x_{\mathrm{CN}} > 0.001 are displayed. FFDs in thick orange boxes correspond to the NCN=184N_{\mathrm{CN}} = 184 neutron shell closure.

:::

Location in the (NN, ZZ) plane of the contributing fission seeds

Due to the specific β\beta-decay and neutron capture properties of nuclei above Th, in particular the “fission bottleneck”, the “fission roof” and the N=184N=184 neutron shell closure, not all the trans-Th nuclei play a role in fission recycling.

The contribution of a given fission seed area (see Figure 7d defining the 5 areas of relevance for the nuclear physics adopted in the present study) to the fission recycling Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} is the sum of xFS(Z,A)x_{\mathrm{FS}}(Z,A) for all nuclei located in such a specific fission seed area. Such contributions evolve with the fission recycling, as shown in Figure 12. The contribution of the “decay roof 0”, corresponding to the fission of nuclei originally accumulated along the neutron shell closure N=184N=184 and Z<94Z<94 and reaching the fission efficient region by β\beta-decay after the freeze-out, decreases with increasing Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} since the fission recycling increases with the initial neutron mass fraction Xn0X_n^0. With increasing Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} and Xn0X_n^0, matter accumulated along N=184N=184 isotone spreads up to Fm thanks to the successive β\beta-decays and neutron captures along this shell closure. More nuclei above Z=94Z=94 are produced and increase the contribution of the “decay roof 1” and consequently decrease the one of the “decay roof 0”. For Xfiss,totcum>10−6X_{\mathrm{fiss,tot}}^{\mathrm{cum}}>10^{-6}, the neutron shell closure N=184N=184 starts to be overcome and nuclei in the “fission bottleneck” area contribute to the fission recycling. In the range 10−6<Xfiss,totcum<10−110^{-6}<X_{\mathrm{fiss,tot}}^{\mathrm{cum}}<10^{-1}, the contribution of the “fission bottleneck” is roughly constant representing 20% of the total cumulative fissioning mass fraction. The matter starts to flow through the “fission bottleneck”, via Cm isotopes, and reach the “fission roof” for Xfiss,totcum>10−4X_{\mathrm{fiss,tot}}^{\mathrm{cum}}>10^{-4}. For high fission recycling (Xfiss,totcum>1X_{\mathrm{fiss,tot}}^{\mathrm{cum}}>1 as in scenario I), the total cumulative fissioning mass fraction is dominated by the “fission bottleneck” and “fission roof” areas, representing around 60% and 30%, respectively. The contribution of the “decay roof 2” is low because matter does not pile up in this region due to the absence of shell closure.

Repartition of the fission recycling between the 5 fission seed areas

Figure 12. (Color online) Repartition of the fission recycling Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} between the 5 fission seed areas defined in Figure 7d. The dashed line separates scenario I (Xfiss,totcum>1X_{\mathrm{fiss,tot}}^{\mathrm{cum}}>1) from scenario II (Xfiss,totcum<1X_{\mathrm{fiss,tot}}^{\mathrm{cum}}<1).

Scenario I

Sixteen fission seeds are found to contribute for more than 1% to the fission recycling (see Figure 13a) within scenario I. They are distributed mainly in the “fission bottleneck” and the “fission roof” areas, representing respectively 57% and 31% of the total cumulative fissioning mass fraction (see Figure 13a).

Fission seed contribution depending on the mass of fission seed for scenarios I and II

Figure 13. (Color online) Fission seed contribution (xFSx_{\mathrm{FS}}) depending on the mass of fission seed (the total height). Each bins contains the isotopic distribution of xFSx_{\mathrm{FS}} for a given AA.

A major part of the flow crossing the N=184N=184 shell closure is concentrated along the Am chain, fed by the favorable β\beta-decay of 94278Pu184{}^{278}_{94}\mathrm{Pu}_{184}, where 95285Am190{}^{285}_{95}\mathrm{Am}_{190} with xFS=45%x_{\mathrm{FS}}=45\% is the major contributor to the fission recycling. This favorable fission is due to the low fission barriers of a few nuclei dropping below 4 MeV according to the BSk14 Skyrme HFB calculations [111]. They are also main contributors in the “fission roof” area, the main one in this area and the second major one of all the contributors is 109344Mt233{}^{344}_{109}\mathrm{Mt}_{233} with xFS=12%x_{\mathrm{FS}}=12\%. In this case, the primary fission barrier drop below 2 MeV. As the main fission seeds are in the 275<A<290275<A<290 and 330<A<346330<A<346 ranges, the FFDs of these fission seeds affect the final production of nuclei in the 80<A<21080<A<210 range.

Scenario II

As the initial neutron mass fraction is low in this case, the fission recycling remains low too. The contribution of the “fission bottleneck” and the “fission roof” amount to 37% and 6%, respectively, to be compared with 57% and 31% in scenario I. The main fission seeds area is the “decay roof 1” (Figure 13b) because the low neutron irradiation makes it difficult to overcome the N=184N = 184 neutron shell closure. It represents 45% of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} and the main contributors are Md and No isotopes. The contribution of the “decay roof 0” increases from 2% in scenario I to the non-negligible value of 9% in scenario II.

The main fission seed is still 95285Am190{}^{285}_{95}\mathrm{Am}_{190} with xFS=25%x_{\mathrm{FS}} = 25\% with additional ones located in the 276<ACN<288276 < A_{\mathrm{CN}} < 288 mass region with xFS=2−10%x_{\mathrm{FS}} = 2 - 10\%. The fission recycling occurs mostly after freeze-out in scenario II. The post-freeze-out fission seeds, located in the “decay roof 0 and 1”, represent about 54% of Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}}.

