betaFeedback

WorksPhysics and Astronomy

Solar-System Abundances of pp-Nuclides Probe Collective Neutrino Oscillations in Supernovae

Abstract

Direct evidence for collective neutrino oscillations in core-collapse supernovae remains elusive. We show that this quantum phenomenon leaves a footprint on the abundance pattern of proton-rich nuclides in the solar system. Modeling the νpνp-process using a 20 M⊙20\,M_\odot progenitor, we map out the dependence of the total yields on the starting radius of the oscillations, self-consistently coupling hydrodynamics and nucleosynthesis. The oscillations boost key pp-nuclides (92,94Mo^{92,94}\text{Mo}, 96,98Ru^{96,98}\text{Ru}) and long-lived 92Nb^{92}\text{Nb} by up to two orders of magnitude, bringing their abundances into agreement with the observations. The best match is found when oscillations commence within 10 km10\text{ km} of the proto-neutron star surface, indicating fast collective oscillations.

Introduction

Just outside the protoneutron star (PNS) in a core-collapse supernova (CCSN), the dense gas of streaming neutrinos creates extreme conditions unattainable in terrestrial laboratories. Coherent ν\nu–ν\nu scattering in this system couples the flavor states of neutrinos on different trajectories, driving collective flavor oscillations [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]. Analytical and numerical studies of this many-body quantum phenomenon have uncovered very rich physics, including fast collective oscillations (FCO) [14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31] driven purely by ν\nu–ν\nu coherent scattering and slow collective oscillations (SCO) [32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45] driven by the interplay between the ν\nu–ν\nu coherent scattering and neutrino-mass-induced transformations. Yet, a complete theoretical description of this phenomenon in realistic CCSN conditions remains elusive, due to the extreme complexity of the problem rooted in the nonlinearity of oscillations, the presence of disparate length- and timescales [46, 47, 48], and essential multi-dimensional dynamics [13, 49]. It is then crucial to identify any observables that carry imprints of FCO or SCO, especially those that are sensitive to how far from the PNS surface the oscillations develop. In this Letter, we argue that the desired probe is provided by the pattern of certain proton-rich isotopes observed in the Solar System. Moreover, we show that the oscillations may be necessary to explain the observations.

It has long been known that if collective oscillations were to occur in the hot bubble surrounding the PNS, they could impact nucleosynthesis there [34, 50, 51]. These early papers considered the rr-process (rapid neutron capture) and invoked sterile neutrinos. It has since been shown, however, that collective oscillations can occur in the hot bubble even with only active neutrinos [35]. Combining representative oscillation scenarios with nucleosynthesis calculations then showed that the νr\nu r-process yields can depend sensitively on where flavor mixing starts [52]. This makes nucleosynthesis a promising complementary probe to neutrino signal detected on Earth, as the latter only reflects the final state of the neutrino evolution.

Our understanding of the type of nucleosynthesis that can occur in the hot bubble has also advanced. The νe\nu_e and νˉe\bar{\nu}_e spectra predicted by modern CCSN simulations tend to yield proton-rich conditions in the neutrino-driven outflow from the PNS surface that forms the hot bubble, especially once the effect of weak magnetism and recoil [53, 54, 55] is accounted for [56]. This is, in fact, a welcome development. While the rr-process may efficiently proceed in neutron star binary mergers [57, 58, 59], one also needs to identify a site where pp-nuclides are made. The origin of these isotopes, especially 92,94Mo{}^{92,94}\mathrm{Mo} and 96,98Ru{}^{96,98}\mathrm{Ru}, which are rather abundant in the Solar System, has been a long-standing mystery [60, 61]. The hot bubble in a CCSN turns out to be a promising site, supplying not only the proton-rich conditions, but also a large flux of streaming electron antineutrinos. The latter helps solve [62, 63, 64] the famous “waiting points problem” of the classic rprp-process [65]. This neutrino-assisted mechanism is called the νp\nu p-process [62].

While there are several recent notable investigations of the impact of FCO on the CCSN explosion mechanism [66, 67, 68, 69, 70, 71, 72, 73, 74], there is only a single study on how the νp\nu p-process is affected by FCO [75]. There is, likewise, a limited number of studies exploring the effects of SCO on the νp\nu p-process [76, 77, 78]. While these papers contain important findings, they do not treat the impact of the flavor transformations on the hydrodynamics self-consistently, which is required to tackle the problem.

**Analysis requirements.—**Flavor oscillations change the spectra of νe\nu_e and νˉe\bar{\nu}_e, by mixing them with the original spectra of νx\nu_x and νˉx\bar{\nu}_x, with x=μ,τx=\mu,\tau. This impacts the νp\nu p-process in three ways. First, it changes the rates of neutron-producing reactions νˉe+p→e++n\bar{\nu}_e+p\rightarrow e^++n during the proton-capture phase of the νp\nu p-process. This is a far effect of the oscillations, as it occurs hundreds of km from the PNS surface [75, 76, 77, 78]. Second, oscillations modify the electron fraction YeY_e in the outflow, set by the reactions νˉe+p→e++n\bar{\nu}_e+p\rightarrow e^++n and νe+n→e−+p\nu_e+n\rightarrow e^-+p. Since the unoscillated energy hierarchy is typically larger for νe\nu_e–νx\nu_x than νˉe\bar{\nu}_e–νˉx\bar{\nu}_x, mixing hardens νe\nu_e more than νˉe\bar{\nu}_e, enhancing νe+n→p+e−\nu_e+n\rightarrow p+e^- relative to νˉe+p→n+e+\bar{\nu}_e+p\rightarrow n+e^+ and increasing YeY_e [75]. This is only relevant if the oscillations occur within 10–20 km from the PNS surface, where YeY_e is set (see Figure 1). It is thus considered a near effect. Third, as another near effect, flavor conversion increases the heating rate Q˙\dot{Q} [75].

Figure 1. Electron fraction YeY_e as a function of radius for our benchmark 20 M⊙20\,M_\odot model at t=3.5 st=3.5\,\mathrm{s}, shown for the outflow without neutrino mixing and with mixing occurring at various distances Δrmix\Delta r_{\rm mix} from the PNS surface.

The last effect energizes the outflow and deserves special consideration. Two key physical factors enter here: (i) Effective νp\nu p-process requires a certain matching between the processes that occur at different stages of the outflow [79, 80]. The number of seeds produced during the stage of quasi-statistical equilibrium (3 GK≲T≲6 GK3\,\mathrm{GK}\lesssim T\lesssim6\,\mathrm{GK}) has to be in a certain ratio with the number of neutrons produced in the νˉe(p,n)e+\bar{\nu}_e(p,n)e^+ reactions in the pp-nuclide formation stage (1.5 GK≲T≲3 GK1.5\,\mathrm{GK}\lesssim T\lesssim3\,\mathrm{GK}). Refs. [80, 81] established the hydrodynamic conditions for optimal yields: a heavy PNS mass and a subsonic outflow persisting for several seconds. (ii) As argued in [82], the outflow in a CCSN, under realistic conditions, can be either subsonic or supersonic, depending on a combination of relevant parameters. Because of this near-critical nature of the outflow, the extra heating induced by neutrino oscillations might cause supersonic transitions, depressing the yields and counteracting the effect of boosted YeY_e. To assess the effect of flavor mixing on the yields then requires us to recompute the hydrodynamics of the outflow for each oscillation scenario.

**Methods.—**We use as benchmark a spherically symmetric long-term supernova simulation of a 20 M⊙20\,M_\odot progenitor with a 1.93 M⊙1.93\,M_\odot PNS from the Garching CCSN Archive [83, 84, 85]. This state-of-the-art simulation meets the hydrodynamic requirements for efficient νp\nu p, as outlined above, yet was not tailored for the νp\nu p-process by its authors. We take as input the PNS mass MM, time-dependent PNS radius, (unoscillated) neutrino luminosities and spectra, and the progenitor profile [86]. We model the luminosity evolution with the fit from Ref. [85], and we fix spectral moments at their values at post-bounce time t=3.5 st=3.5\,\mathrm{s} in the SN simulation (when the νp\nu p-process peaks), assuming pinched Fermi-Dirac distributions [87, 88, 89].

As mentioned above, we are compelled to recompute the hydrodynamics for each oscillation scenario. We do so with the framework of Ref. [81], which solves general-relativistic steady-state hydrodynamics in spherical symmetry, with an equation of state including the baryon gas and variable radiation degrees of freedom. Neutrino interactions with both nucleons and electrons are included, and time-dependent boundary conditions at the PNS surface and the hot-bubble’s edge (with proper merging into the colder outflow) are imposed, as discussed in Supplementary Material (SM) A, B. We construct the time evolution through a sequence of snapshots, representing particles launched 0.1 s apart.

Following existing literature [75], we model flavor conversion as complete instantaneous equilibration at a distance Δrmix\Delta r_{\rm mix} from the PNS surface. Beyond this radius, we replace the original spectra with the equilibrated ones, combining all flavors equally across the energy modes: fνe(νˉe)′=(fνe(νˉe)+2fνx(νˉx))/3f_{\nu_e(\bar{\nu}_e)}'=(f_{\nu_e(\bar{\nu}_e)}+2f_{\nu_x(\bar{\nu}_x)})/3, thereby modifying neutrino energy moments and luminosities (see SM C). [90] This modifies both neutrino heating (thus hydrodynamics) and nucleosynthesis. We vary Δrmix\Delta r_{\rm mix} as a free parameter to quantify the νp\nu p-process’s sensitivity to conversion location.

Nucleosynthesis is computed by post-processing tracer trajectories with the reaction network SkyNet [91, 92], incorporating modifications from Ref. [81] as detailed in SM A (including medium-enhanced triple-α\alpha [93, 94]). We further modify the code to include distance-dependent flavor mixing (SM C). Each trajectory is launched at tlaunch∈[1,10] st_{\rm launch}\in[1,10]\,\mathrm{s} after the core bounce and evolves for ∼30\sim30 years in SkyNet, with conditions evaluated at t=tlaunch+tskyt=t_{\rm launch}+t_{\rm sky}, where tskyt_{\rm sky} is SkyNet’s evolution time (starting from zero at launch). Beyond the benchmark model, we perturb the PNS and progenitor’s properties to test the robustness of our qualitative findings across different regimes.