Fission progenitors

We now turn in the analysis of the so-called fission progenitors, i.e. the nuclei for which fission contributes in a non-negligible way to their final abundance. Through a detailed tracking of the fission processes taking place during the r-process, the impact of fission on the final abundances can be quantified by the analysis of the fission fragments progenitors and their related fission seeds. For each trajectory, the progenitors of a given fission fragment can be identified and the fission contribution to the production of this final nucleus can be estimated. More precisely, the contribution stemming from the fission fragments xFFP(Z,A,ti)x_{\mathrm{FFP}}(Z,A,t_i) (FFP stands for fission fragments progenitors) to a nucleus (Z,A)(Z,A) with the molality Ym(Z,A,ti)Y_m(Z,A,t_i) (in mol/g) at time tit_i is defined as

xFFP(Z,A,ti)=xFFP(Z,A,ti−1)Ym(Z,A,ti−1)+ΔYm,PΔt(Z,A,ti−1)Δt∑rxFFP(Zr,Ar,ti−1)xRP(Z,A,r,ti−1)Ym(Z,A,ti−1)+ΔYm,PΔt(Z,A,ti−1)Δt(11)x_{\mathrm{FFP}}(Z,A,t_i)=\frac{x_{\mathrm{FFP}}(Z,A,t_{i-1})Y_m(Z,A,t_{i-1})+\frac{\Delta Y_{m,P}}{\Delta t}(Z,A,t_{i-1})\Delta t\sum_r x_{\mathrm{FFP}}(Z_r,A_r,t_{i-1})x_{\mathrm{RP}}(Z,A,r,t_{i-1})}{Y_m(Z,A,t_{i-1})+\frac{\Delta Y_{m,P}}{\Delta t}(Z,A,t_{i-1})\Delta t} \tag*{(11)}

where Δt=ti−ti−1\Delta t=t_i-t_{i-1} is an arbitrary time step taken here to be of the order of ≈ti/10\approx t_i/10 and ΔYm,PΔt(Z,A,ti−1)\frac{\Delta Y_{m,P}}{\Delta t}(Z,A,t_{i-1}) is the total production rate (in mol/g/s) of nucleus (Z,A)(Z,A) between ti−1t_{i-1} and tit_i, expressed in terms of the molality YmY_m (if normalized to 1, it corresponds to the molar fraction YY). The reaction fraction xRP(Z,A,r,ti−1)x_{\mathrm{RP}}(Z,A,r,t_{i-1}) is the fraction of the total production rate due to the reaction (r)(r). The reactions considered here are (n,γ)(n,\gamma), (γ,n)(\gamma,n), (β,γ)(\beta,\gamma), (β,n)(\beta,n), (β,2n)(\beta,2n), (β,3n)(\beta,3n), (γ,α)(\gamma,\alpha), (sf) (i.e. spontaneous fission), (β,f)(\beta,f) and (n,f)(n,f). In other words, the contribution of fission fragments to the production of a nucleus ZAX{}^{A}_{Z}\mathrm{X} at time tit_i is the weighted average between the previous contribution xFFP(Z,A,ti−1)x_{\mathrm{FFP}}(Z,A,t_{i-1}) and the contributions xFFP(Zr,Ar,ti−1)x_{\mathrm{FFP}}(Z_r,A_r,t_{i-1}) due to the production of this element by a given reaction rr: ZrArY→ZAX+…{}^{A_r}_{Z_r}\mathrm{Y}\rightarrow{}^{A}_{Z}\mathrm{X}+\ldots, between ti−1t_{i-1} and tit_i, at a rate of ΔYm,PΔt(Z,A,ti−1)xRP(Z,A,r,ti−1)\frac{\Delta Y_{m,P}}{\Delta t}(Z,A,t_{i-1})x_{\mathrm{RP}}(Z,A,r,t_{i-1}). If a nucleus is exclusively produced by fission, xFFP=1x_{\mathrm{FFP}}=1.

This quantity is well suited to quantify the impact of fission fragments on final abundances because it is not affected by the decay of post-neutron fission fragments thanks to the time tracking. To obtain the overall final xFFPx_{\mathrm{FFP}}, hereafter xFFPfx_{\mathrm{FFP}}^{f}, (Figure 16, red dotted curves), the tracking (Eq. (11)) is performed up to t=1t=1 yr when the residual fission does not contribute anymore to the production of light nuclei.

Fission fragment distributions and their impact on the r-process nucleosynthesis in neutron star mergers

Figure 16. (Color online) (α\alpha and γ\gamma) Contribution of the fission fragments to the final production (xFFPfx_{\mathrm{FFP}}^{\mathrm{f}}) (red dotted curves) and fission seeds contributions to the production of fission fragment progenitors (xFS/FFPfx_{\mathrm{FS/FFP}}^{\mathrm{f}}) (color scale) computed with SPY(α\alpha) or GEF (γ\gamma) FFDs. (β\beta and δ\delta) Mean fission yields ⟨YFF⟩FS\langle Y_{\mathrm{FF}}\rangle_{\mathrm{FS}} for SPY (β\beta) or GEF model (δ\delta) averaged over fission seeds from the “fission bottleneck” and the “fission roof” (green curves, denoted as pre-freeze-out) or from “decay roofs” (orange curves, denoted as post-freeze-out).

The contribution of fission fragments to the final production depends on the fission recycling indicator Xfiss,totcumX_{\mathrm{fiss,tot}}^{\mathrm{cum}} (Figure 14, gray curve) and starts to be significant, i.e.

Contribution of the fission fragments to the final production of nuclei with mass number $A$

Figure 14. (Color online) Contribution of the fission fragments to the final production of nuclei with mass number AA xFFP(A,Xfiss,totcum)x_{\mathrm{FFP}}(A,X_{\mathrm{fiss,tot}}^{\mathrm{cum}}) obtained with SPY FFDs. Gray curve delimits xFFPf(A,Xfiss,totcum)=0.1x_{\mathrm{FFP}}^{f}(A,X_{\mathrm{fiss,tot}}^{\mathrm{cum}})=0.1 and white one xFFPf(A,Xfiss,totcum)=0.5x_{\mathrm{FFP}}^{f}(A,X_{\mathrm{fiss,tot}}^{\mathrm{cum}})=0.5.

max⁡A[xFFPf(A,Xfiss,totcum)](Xfiss,totcum)>0.1,\max_A\left[x_{\mathrm{FFP}}^{f}\left(A,X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\right)\right]\left(X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\right)>0.1,

for Xfiss,totcum>10−4X_{\mathrm{fiss,tot}}^{\mathrm{cum}}>10^{-4} (X0n>0.65X_0^n>0.65) which corresponds to the value given in Sec. III.2.2. For Xfiss,totcum>10−2X_{\mathrm{fiss,tot}}^{\mathrm{cum}}>10^{-2}, fission contributes at least to 50% of the production of the final nuclei with 108<A<120108<A<120 and 139<A<155139<A<155. For a high fission recycling (Xfiss,totcum≈1X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\approx1), this contribution can reach nearly 100% for the elements with A≈110A\approx110.

In addition to the strong sensitivity to the fission recycling, xFFPf(A)x_{\mathrm{FFP}}^{f}(A) is obviously affected by the FFDs model adopted, even if the fission recycling does not depend significantly on the FFDs model. This sensitivity to the FFDs model is studied for both scenario in the two subsections below. In order to clarify the link between xFSx_{\mathrm{FS}} and xFFPfx_{\mathrm{FFP}}^{f}, we also introduce here two new quantities, namely the relative contribution of each fission seed to the production of a final nucleus AA by fission

xFS/FFPf(A,AFS)=xFFPf(A,AFS)xFFPf(A)(12)x_{\mathrm{FS/FFP}}^{f}(A,A_{\mathrm{FS}})=\frac{x_{\mathrm{FFP}}^{f}(A,A_{\mathrm{FS}})}{x_{\mathrm{FFP}}^{f}(A)} \tag*{(12)}

where xFFPf(A,AFS)x_{\mathrm{FFP}}^{f}(A,A_{\mathrm{FS}}) is obtained by tracking only fission fragments produced by the specific fission seed AFSA_{\mathrm{FS}}, xRP(Z,A,ti−1)x_{\mathrm{RP}}(Z,A,t_{i-1}) being replaced by xRP(Z,A,AFS,ti−1)x_{\mathrm{RP}}(Z,A,A_{\mathrm{FS}},t_{i-1}) in Eq. (11). and the contribution of a compound nucleus (ZCN,ACN)(Z_{\mathrm{CN},} A_{\mathrm{CN}}) to the fission recycling, which can be expressed from Eq. (10) as

xCN(ZCN,ACN)=∑(Z,A)CN,r={sf,βdf,nif}xFS(Zr,Ar,r)(13)x_{\mathrm{CN}}(Z_{\mathrm{CN}},A_{\mathrm{CN}})=\sum_{(Z,A)_{\mathrm{CN}},r=\{\mathrm{sf},\beta\mathrm{df},\mathrm{nif}\}}x_{\mathrm{FS}}(Z_r,A_r,r) \tag*{(13)}

where xFS(Zr,Ar,r)x_{\mathrm{FS}}(Z_r,A_r,r) is the contribution to the fission recycling ⟨Xfiss,totcum⟩traj\langle X_{\mathrm{fiss,tot}}^{\mathrm{cum}}\rangle_{\mathrm{traj}} of the fission seed (Zr,Ar)(Z_r,A_r) fissioning by a given fission mode rr (spontaneous, β\beta-delayed or neutron induced). In Eq. (13), the summation is performed in such a way that the mass of the compound nucleus formed from the fission seed is equal to ACNA_{\mathrm{CN}} and the proton number to ZCNZ_{\mathrm{CN}}.

Scenario I

In scenario I, fission processes contribute significantly to the production of nuclei with 80<A<20080 < A < 200 for both FFDs models (Figure 16a, red dotted curves) due to the high fission recycling. The shape of xFFPf(A)x_{\mathrm{FFP}}^{f}(A) is affected by the FFDs model (Figure 16a$\alpha$ and $\gamma$, red dotted curves). However the link between xFFPfx_{\mathrm{FFP}}^{f} and FFDs is no straightforward.

Two mean fission yields distributions ⟨YFF⟩FS\langle Y_{\mathrm{FF}}\rangle_{\mathrm{FS}} can be defined, the first one is associated with the “fission bottleneck” and the “fission roof”, denoted as pre-freeze-out mean fission yields, corresponds to the mean fission yields distribution during the neutron irradiation, before freeze-out (Figure 16, green curves). The second one, denoted as post-freeze-out mean fission yields, is associated with the “decay roofs” (Figure 16, orange curves), after freeze-out.

For both FFDs model, the pre-freeze-out mean fission yields is dominated by the FFDs of 286Am{}^{286}\mathrm{Am} (xCN=0.45x_{\mathrm{CN}}=0.45) and 345Mt{}^{345}\mathrm{Mt} (xCN=0.12x_{\mathrm{CN}}=0.12) formed by the neutron capture of 285Am{}^{285}\mathrm{Am} and 344Mt{}^{344}\mathrm{Mt}. With SPY FFDs (see Figure 15, green curves), these two main contributors fission asymmetrically. The three peaks pattern seen in the pre-freeze-out ⟨YFF⟩FS\langle Y_{\mathrm{FF}}\rangle_{\mathrm{FS}} (Figure 16a$\beta$) is due to (i) the light fragments of 286Am{}^{286}\mathrm{Am} feeding the peak around A=120−130A=120-130, (ii) the heavy fragments of 286Am{}^{286}\mathrm{Am} plus the light fragments of 345Mt{}^{345}\mathrm{Mt} feeding the peak around A=145−160A=145-160, and (iii) the heavy fragments of 346Mt{}^{346}\mathrm{Mt} responsible for the peak A=180−190A=180-190. In contrast, the GEF model (see Figure 15, blue curves) predicts a more asymmetric fission of 286Am{}^{286}\mathrm{Am} but a more symmetric one for 346Mt{}^{346}\mathrm{Mt}. As a result, a double hump pattern is obtained for the pre-freeze-out ⟨YFF⟩FS\langle Y_{\mathrm{FF}}\rangle_{\mathrm{FS}} (Figure 16a$\delta$) where the peak around A=100−120A=100-120 is due to the light fragments of 286Am{}^{286}\mathrm{Am} and the one around A=150−180A=150-180 to the heavy fragments of 286Am{}^{286}\mathrm{Am} plus the symmetric fragments of 346Mt{}^{346}\mathrm{Mt}.