**Oscillation-driven physical effects.—**The impact of neutrino oscillations on the νp\nu p-process depends on how far from the PNS flavor mixing takes place. If conversions occur at Δrmix≳40 km\Delta r_{\mathrm{mix}} \gtrsim40\,\mathrm{km} (“far conversions”), only the rate of neutron production during the proton-capture phase is appreciably changed, while YeY_e and Q˙\dot{Q} are not affected. For the model we use in this paper, the νˉe(p,n)e+\bar{\nu}_e(p,n)e^+ rates are enhanced by ∼20%\sim20\%.

Conversions at smaller radii (“near conversions”) induce all three effects simultaneously: ∼20%\sim20\% enhancement of νˉe(p,n)e+\bar{\nu}_e(p,n)e^+ rates, up to ∼30%\sim30\% increase in Q˙\dot{Q} (and also the mass outflow rate M˙\dot{M}), and boosted YeY_e. The impacts on YeY_e and Q˙\dot{Q} exhibit similar dependence on Δrmix\Delta r_{\mathrm{mix}} (see SM C).

Figure Figure 1 shows Ye(r)Y_e(r) at a representative post-bounce time for several values of the flavor equilibration radius. Notice that the important weak-magnetism and recoil corrections [53, 54, 55] are included here and in what follows. With the original neutrino spectra, YeY_e approaches an equilibrium value of ≈0.56\approx0.56 (solid violet). Neutrino mixing increases YeY_e, rendering the outflow more proton-rich, consistent with Ref. [75]. This enhancement is stronger the closer mixing occurs to the PNS. For Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\,\mathrm{km}, YeY_e increases up to ≈0.59\approx0.59 (solid yellow), while there is negligible impact for Δrmix≳30 km\Delta r_{\mathrm{mix}} \gtrsim30\,\mathrm{km}.

Notice that earlier studies found Ye≈0.6Y_e \approx0.6 to be conducive to successful νp\nu p-process, e.g. [79, 80, 81]. That flavor mixing brings YeY_e close to this desired value is, in fact, not an accident of the specific simulation we are considering and rather insensitive to the details of flavor conversion. As can be readily shown (see SM A), in the limit when the νe\nu_e and νˉe\bar{\nu}_e spectra are identical, YeY_e reaches ≈0.63\approx0.63. Since the original νx\nu_x and νˉx\bar{\nu}_x spectra are close, any flavor mixing dynamics that transfers substantial νx\nu_x flux into νe\nu_e boosts YeY_e to near the desired value ∼0.6\sim0.6. For instance, were the outcome of near conversion to be spectral swaps of the kind studied in Refs. [37, 41, 42, 95–98], the resulting YeY_e shift would be comparable.

Results of near flavor conversion.—Oscillation-induced changes to YeY_e (Figure 1) and Q˙\dot{Q} (explicitly shown in SM C) are largest as flavor conversion occurs close to the PNS. Thus, we begin by comparing two extreme cases: no mixing and flavor equilibration occurring 1 km1\,\mathrm{km} above the PNS surface (Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\,\mathrm{km}).

The left panels of Figure 2 show the evolution of the outflows launched at different tlauncht_{\mathrm{launch}} for our benchmark 20 M⊙20\,M_\odot model. Without mixing, the outflow remains subsonic throughout the cooling phase (upper left). For Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\,\mathrm{km} (lower left), the additional heating induces a transonic transition at tlaunch≳4.5 st_{\mathrm{launch}} \gtrsim4.5\,\mathrm{s}, and the termination shocks, visible as discontinuities in the radial temperature profiles around r≃108 cmr \simeq10^8\,\mathrm{cm}, persist until tlaunch≈8 st_{\mathrm{launch}} \approx8\,\mathrm{s}. Such emergence of transonic outflows from a modest increment of heating illustrates the near-critical nature of CCSN hydrodynamics [82], and, practically, the necessity of self-consistently solving the hydrodynamics for small Δrmix\Delta r_{\rm mix}.

Figure 2. Left: Radial temperature profiles for outflows launched at tlaunch∈[1,10] st_{\mathrm{launch}} \in[1,10]\,\mathrm{s}, depicted here in steps of 0.5 s0.5\,\mathrm{s}. The green and yellow bands correspond to the seed-formation and pp-nuclide formation stages of the νp\nu p-process, respectively [79, 81]. Right: Time evolution of the production rates of Mo (red) and Ru isotopes (blue), and 92Nb{}^{92}\mathrm{Nb} (black), with vertical dotted lines delimiting the time window where the outflows contain a supersonic transition. Top (bottom) panels show the unmixed case (flavor equilibration occurring 1 km1\,\mathrm{km} above the PNS surface).

The right panels of Figure 2 show the corresponding production rates as a function of time of several pp-nuclides and the long-lived radionuclide 92Nb{}^{92}\mathrm{Nb}, a key diagnostic of the νp\nu p-process’s efficiency [61, 80, 81].[99] Without mixing (upper right), pp-nuclides are efficiently produced in an optimal window of ∼3\sim3–44 s in the early cooling phase, while 92Nb{}^{92}\mathrm{Nb} is produced later. With Δrmix=1\Delta r_{\rm mix}=1 km (lower right), the synthesis of Mo and Ru isotopes is enhanced by factors of ∼10\sim10–2020 and ∼30\sim30–6060, respectively, reflecting higher YeY_e and (anti)neutrino capture rates. Importantly, the supersonic transition sets in only after the production rates have peaked (vertical dotted line at tlaunch≃4.5t_{\rm launch}\simeq4.5 s). While the onset of mildly supersonic conditions noticeably accelerates the post-peak decline of Mo and Ru nucleosynthesis, the production rates remain higher than the original (unmixed) case at each tlauncht_{\rm launch}. Oscillations increase 92Nb{}^{92}\mathrm{Nb} production by as much as two orders of magnitude. In contrast to pp-nuclides, 92Nb{}^{92}\mathrm{Nb} has a significantly broadened production window and distinct temporal behavior.

Yields as a function of mixing radius.—To compare with observations, we next compute the total yields in a given explosion by integrating over time. We present our results as time-averaged yields normalized by the total mass ejected

⟨YA,Z⟩=∫dt YA,Z(t)M˙(t)∫dt M˙(t).(1)\langle Y_{A,Z}\rangle= \frac{\int dt\,Y_{A,Z}(t)\dot{M}(t)}{\int dt\,\dot{M}(t)} . \tag*{(1)}

With this, we systematically investigate the dependence of ⟨YA,Z⟩\langle Y_{A,Z}\rangle on the mixing location, the first time such a study is undertaken in the literature. Figure 3 shows ⟨YA⟩=∑Z⟨YA,Z⟩\langle Y_A\rangle= \sum_Z \langle Y_{A,Z}\rangle versus mass number AA for near (left) and far (right) conversions for different values of Δrmix\Delta r_{\rm mix}. For the pp-nuclides labeled on the upper axis, ⟨YA⟩\langle Y_A\rangle corresponds to ⟨YA,Z⟩\langle Y_{A,Z}\rangle of the indicated isotope, the only relevant species at that AA produced by the νp\nu p-process.

Figure 3. Time-averaged yields ⟨YA⟩\langle Y_A\rangle for our benchmark 20 M⊙20\,M_\odot model under flavor conversion occurring at various radii from the PNS. Curves show ⟨YA⟩\langle Y_A\rangle versus mass number AA; diamonds mark the corresponding 92Nb{}^{92}\mathrm{Nb} abundance. Left: “Near” conversion, i.e., mixing within 30 km from the PNS surface, which modifies YeY_e and Q˙\dot{Q} and also enhances neutron production rates. Right: “Far” conversion, i.e., mixing at larger radii, affecting only neutron production during the pp-nuclide formation stage. Black arrows indicate the minimum abundances needed to reproduce the solar values (see main text and SM).

From the left panel of Figure 3, we see that the largest enhancement is obtained when Δrmix<10\Delta r_{\rm mix}<10 km. Here, for 90≤A≤10090\leq A\leq100, flavor equilibration enhances the yields up to ∼2\sim2 orders of magnitude relative to the unmixed case (solid violet) for both pp-nuclides (curves) and 92Nb{}^{92}\mathrm{Nb} (diamonds), with factors of ∼10\sim10 for 92Mo{}^{92}\mathrm{Mo} and ∼60\sim60 for 98Ru{}^{98}\mathrm{Ru}. This reflects the enhancement in the production rates illustrated in Figure 2. Counterintuitively, despite larger YeY_e values at smaller Δrmix\Delta r_{\rm mix} (Figure 1), the cases with Δrmix=1,5,10\Delta r_{\rm mix}=1,5,10 km yield similar ⟨YA⟩\langle Y_A\rangle. This is due to the competing effect of the increase in heating: mixing closer to the PNS triggers an earlier transonic transition that shortens the optimal production window, partially offsetting the YeY_e-driven gain.

For 10 km≲Δrmix≲4010\ {\rm km}\lesssim\Delta r_{\rm mix}\lesssim40 km the νp\nu p-yields are strongly sensitive to the value of Δrmix\Delta r_{\rm mix}. Here, no supersonic transition takes place, and the effect of YeY_e is not counterbalanced by hydrodynamics. As expected, the yields fall monotonically with increasing Δrmix\Delta r_{\rm mix}, approaching a factor of 2–3 enhancement relative to the unmixed case for Δrmix=40\Delta r_{\rm mix}=40 km (see SM D).

For 40 km≲Δrmix≲30040\ {\rm km}\lesssim\Delta r_{\rm mix}\lesssim300 km (right panel of Figure 3) the yields are nearly independent of Δrmix\Delta r_{\rm mix}. In this regime, YeY_e and Q˙\dot{Q} are effectively not modified, leaving the far effect as the only relevant process, which operates at the pp-nuclide production stage, starting roughly at 300 km (see Figure 2, top-left panel). At larger Δrmix\Delta r_{\rm mix}, enhancement diminishes, becoming negligible for Δrmix≳3000\Delta r_{\rm mix}\gtrsim3000 km where pp-nuclide production ends.