The main contributors to the post-freeze-out mean fission yields, i.e. to the fission recycling in the “decay roofs” areas, are isotopes from Fm to Lr with ACN=278A_{\mathrm{CN}}=278 to 284 where xCNx_{\mathrm{CN}} reaches values up to 0.016 for 283No{}^{283}\mathrm{No} (see Figure 15a). Those nuclei are characterized by an admixture of slightly asymmetric fission and a triple-peak fission where the maximum yields are found around A=130A=130 for SPY FFDs. GEF FFD is more asymmetric leading to post-freeze-out mean fission yields roughly constant in the A=110−160A=110-160 mass range.

During the neutron irradiation, fission fragments captures neutrons. The A<130A<130 fission fragments accumulate along the neutron shell closure N=82N=82 and the relative contribution from the “fission bottleneck” xFS/FFPfx_{\mathrm{FS/FFP}}^{f} increases at A=120−130A=120-130. The fission fragments produced above N=82N=82 (A≳130A\gtrsim130) spread out up to N=126N=126 (A≲195A\lesssim195). For A>160A>160, the fission contribution to the final production xFFPfx_{\mathrm{FFP}}^{f} and the relative contribution of the “fission bottleneck” xFS/FFPfx_{\mathrm{FS/FFP}}^{f} is higher with GEF than with SPY since the pre-freeze-out distribution is more asymmetric due to the more asymmetric FFD of 246Am{}^{246}\mathrm{Am}. At freeze-out, there is no more fission fragments produced by the “fission bottleneck” and the “fission roof”. The final abundance peak A=130A=130 predicted by SPY model (Figure 8) is due to the high post-freeze-out mean fission yields around A=130A=130 (Figure 16a, orange curves). xFFPfx_{\mathrm{FFP}}^{f} is locally reduced around A=130A=130 due to the accumulation of non-recycled material along the neutron shell closure N=82N=82. This reduction is limited with SPY model since the pre- and post-freeze-out mean fission yields are higher than in the GEF case around A=130A=130. A low xFFPfx_{\mathrm{FFP}}^{f} for 80<A<10080<A<100 could be expected since the mean fission yields are negligible for both FFD models. However, in this case, there is no matter accumulation in this mass region, so that the low 80<A<10080<A<100 production (Yf<10−4Y^{f}<10^{-4}) occurring after freeze-out is essentially due to the asymmetric fission of seeds located in the “decay roof”. This leads to a non-negligible value of xFFPfx_{\mathrm{FFP}}^{f} for 80<A<10080<A<100. The drop of xFFPfx_{\mathrm{FFP}}^{f} for A>180A>180 is due to the accumulation of matter along the neutron shell closure N=126N=126.

Scenario II

In scenario II, the reduced fission recycling leads to a non-negligible xFFPfx_{\mathrm{FFP}}^{f} that does not exceed 30% only around 130<A<185130<A<185, irrespective of the FFDs model adopted (Figure 16b, red dotted curves).

For both FFDs models, the post-freeze-out mean fission yields is roughly similar than the one within scenario I since the main contributors from the “decay roof” regions are the same in both cases (see Figure 15a) where xCNx_{\mathrm{CN}} reaches values up to 0.08 for 282No{}^{282}\mathrm{No}. The fission seed contribution from the “fission roof” is small, so that the A=180−190A=180-190 pre-freeze-out mean fission yields are negligible with the SPY model. The GEF pre-freeze-out mean fission yields is roughly similar to the one found in scenario I, though the A=100−120A=100-120 peak is now found higher than the A=160−180A=160-180 one.

The low neutron irradiation makes the contribution of the “fission bottleneck” and the “fission roof” relatively small within this scenario, leading to a total

xFS<0.25x_{\mathrm{FS}} < 0.25 for 90% the ejected mass (i.e. Xfiss,totcum<0.1X_{\mathrm{fiss,tot}}^{\mathrm{cum}} < 0.1) (Figure 12). The role played by the pre-freeze-out mean fission yields is consequently reduced. It remains more difficult to overcome the neutron shell closure N=82N = 82 and more non-recycled material accumulates along this shell closure. For this reason, the final abundance around A=130A = 130 is higher than in scenario I with a negligible fission contribution to the A≤130A \le130 abundances. As the SPY post-freeze-out mean fission yields is peaked around A=120A = 120–160160, xFFPfx_{\mathrm{FFP}}^{\mathrm{f}} is maximal for A=135A = 135–160160. Compared to SPY, GEF pre- and post-freeze-out mean fission yields are higher for A≳160A \gtrsim160 which leads to a higher xFFPfx_{\mathrm{FFP}}^{\mathrm{f}} in this mass region.

Conclusion

In the present study, new improvements have been brought to our scission-point model, called SPY, to derive the fission fragment distribution for all neutron-rich fissioning nuclei of relevance in r-process calculations. These improvements include a phenomenological modification of the scission distance and an updated smoothing procedure of the distribution to take into account effects that are neglected in the original model. Such corrections lead to a better agreement with experimental fission yields. In particular, SPY FFDs are now in overall good agreement with the experimental symmetric/asymmetric transition in the Th and Pa isotopic chains. Those yields at scission are also used to estimate the number of neutrons emitted by the excited fragments on the basis of different neutron evaporation models. Our new fission yields significantly differ from to those predicted by the widely used GEF model.

The impact of fission on the r-process nucleosynthesis in neutron mergers is also re-analyzed. Two scenarios are considered, the first one with low initial electron fraction is subject to intense fission recycling, in contrast to the second one which includes weak interactions on free nucleons resulting in less pronounced fission recycling. The efficiency of the r-process nucleosynthesis and the corresponding abundance distribution in a given mass element has been discussed essentially in terms of the initial neutron mass fraction Xn0X_{n}^{0} which engulfs information on both the initial electron fraction and entropy. The contribution fission processes may have to the final abundance distribution has been studied in detail in the light of newly defined quantitative indicators, namely indicators for the fission recycling (Xfiss,totX_{\mathrm{fiss,tot}}), fission seeds (xFSx_{\mathrm{FS}}) and fission fragment progenitors (xFFPx_{\mathrm{FFP}}). The various regions of the nuclear chart responsible for fission recycling during the neutron irradiation as well as after freeze-out have been identified for the nuclear physics adopted in the present calculations, what we defined as the “decay roofs”, the “fission bottleneck” and the “fission roof”. Within the “fission bottleneck” region, Am isotopes around A=285A = 285 play a key role by limiting the nuclear flow to higher elements and strongly contributing to the fission recycling, mainly due to the low fission barriers predicted by the BSk14 interaction in this region. Above Z=110Z = 110, the so-called “fission roof” prohibits the production of super-heavy elements. “Decay roofs” essentially recycle heavy material after the neutron irradiation freezes out when the material accumulated at the N=184N = 184 shell closure decays back to the valley of β\beta-stability.

When considering an efficient r-process (scenario I), fission fragments are found to contribute almost entirely to the final abundances of nuclei with 100≲A≲180100 \lesssim A \lesssim180, while in scenario II where weak interactions on nucleons initially limit the number of free neutrons, such a contribution does not reach more than 30% and is concentrated in the A=140A = 140–180180 region depending on the FFD model adopted. Calculations based on the SPY and GEF FFDs have been compared for both r-process scenarios.

The present study allows us to have a deep insight on the role of fission processes in hydrodynamical r-process simulations on the basis of sound well-defined quantitative indicators. Fission remains an important nuclear process taking place in binary NS mergers, though major nuclear as well as astrophysical uncertainties still need to be further addressed before concluding on the definite role of fission processes. These concern in particular the determination of the fission probabilities for all the nuclei potentially involved during the r-process as well as a detailed description of neutrino absorption and hydrodynamics in simulations. Much progress remains to be done along these two directions.

Acknowledgments

J.-F.L. and S.G. acknowledge financial support from FNRS (Belgium). This work was supported by the Fonds de la Recherche Scientifique (F.R.S.-FNRS) and the Fonds Wetenschappelijk Onderzoek - Vlaanderen (FWO) under the EOS Project nr 0022818F. The present research benefited from computational resources made available on the Tier-1 supercomputer of the Fédération Wallonie-Bruxelles, infrastructure funded by the Walloon Region under the grant agreement nr 1117545. H.-T.J. acknowledges financial support from Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) through Sonderforschungsbereich (Collaborative Research Center) SFB-1258 “Neutrinos and Dark Matter in Astro- and Particle Physics (NDM)” and under Germany’s Excellence Strategy through Cluster of Excellence ORIGINS (EXC-2094)—390783311, and from the European Research Council through Grant ERC-AdG No. 341157-COCO2CASA. A.B. acknowledges support by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 759253, and by Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - Project-ID 279384907 - SFB 1245 and - Project-ID 138713538 - SFB 881 (“The Milky Way System”, subproject A10).