**Comparison with solar abundances.—**Black arrows in Figure 3 mark solar-abundance thresholds from meteoritic data [100] for 92Mo{}^{92}\mathrm{Mo}–106Cd{}^{106}\mathrm{Cd}. These are computed by requiring that νp\nu p-process mass fractions, when diluted in the total ejecta, exceed solar values by one order of magnitude [79, 81] (see SM D). The unmixed case falls short by approximately one order of magnitude. Far conversions, increasing yields by only factors of a few, remain insufficient to reach solar thresholds. In contrast, near conversions enable agreement with solar abundances [79, 81]: our model successfully meets solar requirements of the key Mo and Ru isotopes for Δrmix≲20\Delta r_{\mathrm{mix}} \lesssim20–30 km30\ \mathrm{km}, while for Δrmix≲10 km\Delta r_{\mathrm{mix}} \lesssim10\ \mathrm{km}, 92Nb{}^{92}\mathrm{Nb}, as well as 102Pd{}^{102}\mathrm{Pd}, become viable.

**Discussion.—**Near flavor conversion makes our 20 M⊙20\,M_{\odot} model an efficient νp\nu p-process site. Modest perturbations of the benchmark parameters preserve our qualitative results: when conversion occurs sufficiently close to the PNS surface, yields reach the solar thresholds, provided that the PNS and progenitor remain sufficiently massive. This is demonstrated for a representative case in SM E, obtained by reducing the progenitor mass to 18.6 M⊙18.6\,M_{\odot} and modestly increasing the front shock velocity to 7000 km/s7000\ \mathrm{km/s}. In this model, lower confining pressure triggers an earlier transonic transition, clipping the optimal window even without neutrino mixing. Flavor conversion leads to similar enhancement factors of the yields: for Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\ \mathrm{km}, the transonic transition occurs ∼1 s\sim1\ \mathrm{s} earlier, yet an early subsonic interval persists wherein boosted YeY_e and M˙\dot{M} enhance production. Though reduced relative to the benchmark, yields of 92,94Mo{}^{92,94}\mathrm{Mo}, 96,98Ru{}^{96,98}\mathrm{Ru}, and 92Nb{}^{92}\mathrm{Nb} still exceed solar thresholds for Δrmix≲10 km\Delta r_{\mathrm{mix}} \lesssim10\ \mathrm{km}.

More generally, flavor mixing enhances yields in all cases we explored, even when (mild) transonic transitions already occur near the beginning of the cooling phase. However, the νp\nu p-process is inefficient in strongly supersonic conditions [80, 81], rendering mixing-induced enhancements insufficient to reach solar thresholds. An exploration of different models is left for future work [101].

Our systematic study establishes that near conversion—triggered by FCO (see e.g., Refs. [9, 10, 12] for recent reviews) within ∼30 km\sim30\ \mathrm{km} from the PNS surface [21, 22, 23, 24, 25]—increases yields by more than one order of magnitude, allowing agreement with solar observations for sufficiently massive progenitors and PNS. By contrast, far conversion, associated with SCO at 4040–500 km500\ \mathrm{km} [35, 36, 37, 38], enhances yields by only factors of a few. Mixing at larger distances, i.e. Δrmix∼O(103) km\Delta r_{\mathrm{mix}} \sim\mathcal{O}(10^3)\ \mathrm{km}, has negligible impact [102]. These results suggest that FCO may play a crucial role in establishing the observed abundances of light pp-nuclides in the solar system.

The relative importance of far and near conversions depends on the closeness of unoscillated νˉe\bar{\nu}_e and νˉx\bar{\nu}_x spectra in both energy and luminosity during cooling. Similar spectra are commonly observed in state-of-the-art simulations [84, 103, 104, 105, 106, 107], while large differences typically occur in simulations neglecting inelastic neutrino-nucleon scattering corrections [77] (see Fig. 14 in Ref. [108]). The GARCHING group finds an average energy difference Δ⟨Eν⟩≡⟨Eνˉx⟩−⟨Eνˉe⟩≲0.5 MeV\Delta\langle E_{\nu}\rangle\equiv\langle E_{\bar{\nu}_x}\rangle- \langle E_{\bar{\nu}_e}\rangle\lesssim0.5\ \mathrm{MeV} [84, 103], slightly smaller than the one observed by other groups (Δ⟨Eνˉ⟩≲1\Delta\langle E_{\bar{\nu}}\rangle\lesssim1–2 MeV2\ \mathrm{MeV}) [104, 105, 106, 107]. Our far-conversion results agree qualitatively with Ref. [76], which assumes Δ⟨Eνˉ⟩≈0.9 MeV\Delta\langle E_{\bar{\nu}}\rangle\approx0.9\ \mathrm{MeV} based on simulations from Ref. [109]. By contrast, adopting greater νˉe\bar{\nu}_e–νˉx\bar{\nu}_x spectral differences (Δ⟨Eνˉ⟩≳5 MeV\Delta\langle E_{\bar{\nu}}\rangle\gtrsim5\ \mathrm{MeV}), Refs. [77, 78] found substantially larger enhancements in νˉe\bar{\nu}_e-capture rates and νp\nu p-yields. Thus, the impact of far conversion is sensitive to the details of neutrino transport, which motivates careful future investigations, given its importance to the estimation of νp\nu p-yields in the presence of oscillations.

**Conclusions.—**We saw that flavor oscillations close to the PNS can transform the nature of the outflow and with it the efficiency of the νp\nu p-process. This requires the recomputation of the hydrodynamics for each oscillation scenario, while also including the oscillation effects on the electron fraction and the neutron production rate. Such a self-consistent treatment of the hydrodynamics and nucleosynthesis, as well as the systematic investigation of the dependence of the yields on the starting radius of the oscillations, are presented here for the first time.

Our calculations confirm that the yields of the νp\nu p-process are sensitive to the starting radius of collective oscillations. We find that the impact of the oscillations is greatest when they occur within ∼10 km\sim10\ \mathrm{km} of the PNS surface, in which case the yields of the pp-nuclides in the crucial A=90A = 90–100100 mass range are boosted by up to 2020–6060 times. In comparison, when the neutrino mixing radius is ≳40 km\gtrsim40\ \mathrm{km}, the enhancement is an order of magnitude smaller and, moreover, the sensitivity to the exact location of the oscillations is reduced.

Crucially, when the oscillations occur close to the PNS surface, and only then, the predicted abundances line up with the solar-system observations. This applies to all pp-isotopes up to 102Pd{}^{102}\mathrm{Pd}, as well as to 92Nb{}^{92}\mathrm{Nb}, which also serves as a probe of the efficiency of the νp\nu p-process. Thus, collective oscillations may be responsible for the observed pattern of these isotopes. Notably, the agreement is reached entirely within the physical conditions of a modern CCSN simulation, without tuning any aspects of explosion, such as the electron fraction, neutrino fluxes, entropy per baryon, or the outflow profile, the first such example in the literature.

Given today’s best available information on the dynamics of collective oscillation and on the spectra of different neutrino flavors, our results might be taken as an indication that the relevant mechanism is fast collective. Further investigations incorporating collective oscillation dynamics self-consistently into CCSN simulations are needed to investigate the robustness of this finding.

Acknowledgments

The work of all authors at SLAC was supported by the U.S. Department of Energy under contract number DE-AC02-76SF00515. AVP acknowledges partial support from the U.S. Department of Energy under contract number DE-FG02-87ER40328 at the University of Minnesota, and would also like to thank SLAC for their hospitality and support during the period of completion of this project. IPG is supported by NSF Physics Frontier Center Award number 2020275. PM thanks the Neutrino Theory Network Program Grant under award number DE-AC02-07CHI11359, which supported this work while she was at UC Berkeley. Additionally, PM is grateful for the support of the Cambridge-Infosys AI Center for funding her current research.

References

Solar-System Abundances of p-Nuclides Probe Collective Neutrino Oscillations in Supernovae

In this Supplemental Material (SM), we describe the methods employed in modeling the hydrodynamics, nucleosynthesis, and flavor equilibration prescription. We present in detail the model parameters extracted from the GARCHING supernova simulation, define the production factors used to assess the agreement with solar abundances, and extend our analysis in the main text to mildly supersonic models.

Methods: hydrodynamics and nucleosynthesis

We compute nucleosynthetic yields following a two-step approach, akin to Refs. [80, 81]. First, we construct the trajectories of tracer particles for a number of time snapshots. Then, we obtain the yields by post-processing each tracer trajectory with the open-source nuclear reaction network SkyNet [91].

Tracer trajectories are divided into two phases. In the first, described as a steady-state neutrino-driven outflow (NDO), material accelerates from the protoneutron star (PNS) surface and subsequently decelerates through interaction with the surrounding ejecta. In the second phase, the material joins the homologous expansion behind the forward shock and co-moves with it.

Whether the outflow becomes supersonic and develops a termination shock, or instead remains subsonic throughout, is not known a priori. The system often lies close to criticality [82], as our benchmark model demonstrates, with the character of the outflow determined by the interplay of neutrino heating, PNS properties, and the confining pressure imposed by the progenitor profile. Rather than postulating a supersonic or subsonic solution, we solve the hydrodynamic equations as a boundary value problem consistent with the confining pressure, at the edge of the hot bubble, inferred from the progenitor profile and front-shock velocity at each temporal snapshot. This method is described in detail in Ref. [81]. To incorporate the effect of neutrino oscillations, we posit a radius after which the neutrino flavors equilibrate (see SM C), which maximizes the effect of neutrino mixing, and vary this radius as a parameter. Concretely, this means that the neutrino spectra which enter the neutrino capture terms in the outflow calculation as well as in SkyNet, are discretely modified before and after this mixing radius. This rather minimal prescription of positing a radius of equilibration does offer us a straightforward way of studying the radial dependence of oscillations’ impact on nucleosynthesis agnostic of the microphysics.

The main ingredients of our model for the outflow have been delineated in the Letter and detailed in [81]. Here we accentuate a few aspects. Although our hydrodynamic equations assume the steady-state condition, we consider the time evolution of the boundary conditions—including PNS radius, luminosities, confining pressure—and compute the outflow every 0.1 s. This yields a series of snapshots. Our Equation of State (EoS) incorporates variable radiation degrees of freedom and the monatomic baryon gas. A key ingredient in determining the NDO is the neutrino net heating rate Q˙\dot{Q}, computed as in Ref. [81]. We include (i) charged-current absorption and emission on free nucleons, νe+n↔p+e−\nu_e + n \leftrightarrow p + e^{-} and νˉe+p↔n+e+\bar{\nu}_e + p \leftrightarrow n + e^{+}, (ii) elastic neutrino-electron scattering e+ν→e+νe + \nu\rightarrow e + \nu, and (iii) cooling from electron-positron annihilation e−+e+→ν+νˉe^{-} + e^{+} \rightarrow\nu+ \bar{\nu}, following the prescription in Refs. [116, 117]. Charged current absorptions, with a heating rate scaling with the third moment of the neutrino energy distribution—two powers from the cross section and one from the deposited energy—are the main heating processes and also determine the neutron-to-proton ratio in the outflow. Since SN simulations typically provide only the first two energy moments, we parametrize neutrino spectra (see SM B) to reconstruct higher-order moments required to compute Q˙\dot{Q}.

Nucleosynthesis is computed by post-processing each tracer with SkyNet, using nuclear reaction rates from the ReacLib library, version 2.2 [118], with the modifications described in Ref. [81] (see Sec. 8 and App. G therein), including for instance time-evolving neutrino luminosities, the blueshift of neutrino temperature and luminosities, medium-enhanced triple-α\alpha reaction rates from Refs. [93, 94], and the correct decay rate for 92Nb{}^{92}\mathrm{Nb} [80]. Additionally, for this analysis we further modify the code to account for neutrino flavor conversion at a mixing radius rmix=RPNS(t)+Δrmixr_{\rm mix} = R_{\rm PNS}(t) + \Delta r_{\rm mix}. For each tracer, we start nucleosynthesis runs at T=4.5T = 4.5 MeV, using as SkyNet inputs the time evolution of its temperature TT, density ρ\rho, and its location rr, along with a constant PNS mass and time-dependent PNS radius RPNS(t)R_{\rm PNS}(t), assumed to be equal to the neutrinosphere radius. The time evolution of the electron fraction YeY_e is computed by SkyNet and it does incorporate weak-magnetism (WM) and recoil corrections with the method described in Ref. [81].

In the absence of mixing, YeY_e approaches the asymptotic value given by [53, 55, 119]

Ye≈(1+LνˉeLνeϵνˉe−2Δ−7.1ϵνˉe2/mNϵνe+2Δ+1.1ϵνe2/mN)−1,(2)Y_e \approx\left(1+\frac{L_{\bar{\nu}_e}}{L_{\nu_e}}\frac{\epsilon_{\bar{\nu}_e}-2\Delta-7.1\epsilon_{\bar{\nu}_e}^{2}/m_N}{\epsilon_{\nu_e}+2\Delta+1.1\epsilon_{\nu_e}^{2}/m_N}\right)^{-1}, \tag*{(2)}

where mNm_N is the nucleon mass, Δ=1.293 MeV\Delta=1.293\ \mathrm{MeV}, ϵνi=⟨Eνi2⟩/⟨Eνi⟩\epsilon_{\nu_i}=\langle E_{\nu_i}^{2}\rangle/\langle E_{\nu_i}\rangle, and ενi=(⟨Eνi3⟩/⟨Eνi⟩)1/2\varepsilon_{\nu_i}=(\langle E_{\nu_i}^{3}\rangle/\langle E_{\nu_i}\rangle)^{1/2}, with ⟨Eνin⟩\langle E_{\nu_i}^{n}\rangle the nn-th moment of the energy distribution. Note that this expression is derived using an approximation to the exact cross section, only including terms up to first order in Δ/Eνi\Delta/E_{\nu_i} and Eνi/mNE_{\nu_i}/m_N [119]. For a more exact treatment, and how it reduces to this expression at leading order, see Eqs. (22)–(25) from Ref. [55]. In presence of flavor conversion, the asymptotic value of YeY_e increases and depends on the radius at which conversion occurs, as discussed in the main Letter.

To incorporate flavor mixing, we modify SkyNet to include νx\nu_x and νˉx\bar{\nu}_x in addition to νe\nu_e and νˉe\bar{\nu}_e, which were the only species considered in our previous analyses. As input, we provide time-dependent unmixed neutrino luminosities and spectra. Each (anti)neutrino spectrum is modeled using a fit that enables the reconstruction of higher-order energy moments from those provided by supernova simulations (see SM B). Starting from the original spectra of νe\nu_e, νˉe\bar{\nu}_e, νx\nu_x, and νˉx\bar{\nu}_x, we compute distance-dependent spectra and determine the corresponding mixed quantities at the radius rmixr_{\mathrm{mix}}, as described in SM C.

Model parameters from simulations

We model the NDO using input quantities extracted from simulations performed with the PROMETHEUS-VERTEX neutrino-hydrodynamics code [120]. This code solves the fully energy- and velocity-dependent neutrino transport equations for all six species of neutrinos and antineutrinos, employing a state-of-the-art treatment of neutrino interactions [109, 121, 122]. As discussed in Ref. [84], these simulations include a one-dimensional treatment of PNS convection through a mixing-length description of the convective fluxes [7]. They also account for the presence of muons in the hot PNS and the associated muonic-neutrino interactions [122]. However, the resulting differences between the luminosities and spectra of μ\mu and τ\tau neutrinos remain relatively small throughout most of the PNS cooling phase.

For our benchmark model, we adopt parameters from a 1D SN simulation from a 20 M⊙20\,M_\odot progenitor [86] performed by the GARCHING collaboration [83]. This model possesses the qualitative features required for efficient νp\nu p nucleosynthesis [80], specifically a massive PNS (with baryonic mass of M=1.93 M⊙M=1.93\,M_\odot) and subsonic outflows. We notice that such heavy PNS masses naturally arise in multi-dimensional simulations with massive progenitors [103, 123].

In this model, the evolution of the PNS radius (solid line in Figure S1) is well fitted by

Figure S1. Evolution of the PNS radius as a function of post-bounce time for our benchmark 1.93 M⊙1.93\,M_\odot model, parametrized according to Eq. (3).

RPNS(t)=[a+b(t1 s)−c] km,(3)R_{\mathrm{PNS}}(t) = \left[a + b\left(\frac{t}{1\ \mathrm{s}}\right)^{-c}\right]\ \mathrm{km}, \tag*{(3)}

where tt denotes the post-bounce time in seconds and the coefficients are a=9.97a = 9.97, b=11.40b = 11.40 and c=0.59c = 0.59. We adopt this as the near boundary condition for our outflow calculation. The far boundary conditions are fixed in our model by taking a front-shock (FS) velocity of vFS=6000 km/sv_{\mathrm{FS}} = 6000\ \mathrm{km}/\mathrm{s}, consistent with the simulation results.

To assess the impact of an early transition from subsonic to supersonic outflows, we also perturb our benchmark model by considering a slightly lighter progenitor of 18.6 M⊙18.6\,M_\odot [86], with vFS=7000 km/sv_{\mathrm{FS}} = 7000\ \mathrm{km}/\mathrm{s}, while keeping the PNS properties fixed. A more detailed discussion about this model is reported in SM E.

Our reference SN simulation employs the SFHo EoS [124, 125] and corresponds to the 1.93-SFHo model in Refs. [84, 85]. We describe below its main features.

The presence of convection accelerates PNS cooling and modifies the time evolution of the neutrino luminosity. As shown in Ref. [126], a kink in the neutrino light curve when plotted in a doubly logarithmic scale marks the end of convection in the mantle region of the PNS, while convection remains active in the deeper high-density core. This feature typically occurs at t≲10 st \lesssim10\ \mathrm{s} [84, 85, 127]. Consequently, the neutrino luminosity in the cooling phase cannot be accurately described by a simple exponential or power-law decay when convection is present.

Following Ref. [85], we parametrize the neutrino luminosity as

Lν(t)=Cνt−ανexp⁡[−(t/τν)−nν],(4)L_\nu(t) = C_\nu t^{-\alpha_\nu}\exp\left[-\left(t/\tau_\nu\right)^{-n_\nu}\right], \tag*{(4)}

for each (anti)neutrino flavor ν=νe,νˉe,νx,νˉx\nu= \nu_e, \bar{\nu}_e, \nu_x, \bar{\nu}_x, with xx representing non-electron species. The parameters used in this work are taken from Tabs. V and VI of Ref. [85] for electron and non-electron (anti)neutrinos, respectively. We mention here that Tab. VI in Ref. [85] reports the parameter values for νμ\nu_\mu. However, as discussed in Ref. [85], the differences between νμ\nu_\mu and ντ\nu_\tau are small, and we therefore use νμ\nu_\mu and νˉμ\bar{\nu}_\mu as representative of νx\nu_x and νˉx\bar{\nu}_x.

To simplify the implementation of the time dependence of LνL_\nu in the nuclear-reaction network SkyNet [91], we assume a common time evolution for all (anti)neutrino species. Specifically, the parameters αν\alpha_\nu, τν\tau_\nu, and nνn_\nu are taken as the averages over all species. With this prescription, the optimal production window for the νp\nu p process occurs at t≃3.5 st \simeq3.5\ \mathrm{s} in our benchmark model. We therefore fix the normalization constants CνC_\nu in order to reproduce the luminosities of all species at t=3.5 st = 3.5\ \mathrm{s}. The fit parameters for LνL_\nu in the 1.93 M⊙1.93\,M_\odot model are listed in Table S1.

νe\nu_eνˉe\bar{\nu}_eνx\nu_xνˉx\bar{\nu}_x
CνC_\nu [105110^{51} erg/s]10.6111.1412.9013.56
αν\alpha_\nu0.5840.5840.5840.584
τν\tau_\nu [s]9.319.319.319.31
nνn_\nu3.953.953.953.95
⟨Eν⟩\langle E_\nu\rangle [MeV]10.8013.5813.1013.70
⟨Eν2⟩\langle E_\nu^2\rangle [MeV2^2]148.03235.25223.34244.98
TνT_\nu [MeV]3.214.094.144.36
ην\eta_\nu1.170.970.06−0.12-0.12

Table S1. Input parameters for neutrino luminosities (upper blocks) and energy spectra (lower blocks) of all (anti)neutrino species used for our benchmark 1.93 M⊙1.93\,M_\odot model.

The resulting time evolution of Lν(t)L_\nu(t) in our model is shown in the left panel of Figure S2, exhibiting the hierarchy Lνe<Lνˉe<Lνx<LνˉxL_{\nu_e}<L_{\bar{\nu}_e}<L_{\nu_x}<L_{\bar{\nu}_x} at any given time. The small difference between LνxL_{\nu_x} and LνˉxL_{\bar{\nu}_x} is mainly related to WM effects [55], which increase neutrino interaction rates and lower the rates for antineutrinos (see, e.g., the discussion in Ref. [89]). These effects also lead to small differences between the νx\nu_x and νˉx\bar{\nu}_x spectra, as discussed below.