References

References

  1. [1]M. Arnould, S. Goriely, and K. Takahashi, Phys. Repts. 450, 97 (2007).
  2. [3]J. J. Cowan, C. Sneden, J. E. Lawler, A. Aprahamian, M. Wiescher, K. Langanke, G. Martínez-Pinedo, and F.-K. Thielemann, Rev. Mod. Phys. accepted (2020).
  3. [4]S. Wanajo, H.-T. Janka, and B. Müller, Astrophys. J. 726, L15 (2011).
  4. [5]H.-T. Janka, Ann. Rev. of Nucl. Part. Sci. 62, 407 (2012).
  5. [6]L. Hüdepohl, B. Müller, H.-T. Janka, A. Marek, and G. G. Raffelt, Phys. Rev. Lett. 104, 251101 (2010).
  6. [7]L. F. Roberts, S. E. Woosley, and R. D. Hoffman, The Astrophysical Journal 722, 954 (2010).
  7. [8]T. Fischer, S. C. Whitehouse, A. Mezzacappa, F. K. Thielemann, and M. Liebendörfer, Astronomy and Astrophysics 517, A80 (2010), arXiv:0908.1871 [astro-ph.HE].DOI
  8. [9]D. Siegel, J. Barnes, and B. Metzger, Nature 569, 241 (2019).
  9. [10]C. Winteler, R. Käppeli, A. Perego, A. Arcones, N. Vasset, N. Nishimura, M. Liebendörfer, and F.-K. Thielemann, The Astrophysical Journal 750, L22 (2012).
  10. [11]C. Freiburghaus, S. Rosswog, and F. K. Thielemann, J. Lett. 525, L121 (1999).
  11. [12]M. Ruffert and H.-T. Janka, Astronomy and Astrophysics 380, 544 (2001).
  12. [13]H.-T. Janka, T. Eberl, M. Ruffert, and C. Fryer, Astrophys. J. Lett. 527, L39 (1999).
  13. [14]O. Korobkin, S. Rosswog, A. Arcones, and C. Winteler, MNRAS 26, 1940 (2012).
  14. [15]L. Roberts, D. Kasen, W. Lee, and E. Ramirez-Ruiz, Astrophys. J. Lett. 736, L21 (2011).
  15. [16]R. Oechslin, H.-T. Janka, and A. Marek, Astronomy and Astrophysics 467, 395 (2007).
  16. [17]S. Goriely, A. Bauswein, and H.-T. Janka, Astrophys. J. Lett. 738, L32 (2011).
  17. [18]O. Just, A. Bauswein, R. Ardevol-Pulpillo, S. Goriely, and H.-T. Janka, MNRAS 448, 541 (2015).
  18. [19]K. Kyutoku, K. Ioka, and M. Shibata, Phys. Rev. D 88, 041503(R) (2013).
  19. [20]K. Hotokezaka, K. Kyutoku, M. Tanaka, K. Kiuchi, Y. Sekiguchi, M. Shibata, and S. Wanajo, Astrophys. J. Lett. 778, L16 (2013).
  20. [21]A. Bauswein, R. Ardevol Pulpillo, H.-T. Janka, and S. Goriely, Astrophysical J. Lett 773, L9 (2014).
  21. [22]S. Wanajo, Y. Sekiguchi, N. Nishimura, K. Kiuchi, K. Kyutoku, and M. Shibata, Astrophys. J. Lett 789, L39 (2014).
  22. [23]F. Foucart, M. B. Deaton, M. D. Duez, E. O’Connor, C. D. Ott, R. Haas, L. E. Kidder, H. P. Pfeiffer, M. A. Scheel, and B. Szilagyi, Phys. Rev. D 90, 024026 (2014).
  23. [24]T. Baumgarte, S. Shapiro, and M. Shibata, Astrophys. J. 528, L29 (2000).
  24. [25]M. Ruffert and H.-T. Janka, Astronomy and Astrophysics 344, 573 (1999).
  25. [26]L. Dessart, C. Ott, A. Burrows, S. Rosswog, and E. Livne, Astrophys. J. 690, 1681 (2009).
  26. [27]R. Fernandez and B. Metzger, MNRAS 435, 502 (2013).
  27. [28]D. Siegel, R. Ciolfi, and L. Rezzolla, Astrophys. J. Lett. 785, L6 (2014).
  28. [29]B. Metzger and R. Fernandez, MNRAS 441, 3444 (2014).
  29. [30]A. Perego, S. Rosswog, R. Cabezon, O. Korobkin, R. Kappeli, A. Arcones, and M. Liebendorfer, MNRAS 443, 3134 (2014).
  30. [31]A. Bauswein, S. Goriely, and H.-T. Janka, Astrophysical J. 773, 78 (2013).
  31. [32]S. Goriely, The European Physical Journal A 51, 22 (2015).
  32. [33]M. Dominik, K. Belczynski, C. Fryer, D. Holz, E. Berti, T. Bulik, I. Mandel, and R. O’Shaughnessy, Astrophys. J. 759, 52 (2012).
  33. [34]Y. Z. Qian, Astrophys. J. Lett. 534, L67 (2000), arXiv:astro-ph/0003242 [astro-ph].DOI
  34. [35]B. D. Metzger, G. Martinez-Pinedo, S. Darbha, E. Qataert, A. Arcones, D. Kasen, R. Thomas, P. Nugent, I. V. Panov, and N. T. Zinner, MNRAS 406, 2650 (2010), arXiv:1001.5029 [astro-ph.HE].DOI
  35. [36]B. Abbott, R. Abbott, T. Abbott, et al., Phys. Rev. Lett. 119, 161101 (2017).
  36. [37]B. Abbott, R. Abbott, T. Abbott, et al., Astrophys. J. Lett. 848, L12 (2017).
  37. [38]P. S. Cowperthwaite, E. Berger, V. A. Villar, B. D. Metzger, M. Nicholl, et al., Astrophys. J. Lett. 848, L17 (2017).
  38. [39]M. Nicholl, E. Berger, D. Kasen, B. D. Metzger, J. Elias, et al., Astrophys. J. Lett. 848, L18 (2017).
  39. [40]R. Chornock, E. Berger, D. Kasen, P. S. Cowperthwaite, M. Nicholl, et al., Astrophys. J. Lett. 848, L19 (2017).
  40. [41]E. Waxman, E. O. Ofek, D. Kushnir, and A. Gal-Yam, MNRAS 481, 3423 (2018).
  41. [42]D. Watson, C. J. Hansen, J. Selsing, A. Koch, D. B. Malesani, A. C. Andersen, J. P. U. Fynbo, A. Arcones, A. Bauswein, S. Covino, A. Grado, K. E. Heintz, L. Hunt, C. Kouveliotou, G. Leloudas, A. J. Levan, P. Mazzali, and E. Pian, Nature 574, 497 (2019).
  42. [43]I. Y. Korneev and I. V. Panov, Astronomy Letters 37, 864 (2011).
  43. [44]S. Goriely, J.-L. Sida, J.-F. Lemaître, S. Panebianco, N. Dubray, S. Hilaire, A. Bauswein, and H.-T. Janka, Phys. Rev. Lett. 111, 242502 (2013).
  44. [45]M. Eichler, A. Arcones, A. Kelic, O. Korobkin, K. Langanke, T. Marketin, G. Martinez-Pinedo, I. Panov, T. Rauscher, S. Rosswog, C. Winteler, N. T. Zinner, and F.-K. Thielemann, The Astrophysical Journal 808, 30 (2015).