Figure S2. Time evolution of the neutrino luminosities (left panel) and production spectra at t=3.5 st = 3.5\ \mathrm{s} (right panel) for νe\nu_e (solid black), νˉe\bar{\nu}_e (dashed black), νx\nu_x (solid red), νˉx\bar{\nu}_x (dashed red), in our 1.93 M⊙1.93\,M_\odot model.

Following the default option in SkyNet, in this analysis we model neutrino spectra with a normalized “pinched” Fermi-Dirac distribution [87, 88]

fνi(E)=NνieE/Tνi−ηνi+1,(5)f_{\nu_i}(E)=\frac{N_{\nu_i}}{e^{E/T_{\nu_i}-\eta_{\nu_i}}+1}, \tag*{(5)}

where the spectral parameters TνiT_{\nu_i} and ηνi\eta_{\nu_i} are obtained directly from the same SN simulation, and NνiN_{\nu_i} is fixed by the luminosity. This parametrization provides an accurate fit to simulated spectra [89] and enables the reconstruction of higher-order energy moments from the moments supplied by the simulations.

Since SN simulations provide only the first ⟨Eνi⟩\langle E_{\nu_i}\rangle and second moments ⟨Eνi2⟩\langle E_{\nu_i}^2\rangle of the energy distributions, we relate the effective neutrino temperature TνiT_{\nu_i} and the degeneracy parameter ηνi\eta_{\nu_i} to these quantities as

⟨Eνi⟩=TνiF3(ηνi)F2(ηνi),⟨Eνi2⟩⟨Eνi⟩=TνiF4(ηνi)F3(ηνi),(6)\langle E_{\nu_i}\rangle=T_{\nu_i}\frac{F_3(\eta_{\nu_i})}{F_2(\eta_{\nu_i})},\qquad\frac{\langle E_{\nu_i}^2\rangle}{\langle E_{\nu_i}\rangle}=T_{\nu_i}\frac{F_4(\eta_{\nu_i})}{F_3(\eta_{\nu_i})}, \tag*{(6)}

where Fn(ηνi)F_n(\eta_{\nu_i}) are the Fermi integrals defined as

Fn(ηνi)≡∫0∞xnex−ηνi+1 dx=−Γ(n+1)Li⁡n+1(−eηνi),(7)F_n(\eta_{\nu_i})\equiv\int_0^\infty\frac{x^n}{e^{x-\eta_{\nu_i}}+1}\,dx=-\Gamma(n+1)\operatorname{Li}_{n+1}(-e^{\eta_{\nu_i}}), \tag*{(7)}

where Li⁡n\operatorname{Li}_n is the polylogarithm of order nn.

Since ⟨Eνi⟩\langle E_{\nu_i}\rangle and ⟨Eνi2⟩\langle E_{\nu_i}^2\rangle vary only mildly during the first ∼5\sim5 s after core bounce (see, e.g., Figure 1 of Ref. [84]), we use time-independent spectral parameters TνiT_{\nu_i} and ηνi\eta_{\nu_i}. These are fixed by matching ⟨Eνi⟩\langle E_{\nu_i}\rangle and ⟨Eνi2⟩\langle E_{\nu_i}^2\rangle to the values extracted from the simulations at the reference time t=3.5t=3.5 s. The adopted values of ⟨Eνi⟩\langle E_{\nu_i}\rangle, ⟨Eνi2⟩\langle E_{\nu_i}^2\rangle, TνiT_{\nu_i}, and ηνi\eta_{\nu_i} for our benchmark model are reported in Table S1.

The right panel of Figure S2 shows the production spectra (in MeV−1^{-1} s−1^{-1}) for electron (black) and non-electron (red) neutrinos (solid lines) and antineutrinos (dashed lines) at t=3.5t=3.5 s, normalized to reproduce the corresponding luminosities at that time. For each flavor, neutrino spectra peak at lower energies than the corresponding antineutrino spectra, with ⟨Eνˉe⟩\langle E_{\bar{\nu}_e}\rangle about 25% larger than ⟨Eνe⟩\langle E_{\nu_e}\rangle and ⟨Eνˉx⟩\langle E_{\bar{\nu}_x}\rangle about 4% larger than ⟨Eνx⟩\langle E_{\nu_x}\rangle. The few-percent difference between the νx\nu_x and νˉx\bar{\nu}_x spectra arises from WM effects; in their absence the two spectra would be essentially identical.

With the assumed values of neutrino luminosity and spectra, the equilibrium value of YeY_e in the NDO (see Eq. (2)) is approximately 0.57.

Mixing of neutrino flavors

As a benchmark for the effect of collective neutrino oscillations, we consider complete flavor equilibration, which transforms distributions as

fνe′=fνx′=fνe+2fνx3,fνˉe′=fνˉx′=fνˉe+2fνˉx3,(8)f'_{\nu_e}=f'_{\nu_x}=\frac{f_{\nu_e}+2f_{\nu_x}}{3},\qquad f'_{\bar{\nu}_e}=f'_{\bar{\nu}_x}=\frac{f_{\bar{\nu}_e}+2f_{\bar{\nu}_x}}{3}, \tag*{(8)}

where, under the assumption of flavor equipartition, the mixed (anti)neutrino distribution function receives equal contributions from νe\nu_e (νˉe\bar{\nu}_e) and the individual heavy-lepton (anti)neutrinos. As discussed in the main text, our results remain valid provided that oscillations convert a substantial fraction of the heavy-lepton-flavor (anti)neutrino flux into electron (anti)neutrinos. Assuming the pre-mixing spectra are of Fermi-Dirac shape as reported in Eq. (5), the mixed luminosities are

Lνe′=Lνx′=Lνe+2Lνx3,Lνˉe′=Lνˉx′=Lνˉe+2Lνˉx3.(9)L'_{\nu_e} = L'_{\nu_x} = \frac{L_{\nu_e} + 2L_{\nu_x}}{3}, \qquad L'_{\bar{\nu}_e} = L'_{\bar{\nu}_x} = \frac{L_{\bar{\nu}_e} + 2L_{\bar{\nu}_x}}{3}. \tag*{(9)}

and the energy moments are

⟨Eνen⟩′=⟨Eνxn⟩′=NνeTνen+3Fn+2(ηνe)+2NνxTνxn+3Fn+2(ηνx)NνeTνe3F2(ηνe)+2NνxTνx3F2(ηνx).(10)\langle E^n_{\nu_e}\rangle' = \langle E^n_{\nu_x}\rangle' = \frac{N_{\nu_e}T_{\nu_e}^{n+3}F_{n+2}(\eta_{\nu_e}) + 2N_{\nu_x}T_{\nu_x}^{n+3}F_{n+2}(\eta_{\nu_x})}{N_{\nu_e}T_{\nu_e}^{3}F_{2}(\eta_{\nu_e}) + 2N_{\nu_x}T_{\nu_x}^{3}F_{2}(\eta_{\nu_x})}. \tag*{(10)}

with FnF_n as the Fermi-Dirac integral of power nn defined in Eq. (7). An analogous expression for ⟨Eνˉen⟩′=⟨Eνˉxn⟩′\langle E^n_{\bar{\nu}_e}\rangle' = \langle E^n_{\bar{\nu}_x}\rangle' follows by replacing να→νˉα\nu_\alpha\to\bar{\nu}_\alpha (α=e,x\alpha= e, x) everywhere in Eq. (10).

The left panel of Figure S3 shows the unmixed (black) and mixed (orange) production spectra of electron neutrinos (solid) and antineutrinos (dashed) at t=3.5 st = 3.5\,\mathrm{s}. The contribution of non-electron (anti)neutrinos—whose spectra are more similar (see the right panel of Figure S2)—reduces the difference between neutrino and antineutrino spectra after mixing. The resulting mixed luminosities and spectra yield Ye≈0.6Y_e \approx0.6 via Eq. (2) and are used in the outflow calculation described in SM A.

Figure S3. Left: Production spectra at t=3.5 st = 3.5\,\mathrm{s} for unmixed (black) and mixed (orange) electron neutrinos (solid curves) and antineutrinos (dashed curves), in our 1.93 M⊙1.93\,M_\odot model. Right: Ratio of neutrino to antineutrino spectra for the unmixed electron flavor (black) and non-electron flavors (red), as well as for the electron flavor in the mixed case (orange) in the 1.93 M⊙1.93\,M_\odot model. The vertical orange lines correspond to the average energy ⟨E⟩\langle E\rangle (solid) and its inverse ratio with the second ϵ=⟨E2⟩/⟨E⟩\epsilon= \langle E^2\rangle/\langle E\rangle (dashed) and third ε=⟨E3⟩/⟨E⟩\varepsilon= \sqrt{\langle E^3\rangle/\langle E\rangle} (dotted) moments of the mixed electron antineutrino’s energy distribution.

A more quantitative comparison is obtained by considering the ratio of neutrino to antineutrino spectra, shown in the right panel of Figure S3 for the unmixed electron flavor (black), non-electron flavors (red), and the mixed electron flavor (orange). In the mixed case, the spectrum receives a 1/3 contribution from electron flavors and 2/3 from non-electron species, so the ratio lies between the unmixed cases. Vertical lines indicate the average energy ⟨Eνe⟩\langle E_{\nu_e}\rangle (solid), as well as the parameters ϵν\epsilon_{\nu} (dashed) and εν\varepsilon_{\nu} (dotted) for the mixed antineutrino spectra. These quantities enter the determination of Q˙\dot{Q} and YeY_e in Eq. (2). Due to the similarity between νx\nu_x and νˉx\bar{\nu}_x spectra, the ratio of mixed νe\nu_e to νˉe\bar{\nu}_e spectra is close to unity near ϵ\epsilon and ε\varepsilon, explaining why Ye≈0.6Y_e \approx0.6 is obtained.

Larger values of ϵν\epsilon_{\nu} enhance the neutrino net heating rate Q˙\dot{Q} and consequently M˙\dot{M}. The left panel of Figure S4 shows the radial profile of Q˙\dot{Q} at t=3.5 st = 3.5\ \mathrm{s} near the PNS for our benchmark model, for flavor mixing at different distances Δrmix\Delta r_{\mathrm{mix}} from the PNS surface. For each mixing scenario, deviations from the unmixed case are negligible at r<RPNS+Δrmixr < R_{\mathrm{PNS}}+\Delta r_{\mathrm{mix}}, while at larger radii Q˙\dot{Q} increases by ∼30%\sim30\% due to higher neutrino luminosities and harder spectra. Enhanced Q˙\dot{Q} closer to the PNS accelerates the outflow and increases M˙\dot{M}.

Image not extracted

Figure S4. Left: Neutrino net heating rate Q˙\dot{Q} as a function of radius rr at t=3.5 st = 3.5\ \mathrm{s} for flavor conversion at different distances Δrmix\Delta r_{\mathrm{mix}} from the PNS surface. Right: Relative enhancement of the mass outflow rate M˙\dot{M} as a function of launch time tlauncht_{\mathrm{launch}} for different mixing radii Δrmix\Delta r_{\mathrm{mix}}. Vertical lines denote the beginning of the transonic case for Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\ \mathrm{km} (solid), 5 km5\ \mathrm{km} (dashed) and 10 km10\ \mathrm{km} (dash-dotted). Color coding follows Figure 3.

The right panel of Figure S4 shows the time evolution of M˙\dot{M} enhancement relative to the unmixed case for different mixing radii Δrmix\Delta r_{\mathrm{mix}}. The relative enhancement decreases with time: it is ≲30%\lesssim30\% for Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\ \mathrm{km}, ≲10%\lesssim10\% for Δrmix=5 km\Delta r_{\mathrm{mix}} = 5\ \mathrm{km}, and <5%< 5\% for larger mixing radii. The increased Q˙\dot{Q} also raises the outflow velocity, inducing a transonic transition at Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\ \mathrm{km} (vertical solid line), 5 km5\ \mathrm{km} (dashed line), and 10 km10\ \mathrm{km} (dash-dotted line). For Δrmix≲10 km\Delta r_{\mathrm{mix}} \lesssim10\ \mathrm{km}, the transonic transition occurs at progressively later times as Δrmix\Delta r_{\mathrm{mix}} increases. For Δrmix≳10 km\Delta r_{\mathrm{mix}} \gtrsim10\ \mathrm{km}, the transonic transition disappears and the velocity enhancement remains below 10%10\% relative to the unmixed case.

Integrated yields and production factors

To compare νp\nu p yields with solar-system abundances [100], we compute time-integrated yields. For a nuclide (A,Z)(A,Z), the time-averaged abundance is defined as [80]

⟨YA,Z⟩=∫dt YA,Z(t)M˙(t)∫dt M˙(t)(11)\langle Y_{A,Z}\rangle= \frac{\int dt\,Y_{A,Z}(t)\dot{M}(t)}{\int dt\,\dot{M}(t)} \tag*{(11)}

where YA,Z(t)Y_{A,Z}(t) and M˙(t)\dot{M}(t) denote the instantaneous abundance and mass outflow rate, respectively. The resulting time-averaged yields ⟨YA⟩=∑ZYA,Z\langle Y_A\rangle= \sum_Z Y_{A,Z} are shown for the benchmark 20 M⊙20\ M_{\odot} model in Figure 3 and for the 18.6 M⊙18.6\ M_{\odot} model in the left panel of Figure S8.

Figure S8. Left: Time-averaged yields ⟨YA⟩\langle Y_A\rangle for the 18.6 M⊙M_\odot model under flavor conversion occurring at different radii from the PNS. Curves show ⟨YA⟩\langle Y_A\rangle versus mass number AA; diamonds mark the corresponding 92Nb{}^{92}\mathrm{Nb} yield. Black arrows indicate the minimum abundances needed to reproduce the solar values. Color coding follows Figure 3. Right: Production factors of pp-nuclides in the 18.6 M⊙M_\odot model, for the unmixed case (upper right panel) and flavor mixing occurring 1 km above the PNS (Δrmix=1\Delta r_{\rm mix}=1 km—lower right panel). In each panel, the cyan band indicates the co-production region, and the dashed orange line marks the threshold corresponding to an overproduction factor of 10, required to account for solar abundances.

Time-averaged yields increase as mixing occurs closer to the PNS, as discussed in the main text. Figure S5 shows enhancement factors ⟨YA⟩/⟨YA⟩unmixed\langle Y_A\rangle/\langle Y_A\rangle_{\mathrm{unmixed}} for pp-nuclides (curves) and 92Nb{}^{92}\mathrm{Nb} (diamonds) under near (left) and far (right) conversions. The enhancement grows with mass number AA in all scenarios.

Figure S5. Enhancement factors ⟨YA⟩/⟨YA⟩unmixed\langle Y_A\rangle/\langle Y_A\rangle^{\mathrm{unmixed}} for time-averaged yields under flavor equilibration at various radii, for near (left) and far (right) conversions. Curves show enhancement versus mass number AA; diamonds mark 92Nb{}^{92}\mathrm{Nb} enhancement. Vertical lines denote 92Mo{}^{92}\mathrm{Mo} (solid red), 94Mo{}^{94}\mathrm{Mo} (dashed red), 96Ru{}^{96}\mathrm{Ru} (solid blue), and 98Ru{}^{98}\mathrm{Ru} (dashed blue). Color coding follows Figure 3.

For near conversions, the Δrmix=1,5 km\Delta r_{\mathrm{mix}} = 1, 5\ \mathrm{km} cases nearly coincide, yielding factors of ∼10\sim10 for 92Mo{}^{92}\mathrm{Mo} and ∼60\sim60 for 98Ru{}^{98}\mathrm{Ru}. At Δrmix=10 km\Delta r_{\mathrm{mix}} = 10\ \mathrm{km}, enhancements remain within a factor of 2 of these values: ∼8\sim8 for 92Mo{}^{92}\mathrm{Mo} and ∼30\sim30 for 98Ru{}^{98}\mathrm{Ru}. The Δrmix=20,30 km\Delta r_{\mathrm{mix}} = 20, 30\ \mathrm{km} cases produce smaller boosts: ≲10\lesssim10 and ≲5\lesssim5, respectively.

Far conversions yield more modest increases. For Δrmix=40\Delta r_{\mathrm{mix}} = 40–300 km300\ \mathrm{km}, enhancement factors remain ≲3\lesssim3. At larger radii, where conversion occurs during the pp-nuclide production phase, enhancement diminishes further (dotted purple line, Δrmix=1000 km\Delta r_{\mathrm{mix}} = 1000\ \mathrm{km}) and becomes negligible for Δrmix≳3000 km\Delta r_{\mathrm{mix}} \gtrsim3000\ \mathrm{km} (dashed violet line).

The 92Nb{}^{92}\mathrm{Nb} increase differs qualitatively. Near conversion enhances its abundance by up to 2 orders of magnitude, with factors of ∼50\sim50 for Δrmix=1,5 km\Delta r_{\mathrm{mix}} = 1, 5\ \mathrm{km}. Far conversion produces a factor of ∼3\sim3 increase at Δrmix=40,50 km\Delta r_{\mathrm{mix}} = 40, 50\ \mathrm{km}, and enhancement persists even beyond 1000 km, reaching 1.4 at Δrmix=3000 km\Delta r_{\mathrm{mix}} = 3000\ \mathrm{km} (violet diamond).

From these abundances we define the isotopic production factor [79, 80, 81]

fA,Z=⟨XA,Z⟩XA,Z⊙,(12)f_{A,Z} = \frac{\langle X_{A,Z}\rangle}{X^\odot_{A,Z}}, \tag*{(12)}

where XA,Z⊙X^\odot_{A,Z} is the solar-system mass fraction of isotope (A,Z)(A, Z), and ⟨XA,Z⟩=A⟨YA,Z⟩\langle X_{A,Z}\rangle= A\langle Y_{A,Z}\rangle is the time-averaged mass fraction obtained from the SkyNet output, with ∑A,ZAYA,Z=1\sum_{A,Z} A Y_{A,Z} = 1. Solar abundances are taken from meteoritic measurements of carbonaceous chondrites [100], following Ref. [81].

Production factors quantify both relative and absolute abundances. Accounting for astrophysical uncertainties, nuclides are considered co-produced if fA,Z≳fmax⁡/10f_{A,Z} \gtrsim f_{\max}/10, where fmax⁡f_{\max} is the largest production factor in a given model [79, 128]. Absolute yields are characterized by the overproduction factor [79, 80, 129]

fA,Zov=fA,ZMoutMejec,(13)f^{\mathrm{ov}}_{A,Z} = f_{A,Z}\frac{M_{\mathrm{out}}}{M_{\mathrm{ejec}}}, \tag*{(13)}

where the ratio (Mout/Mejec)(M_{\mathrm{out}/M_{\mathrm{ejec}}}) is the astrophysical dilution factor, with Mout=∫dt M˙(t)M_{\mathrm{out}} = \int dt\,\dot{M}(t) the total mass driven out in the NDO and Mejec≃Mprog−MPNSM_{\mathrm{ejec}} \simeq M_{\mathrm{prog}} - M_{\mathrm{PNS}} the total mass ejected in the explosion. Values fA,Zov≳10f^{\mathrm{ov}}_{A,Z} \gtrsim10 are typically required to reproduce solar abundances [79, 80, 129]. The ejected mass MejecM_{\mathrm{ejec}} is fixed by the underlying SN simulation and is therefore independent of the mixing prescription, with Mejec=18.07 M⊙M_{\mathrm{ejec}} = 18.07\,M_\odot for the 20 M⊙20\,M_\odot progenitor and Mejec=16.67 M⊙M_{\mathrm{ejec}} = 16.67\,M_\odot for the 18.6 M⊙18.6\,M_\odot progenitor. Flavor mixing increases neutrino heating in the hot bubble and thus raises the mass-outflow rate M˙\dot{M}. As a result, flavor equilibration occurring closer to the PNS yields a larger total outflow mass, MoutM_{\mathrm{out}}, and a greater dilution factor. For the 20 M⊙20\,M_\odot model, we find Mout≃3.5×10−3 M⊙M_{\mathrm{out}} \simeq3.5 \times10^{-3}\,M_\odot (dilution factor ≃2.0×10−4\simeq2.0 \times10^{-4}) without mixing, rising to Mout≃4.4×10−3 M⊙M_{\mathrm{out}} \simeq4.4 \times10^{-3}\,M_\odot (dilution factor ≃2.4×10−4\simeq2.4 \times10^{-4}) for the Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\ \mathrm{km} case. For the 18.6 M⊙18.6\,M_\odot model, we obtain Mout≃3.6×10−3 M⊙M_{\mathrm{out}} \simeq3.6 \times10^{-3}\,M_\odot (dilution factor ≃2.1×10−4\simeq2.1 \times10^{-4}) in the unmixed case, and Mout≃4.4×10−3 M⊙M_{\mathrm{out}} \simeq4.4 \times10^{-3}\,M_\odot (dilution factor ≃2.6×10−4\simeq2.6 \times10^{-4}) for Δrmix=1 km\Delta r_{\mathrm{mix}} = 1\ \mathrm{km}.