  45. [46]J. d. J. Mendoza-Temis, M.-R. Wu, K. Langanke, G. Martínez-Pinedo, A. Bauswein, and H.-T. Janka, Phys. Rev. C 92, 055805 (2015), arXiv:1409.6135 [astro-ph.HE].
  46. [47]S. A. Giuliani, G. Martínez-Pinedo, and L. M. Robledo, Phys. Rev. C 97, 034323 (2018), arXiv:1704.00554 [nuc l-th].DOI
  47. [48]M. R. Mumpower, T. Kawano, T. M. Sprouse, N. Vassh, E. M. Holmbeck, R. Surman, and P. Möller, The Astrophysical Journal 869, 14 (2018).
  48. [49]Y. Zhu, R. T. Wollaeger, N. Vassh, R. Surman, T. M. Sprouse, M. R. Mumpower, N. Vassh, P. Jaffke, E. M. Holmbeck, C. L. Fryer, W. P. Even, A. J. Couture, and J. Barnes, The Astrophysical Journal 863, L23 (2018).
  49. [50]E. M. Holmbeck, T. M. Sprouse, M. R. Mumpower, N. Vassh, R. Surman, T. C. Beers, and T. Kawano, The Astrophysical Journal 870, 23 (2018).
  50. [51]M.-R. Wu, J. Barnes, G. Martínez-Pinedo, and B. D. Metzger, Phys. Rev. Lett. 122, 062701 (2019), arXiv:1808.10459 [astro-ph.HE].DOI
  51. [52]N. Vassh, R. Vogt, R. Surman, J. Randrup, T. M. Sprouse, M. R. Mumpower, P. Jaffke, D. Shaw, E. M. Holmbeck, Y. Zhu, and G. C. McLaughlin, Journal of Physics G: Nuclear and Particle Physics 46, 065202 (2019).
  52. [53]J.-F. Lemaître, S. Goriely, S. Hilaire, and J.-L. Sida, Phys. Rev. C 99, 034612 (2019).
  53. [54]K.-H. Schmidt, B. Jurado, C. Amouroux, and C. Schmitt, Nuclear Data Sheets 131, 107 (2016), special Issue on Nuclear Reaction Data.
  54. [55]Y. Sekiguchi, K. Kiuchi, K. Kyutoku, and M. Shibata, Phys. Rev. D 91, 064059 (2015), arXiv:1502.06660 [astro-ph.HE].DOI
  55. [56]S. Goriely, A. Bauswein, O. Just, E. Plumbbi, and H.-T. Janka, MNRAS 452, 3894 (2015).
  56. [57]F. Foucart, R. Haas, M. D. Duez, E. O’Connor, C. D. Ott, L. Roberts, L. E. Kidder, J. Lippuner, H. P. Pfeiffer, and M. A. Scheel, Phys. Rev. D 93, 044019 (2016), arXiv:1510.06398 [astro-ph.HE].DOI
  57. [58]L. Lehner, S. L. Liebling, C. Palenzuela, O. L. Caballero, E. O’Connor, M. Anderson, and D. Neilsen, Classical and Quantum Gravity 33, 184002 (2016), arXiv:1603.00501 [gr-qc].DOI
  58. [59]D. Radice, F. Galeazzi, J. Lippuner, L. F. Roberts, C. D. Ott, and L. Rezzolla, MNRAS 460, 3255 (2016), arXiv:1601.02426 [astro-ph.HE].DOI
  59. [60]L. Bovard, D. Martin, F. Guercilena, A. Arcones, L. Rezzolla, and O. Korobkin, Phys. Rev. D 96, 124005 (2017), arXiv:1709.09630 [gr-qc].DOI
  60. [61]D. Martin, A. Perego, W. Kastaun, and A. Arcones, Classical and Quantum Gravity 35, 034001 (2018), arXiv:1710.04900 [astro-ph.HE].DOI
  61. [62]S. Goriely, Il Nuovo Cimento C 42, 115 (2019).
  62. [63]R. Ardevol-Pulpillo, H.-T. Janka, O. Just, and A. Bauswein, MNRAS 485, 4754 (2019).
  63. [64]J.-F. Lemaître, S. Panebianco, J.-L. Sida, S. Hilaire, and S. Heinrich, Phys. Rev. C 92, 034617 (2015).
  64. [65]S. Goriely, N. Chamel, and J. M. Pearson, Phys. Rev. C 88, 061302(R) (2013).
  65. [66]P. Decowski, W. Grochulski, A. Marcinkowski, K. Siwek, and Z. Wilhelmi, Nucl. Phys. A 110, 129 (1968).
  66. [67]L. Moretto, Physics Letters B 40, 1 (1972).
  67. [68]L. Moretto, Nucl. Phys. A 185, 145 (1972).
  68. [69]P. Demetriou and S. Goriely, Nucl. Phys. A 695, 95 (2001).
  69. [70]D. Madland and J. Nix, Nucl. Phys. A 476, 1 (1988).
  70. [71]C. Schmitt, A. Guessous, J. Bocquet, H.-G. Clerc, R. Brissot, D. Engelhardt, H. Faust, F. Gönnenwein, M. Mutterer, H. Nifennecker, J. Pannicke, C. Ristori, and J. Theobald, Nuclear Physics A 430, 21 (1984).
  71. [72]C. Romano, Y. Danon, R. Block, J. Thompson, E. Blain, and E. Bond, Phys. Rev. C 81, 014607 (2010).
  72. [73]G. Mariolopoulos, C. Hamelin, J. Blachot, J. Bocquet, R. Brissot, J. Crençon, H. Nifenfecker, and C. Ristori, Nuclear Physics A 361, 213 (1981).
  73. [74]C. Tsuchiya, Y. Nakagome, H. Yamana, H. Moroyama, K. Nishio, I. Kanno, K. Shin, and I. Kimura, Journal of Nuclear Science and Technology 37, 941 (2000).
  74. [75]J. Bocquet and R. Brissot, Nuclear Physics A 502, 213 (1989).
  75. [76]S. Zeynalov, F.-J. Hambsch, and S. Oberstedt, J. Korean Phy. Soc. 59, 1396 (2011).
  76. [77]Vorobyev, A.S., Shcherbakov, O.A., Gagarski, A.M., Val’ski, G.V., and Petrov, G.A., EPJ Web of Conferences 8, 03004 (2010).
  77. [78]K. Nishio, Y. Nakagome, I. Kanno, and I. Kimura, Journal of Nuclear Science and Technology 32, 404 (1995).
  78. [79]V. Dushin, F.-J. Hambsch, V. Jakovlev, V. Kalinin, I. Kraev, A. Laptev, D. Nikolaev, B. Petrov, G. Petrov, V. Petrova, Y. Pleva, O. Shcherbakov, V. Shpakov, V. Sokolov, A. Vorobyev, and T. Zavarukhina, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 516, 539 (2004).
  79. [80]O. Litaize, O. Serot, and L. Berge, The European Physical Journal A 51, 177 (2015).
  80. [81]P. Talou, B. Becker, T. Kawano, M. B. Chadwick, and Y. Danon, Phys. Rev. C 83, 064612 (2011).