We show the production factors of different pp-nuclides for our benchmark 20 M⊙20\,M_\odot model in Figure S6 for mixing occurring at different distances Δrmix\Delta r_{\mathrm{mix}} from the PNS surface, comparing “far” conversion (left panel) and “near” conversion (right panel). Each panel also displays the production factor of the extinct radionuclide 92Nb{}^{92}\mathrm{Nb} (green diamond), normalized to 3×10−33 \times10^{-3} of the 92Mo{}^{92}\mathrm{Mo} solar abundance. The associated error bars reflect the inferred production ratio 92Nb/92Mo^{92}\mathrm{Nb}/^{92}\mathrm{Mo}, constrained to lie in the range ∼10−3\sim10^{-3}–10−210^{-2} to account for the solar-system abundance of 92Zr^{92}\mathrm{Zr}, into which 92Nb^{92}\mathrm{Nb} decays [61, 130, 131] [112].

Figure S6. Production factors of pp-nuclides (purple dots) and 92Nb^{92}\mathrm{Nb} (green diamond with error bars) in the 20 M⊙M_{\odot} model, under flavor conversion occurring at different radii Δrmix\Delta r_{\mathrm{mix}} from the PNS. In each panel, the cyan band indicates the co-production region, and the dashed orange line marks the threshold required to satisfy solar observations. The uncertainty on f(92Nb)f(^{92}\mathrm{Nb}) arises because 92Nb^{92}\mathrm{Nb} is an extinct radionuclide: its solar abundance cannot be measured directly, but is inferred from meteoritic data to correspond to a production fraction 92Nb/92Mo^{92}\mathrm{Nb}/^{92}\mathrm{Mo} within ∼10−3\sim10^{-3}–10−210^{-2}.

Production factors increase as mixing occurs closer to the PNS. For far conversion, Mo and Ru isotope production factors increase by more than a factor of 2 from the unmixed case to Δrmix=40\Delta r_{\mathrm{mix}} = 40 km, while results at Δrmix=300\Delta r_{\mathrm{mix}} = 300 km and 40 km are the same. For near conversion, production factors increase monotonically from Δrmix=30\Delta r_{\mathrm{mix}} = 30 km to 1 km due to enhanced YeY_e and M˙\dot{M}.

In all cases, 84Sr^{84}\mathrm{Sr} has the highest production factor. The co-production band (cyan) includes all isotopes with fA,Z≥0.1f(84Sr)f_{A,Z} \ge0.1 f(^{84}\mathrm{Sr}). Without mixing, all species with A≥90A \ge90 fall below this threshold. The increase induced by far conversion is not enough to reach the co-production region. For Δrmix≲30\Delta r_{\mathrm{mix}} \lesssim30 km, production factors approach co-production, and for Δrmix≲10\Delta r_{\mathrm{mix}} \lesssim10 km, isotopes up to 102Pd^{102}\mathrm{Pd} are co-produced. For Δrmix=1\Delta r_{\mathrm{mix}} = 1 km, flavor mixing increases Mo and Ru isotopes by a factor ∼10\sim10, 92Nb^{92}\mathrm{Nb} by ∼100\sim100, and enhances 102Pd^{102}\mathrm{Pd} and 106Cd^{106}\mathrm{Cd} by ∼2\sim2 and more than 3 orders of magnitude, respectively.

The horizontal dashed orange line marks the threshold fA,Zov=10f^{\mathrm{ov}}_{A,Z}=10 required to reproduce solar abundances [Eq. (13)]. Mixing increases M˙\dot{M} through enhanced heating, yielding larger MoutM_{\mathrm{out}} and thus slightly lowering this threshold for smaller Δrmix\Delta r_{\mathrm{mix}}. In the unmixed scenario, nuclides with 90≤A≤10090 \le A \le100 are neither co-produced nor sufficiently abundant to meet solar constraints. Far conversion cannot reach the solar threshold (all nuclides remain below the orange line), while Mo and Ru isotopes satisfy solar observations for Δrmix≲30\Delta r_{\mathrm{mix}} \lesssim30 km. For Δrmix≲10\Delta r_{\mathrm{mix}} \lesssim10 km, Mo and Ru isotopes—as well as 92Nb^{92}\mathrm{Nb} within uncertainties—are co-produced and meet solar-abundance requirements.

From the overproduction factor, we derive the solar-abundance thresholds shown in Figure 3 and Figure S8 by inverting the condition fA,Zov≳10f^{\mathrm{ov}}_{A,Z} \gtrsim10 [79, 80, 129]:

⟨YA⟩thr=10MejecMoutXA,Z⊙A.(14)\langle Y_A \rangle_{\mathrm{thr}} = 10 \frac{M_{\mathrm{ejec}}}{M_{\mathrm{out}}} \frac{X^\odot_{A,Z}}{A}. \tag*{(14)}

For 92Nb^{92}\mathrm{Nb}, the minimum threshold corresponds to normalizing its production to 10−310^{-3} of the 92Mo^{92}\mathrm{Mo} abundance, while the black dot denotes the benchmark normalization of 3×10−33 \times10^{-3}.

To remain conservative, we always show ⟨YA⟩thr\langle Y_A\rangle_{\mathrm{thr}} only for the unmixed case, which yields the largest thresholds relative to the corresponding mixed scenarios.

Mildly-supersonic models

A natural question one has, after seeing the results of our benchmark 20 M⊙20\,M_\odot model, is whether the conditions of this simulation, albeit taken off the shelf, are “fine-tuned” or coincidental so that when mixing is introduced, the supersonic transition enters only after the nucleosynthesis has mostly concluded. Had the transition occurred earlier and nucleosynthesis operated in a supersonic environment, there could be a real competition between the favorable influence of increased YeY_e and the adverse effect of the supersonic regime.

To this end, we perturb the far boundary condition of our benchmark model by adopting a progenitor of 18.6 M⊙18.6\,M_\odot with vFS=7000v_{\mathrm{FS}}=7000 km/s, while retaining the PNS properties and neutrino spectra. This lowers the confining pressure of the hot bubble and increases the outflow’s velocity, inducing a supersonic transition within the first seconds of the cooling phase without the introduction of neutrino oscillations. Considering such a mildly supersonic model affords us insights into the other side of near-critical supernova outflows.

Figure S7 compares the time evolution of the benchmark 20 M⊙20\,M_\odot model (left panels) and the perturbed 18.6 M⊙18.6\,M_\odot model (right panels) during the cooling phase (tlaunch∈[1,10]t_{\mathrm{launch}} \in[1,10] s). The upper panels show the evolution of the outflows in the unmixed (panels a1a_1 and b1b_1) and in the Δrmix=1\Delta r_{\mathrm{mix}}=1 km cases (panels c1c_1 and d1d_1), while the lower panels display the corresponding production rates as a function of the launch time. In contrast to the benchmark model, this 18.6 M⊙18.6\,M_\odot (panel b1b_1) model exhibits a transonic transition at tlaunch≃3.3t_{\mathrm{launch}} \simeq3.3 s, followed by a return to subsonic outflows at t≃9.2t \simeq9.2 s with the decline of neutrino luminosities. Flavor equilibration accelerates the onset of this transonic transition, which takes place at t≃2.2t \simeq2.2 s when mixing occurs 1 km above the PNS surface (panel d1d_1). This presents an interesting study as the induced supersonic regime now intrudes on the early cooling phase, when the νp\nu p process is most efficient.

Figure S7. Upper panels: Radial temperature profiles for outflows launched at tlaunch∈[1,10]t_{\mathrm{launch}} \in[1,10] s, depicted here in steps of 0.5 s. The green and yellow bands correspond to the seed-formation and pp-nuclide formation stages of the νp\nu p-process, respectively [79, 81]. Upper panels: Time evolution of the production rates of Mo (red) and Ru isotopes (blue), and 92Nb{}^{92}\mathrm{Nb} (black), with vertical dotted lines delimiting the time window where the outflows contain a supersonic transition. In each set of plots, the left (right) panels refer to the 20 M⊙M_{\odot} (18.6 M⊙M_{\odot}) progenitor, while the upper (lower) panels are dedicated to the unmixed (Δrmix=1\Delta r_{\mathrm{mix}} = 1 km) case.

What is the impact on nucleosynthesis? As shown in the lower panels of Figure S7, without mixing, the production rates in the optimal window (prior to the transonic transition at t≃3.3t \simeq3.3 s) are smaller by a factor of ∼2\sim2 compared to the benchmark model (panel a2a_2). When mixing occurs 1 km above the PNS, the supersonic transition occurs near the end of the optimal production window in the 20 M⊙20\,M_\odot model (panel c2c_2), whereas in this 18.6 M⊙18.6\,M_\odot model (panel d2d_2) it clips the synthesis of pp-nuclides, preventing the νp\nu p process from attaining its peak rates.

Despite this hydrodynamic effect, in both models mixing at Δrmix=1\Delta r_{\mathrm{mix}}=1 km enhances the production rates by a factor of ∼20\sim20 for Mo and ∼50\sim50 for Ru isotopes during the optimal window. The ∼1\sim1 s reduction in the duration of this window is subdominant relative to this enhancement from the increased YeY_e and M˙\dot{M}. Thus, in both the 20 M⊙20\,M_\odot and 18.6 M⊙18.6\,M_\odot models, effective neutrino mixing near the PNS renders higher nucleosynthetic yields. The production of 92Nb^{92}\mathrm{Nb} (black lines) is similarly improved in both models, with flavor equilibration at Δrmix=1\Delta r_{\mathrm{mix}}=1 km extending the production window up to ∼10\sim10 s.

The time-integrated results are reported in Figure S8. Its left panel shows the time-averaged abundances for the 18.6 M⊙18.6\,M_\odot model, comparing the unmixed case (solid black line) with scenarios of varying radii of flavor equilibration, given by the displacement from the PNS radius Δrmix\Delta r_{\mathrm{mix}}. The qualitative behavior is similar to that of the benchmark 20 M⊙20\,M_\odot model (Figure 3): mixing magnifies νp\nu p yields, generally as the equilibration radius approaches the PNS. For Δrmix=20\Delta r_{\mathrm{mix}}=20–3030 km, yields increase by less than one order of magnitude, while when flavor conversion occurs sufficiently close to the PNS (Δrmix≲10\Delta r_{\mathrm{mix}} \lesssim10 km), νp\nu p yields increase by up to a factor of ∼100\sim100 relative to their original values. Notably, the Δrmix=5\Delta r_{\mathrm{mix}}=5 km case produces slightly larger yields than the Δrmix=1\Delta r_{\mathrm{mix}}=1 km scenario. This occurs because the earlier transonic transition in the Δrmix=1\Delta r_{\mathrm{mix}} = 1 km case partially counteracts the larger YeY_{\mathrm{e}} boost, resulting in a net reduction compared to Δrmix=5\Delta r_{\mathrm{mix}} = 5 km. Overall, the yields from the 18.6 M⊙M_{\odot} model are systematically lower by about one order of magnitude compared to the corresponding cases in the benchmark model. Consequently, the time-averaged yields of 92,94Mo{}^{92,94}\mathrm{Mo} and 96,98Ru{}^{96,98}\mathrm{Ru} exceed the solar-abundance thresholds (blue arrows) only if mixing occurs at Δrmix≲10\Delta r_{\mathrm{mix}} \lesssim10 km, i.e., at smaller radii than for the 20 M⊙M_{\odot} case, where Δrmix≲20\Delta r_{\mathrm{mix}} \lesssim20 km is sufficient. A similar conclusion applies to 92Nb{}^{92}\mathrm{Nb} production. In the absence of neutrino mixing, the predicted abundance lies well below the solar threshold (see the black diamond in the left panel of Figure S8), whereas mixing at small radii (see, e.g., the red diamond corresponding to Δrmix=1\Delta r_{\mathrm{mix}} = 1 km) brings agreement with observations.

Finally, the right panel of Figure S8 shows the production factors of the pp-nuclides (purple dots) and 92Nb{}^{92}\mathrm{Nb} (green diamond with error bars), as defined in SM D, for the unmixed case (upper panel) and for mixing at Δrmix=1\Delta r_{\rm mix}=1 km (lower panel). As in the benchmark model, 84Sr{}^{84}\mathrm{Sr} exhibits the largest production factor. Without neutrino oscillations, νp\nu p yields beyond 84Sr{}^{84}\mathrm{Sr} are strongly suppressed, while with flavor equilibration at Δrmix=1\Delta r_{\rm mix}=1 km, 92,94Mo{}^{92,94}\mathrm{Mo} and 96,98Ru{}^{96,98}\mathrm{Ru} lie within a factor of ∼2\sim2 below the co-production band (cyan region), indicating non-negligible production within uncertainties. The horizontal dashed orange line denotes the threshold required to reproduce solar abundances, corresponding to fA,Zov=10f^{\rm ov}_{A,Z}=10 (see Eq. (13) and the discussion below). Mixing enhances the mass outflow rate M˙\dot{M} through increased heating, leading to a larger MoutM_{\rm out} and thus a slightly lower threshold in the Δrmix=1\Delta r_{\rm mix}=1 km case relative to the unmixed scenario (see SM D). Overall, production factors in the 18.6 M⊙M_\odot model are smaller than in the benchmark 20 M⊙M_\odot case (see Figure S6), implying that pp-nuclides with A>90A>90 are severely underproduced without mixing, while with mixing only species up to 98Ru{}^{98}\mathrm{Ru} exceed the solar threshold.

In summary, the 18.6 M⊙M_\odot model has reduced yields relative to the benchmark, yet effective neutrino mixing still supports efficient νp\nu p process. Mo and Ru isotopes are produced at non-negligible levels—just below the co-production band yet above the solar threshold. In both models, mixing salvages the otherwise underproduced 92Nb{}^{92}\mathrm{Nb}, putting it close to the co-production band (within ∼1σ\sim1\sigma) and satisfying observational constraints.

References

  1. [1]J. T. Pantaleone, Neutrino oscillations at high densities, Phys. Lett. B287 (1992) 128.DOI
  2. [2]G. Sigl and G. Raffelt, General kinetic description of relativistic mixed neutrinos, Nucl. Phys. B406 (1993) 423.
  3. [3]B. H. J. McKellar and M. J. Thomson, Oscillating doublet neutrinos in the early universe, Phys. Rev. D49 (1994) 2710.pubmed.ncbi.nlm.nih.gov/10017264
  4. [4]S. Samuel, Neutrino oscillations in dense neutrino gases, Phys. Rev. D 48 (1993) 1462.pubmed.ncbi.nlm.nih.gov/10016384
  5. [5]Y. Z. Qian and G. M. Fuller, Neutrino-neutrino scattering and matter enhanced neutrino flavor transformation in Supernovae, Phys. Rev. D51 (1995) 1479 [astro-ph/9406073].
  6. [6]H. Duan, G. M. Fuller and Y.-Z. Qian, Collective neutrino oscillations, Annual Review of Nuclear and Particle Science 60 (2010) 569.
  7. [7]A. Mirizzi, I. Tamborra, H.-T. Janka, N. Saviano, K. Scholberg, R. Bollig, L. Hüdepohl and S. Chakraborty, Supernova Neutrinos: Production, Oscillations and Detection, Riv. Nuovo Cim. 39 (2016) 1 [1508.00785].DOI
  8. [8]S. Chakraborty, R. Hansen, I. Izaguirre and G. Raffelt, Collective neutrino flavor conversion: Recent developments, Nucl. Phys. B 908 (2016) 366 [1602.02766].DOI
  9. [9]I. Tamborra and S. Shalgar, New Developments in Flavor Evolution of a Dense Neutrino Gas, Ann. Rev. Nucl. Part. Sci. 71 (2021) 165 [2011.01948].DOI
  10. [10]F. Capozzi and N. Saviano, Neutrino Flavor Conversions in High-Density Astrophysical and Cosmological Environments, Universe 8 (2022) 94 [2202.02494].DOI
  11. [11]S. Richers and M. Sen, Fast Flavor Transformations, in Handbook of Nuclear Physics (I. Tanihata, H. Toki and T. Kajino, eds.), pp. 1–17. Springer, 2023. 2207.03561. DOI.
  12. [12]M. C. Volpe, Neutrinos from dense environments: Flavor mechanisms, theoretical approaches, observations, and new directions, Rev. Mod. Phys. 96 (2024) 025004 [2301.11814].arxiv.org/abs/2301.11814
  13. [13]L. Johns, S. Richers and M.-R. Wu, Neutrino Oscillations in Core-Collapse Supernovae and Neutron Star Mergers, Ann. Rev. Nucl. Part. Sci. 75 (2025) 399 [2503.05959].DOI
  14. [14]R. F. Sawyer, Speed-up of neutrino transformations in a supernova environment, Phys. Rev. D 72 (2005) 045003 [hep-ph/0503013].DOI
  15. [15]R. F. Sawyer, The multi-angle instability in dense neutrino systems, Phys. Rev. D 79 (2009) 105003 [0803.4319].DOI
  16. [16]R. F. Sawyer, Neutrino cloud instabilities just above the neutrino sphere of a supernova, Phys. Rev. Lett. 116 (2016) 081101 [1509.03323].DOI
  17. [17]B. Dasgupta, A. Mirizzi and M. Sen, Fast neutrino flavor conversions near the supernova core with realistic flavor-dependent angular distributions, JCAP 02 (2017) 019 [1609.00528].DOI
  18. [18]I. Izaguirre, G. Raffelt and I. Tamborra, Fast Pairwise Conversion of Supernova Neutrinos: A Dispersion-Relation Approach, Phys. Rev. Lett. 118 (2017) 021101 [1610.01612].DOI
  19. [19]S. Chakraborty, R. S. Hansen, I. Izaguirre and G. Raffelt, Self-induced flavor conversion of supernova neutrinos on small scales, JCAP 01 (2016) 028 [1507.07569].DOI
  20. [20]S. Chakraborty, R. S. Hansen, I. Izaguirre and G. Raffelt, Self-induced neutrino flavor conversion without flavor mixing, JCAP 03 (2016) 042 [1602.00698].DOI
  21. [21]H. Nagakura, T. Morinaga, C. Kato and S. Yamada, Fast-pairwise collective neutrino oscillations associated with asymmetric neutrino emissions in core-collapse supernova, 1910.04288.
  22. [22]T. Morinaga, H. Nagakura, C. Kato and S. Yamada, Fast neutrino-flavor conversion in the preshock region of core-collapse supernovae, Phys. Rev. Res. 2 (2020) 012046 [1909.13131].DOI
  23. [23]S. Abbar, H. Duan, K. Sumiyoshi, T. Takiwaki and M. C. Volpe, Fast Neutrino Flavor Conversion Modes in Multidimensional Core-collapse Supernova Models: the Role of the Asymmetric Neutrino Distributions, Phys. Rev. D 101 (2020) 043016 [1911.01983].DOI
  24. [24]R. Glas, H. T. Janka, F. Capozzi, M. Sen, B. Dasgupta, A. Mirizzi and G. Sigl, Fast Neutrino Flavor Instability in the Neutron-star Convection Layer of Three-dimensional Supernova Models, Phys. Rev. D 101 (2020) 063001 [1912.00274].DOI
  25. [25]H. Nagakura, L. Johns, A. Burrows and G. M. Fuller, Where, when, and why: Occurrence of fast-pairwise collective neutrino oscillation in three-dimensional core-collapse supernova models, Phys. Rev. D 104 (2021) 083025 [2108.07281].DOI
  26. [26]I. Padilla-Gay, I. Tamborra and G. G. Raffelt, Neutrino Flavor Pendulum Reloaded: The Case of Fast Pairwise Conversion, Phys. Rev. Lett. 128 (2022) 121102 [2