  81. [82]R. Vogt and J. Randrup, Physics Procedia 47, 82 (2013), scientific Workshop on Nuclear Fission Dynamics and the Emission of Prompt Neutrons and Gamma Rays, Biarritz, France, 28-30 November 2012.
  82. [83]M. Chadwick, P. Obložinský, M. Herman, N. Greene, R. McKnight, D. Smith, P. Young, R. MacFarlane, G. Hale, S. Frankle, A. Kahler, T. Kawano, R. Little, D. Madland, P. Moller, R. Mosteller, P. Page, P. Talou, H. Trellue, M. White, W. Wilson, R. Arcilla, C. Dunford, S. Mughabghab, B. Pritychenko, D. Rochman, A. Sonzogni, C. Lubitz, T. Trumbull, J. Weinman, D. Brown, D. Cullen, D. Heinrichs, D. McNabb, H. Derrien, M. Dunn, N. Larson, L. Leal, A. Carlson, R. Block, J. Briggs, E. Cheng, H. Huria, M. Zerkle, K. Kozier, A. Courcelle, V. Pronyaev, and S. van der Marck, Nuclear Data Sheets 107, 2931 (2006), evaluated Nuclear Data File ENDF/B-VII.0.
  83. [84]T. Kawano, P. Talou, I. Stetcu, and M. Chadwick, Nuclear Physics A 913, 51 (2013).
  84. [85]V. Weisskopf, Phys. Rev. 52, 295 (1937).
  85. [86]V. F. Weisskopf and D. H. Ewing, Phys. Rev. 57, 472 (1940).
  86. [87]H. A. Bethe, Rev. Mod. Phys. 9, 69 (1937).
  87. [88]J. Tőke and W. Świątecki, Nuclear Physics A 372, 141 (1981).
  88. [89]A. Koning and D. Rochman, Nuclear Data Sheets 113, 2841 (2012), special Issue on Nuclear Reaction Data.
  89. [90]K.-H. Schmidt, S. Steinhäuser, C. Böckstiegel, A. Grewe, A. Heinz, A. Junghans, J. Benlliure, H.-G. Clerc, M. de Jong, J. Müller, M. Pfützner, and B. Voss, Nuclear Physics A 665, 221 (2000).
  90. [91]H. Baba, T. Saito, N. Takahashi, A. Yokoyama, T. Miyauchi, S. Mori, D. Yano, T. Hakoda, K. Takamiya, K. Nakanishi, and Y. Nakagome, Journal of Nuclear Science and Technology 34, 871 (1997).
  91. [92]D. Ramos, M. Caamaño, F. Farget, C. Rodríguez-Tajes, L. Audouin, J. Benlliure, E. Casarejos, E. Clement, D. Cortina, O. Delaune, X. Derkx, A. Dijon, D. Doré, B. Fernández-Domínguez, G. de France, A. Heinz, B. Jacquot, A. Navin, C. Paradela, M. Rejmund, T. Roger, M.-D. Salsac, and C. Schmitt, Phys. Rev. C 97, 054612 (2018).
  92. [93]F.-J. Hambsch, F. Vivès, P. Siegler, and S. Oberstedt, Nuclear Physics A 679, 3 (2000).
  93. [94]H. Thierens, A. De Clercq, E. Jacobs, D. De Frenne, P. D’hondt, P. De Gelder, and A. J. Deruytter, Phys. Rev. C 23, 2104 (1981).
  94. [95]V. Rubchenya, W. Trzaska, I. Itkis, M. Itkis, J. Kliman, G. Kniajeva, N. Kondratiev, E. Kozulin, L. Krupa, I. Pokrovski, V. Voskressenski, F. Hanappe, T. Materna, O. Dervaux, L. Stutte, G. Chubarian, S. Khelbnikov, D. Vakhtin, and V. Lyapin, Nuclear Physics A 734, 253 (2004).
  95. [96]A. Goverdovskij, M. Mitrofanov, and V. Khryachkov, Yadernaya Fizika 58, 1948 (1995).
  96. [97]Y. L. Zhao, I. Nishinaka, Y. Nagame, K. Tsukada, K. Sueki, S. Goto, M. Tanikawa, and H. Nakahara, Journal of Radioanalytical and Nuclear Chemistry 255, 67 (2003).
  97. [98]J. Unik, J. Gindler, L. Glendenin, K. Flyn, A. Gorski, and R. Sjoblom, Proc. Symp. on physics and chemistry of fission, Rochester 2, 19 (1974).
  98. [99]D. C. Hoffman, J. B. Wilhelmy, J. Weber, W. R. Daniels, E. K. Hulet, R. W. Lougheed, J. H. Landrum, J. F. Wild, and R. J. Dupzyk, Phys. Rev. C 21, 972 (1980).
  99. [100]D. Hoffman, D. Lee, A. Ghiorso, M. Nurmia, and K. Aleklett, Phys. Rev. C 22, 1581 (1980).
  100. [101]T. M. Hamilton, K. E. Gregorich, D. M. Lee, K. R. Czerwinski, N. J. Hannik, C. D. Kacher, B. Kadkhodayan, S. A. Kreek, M. J. Nurmia, M. R. Lane, M. P. Neu, A. Türler, and D. C. Hoffman, Phys. Rev. C 46, 1873 (1992).
  101. [102]M. R. Lane, K. E. Gregorich, D. M. Lee, M. F. Mohar, M. Hsu, C. D. Kacher, B. Kadkhodayan, M. P. Neu, N. J. Stoyer, E. R. Sylwester, J. C. Yang, and D. C. Hoffman, Phys. Rev. C 53, 2893 (1996).
  102. [103]http://www.khs-erzhausen.de/GEF.html.
  103. [104]A. Bauswein, H.-T. Janka, and R. Oechslin, Phys. Rev. D 82, 084043 (2010).
  104. [105]A. W. Steiner, M. Hempel, and T. Fischer, The Astrophysical Journal 774, 17 (2013).
  105. [106]Y. Xu, S. Goriely, A. Jorissen, G. Chen, and M. Arnould, A&A 549, 10 (2013), A106.
  106. [107]S. Goriely, S. Hilaire, and A. J. Koning, Astronomy and Astrophysics 487, 767 (2008).
  107. [108]S. Goriely, N. Chamel, and J. M. Pearson, Phys. Rev. C 82, 035804 (2010).
  108. [109]T. Marketin, L. Huther, and G. Martinez-Pinedo, Phys. Rev. C 93, 025805 (2016).
  109. [110]S. Goriely, M. Samyn, and J. M. Pearson, Phys. Rev. C 75, 064312 (2007).
  110. [111]S. Goriely, S. Hilaire, A. J. Koning, M. Sin, and R. Capote, Phys. Rev. C 79, 024612 (2009).
  111. [112]H. Koura, T. Tachibana, and T. Yoshida, J. Nucl. Sci. Tech. Suppl. 2, 774 (2002).
  112. [113]S. Goriely, Astronomy and Astrophysics 342, 881 (1999).
  113. [114]R. Hoffman, S. Woosley, and Y.-Z. Qian, Astrophys. J 482, 951 (1997).
  114. [115]Y.-Z. Qian and S. Woosley, Astrophys. J 471, 331 (1996).

Paper details

vlnc:W_fpb9kg7h
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents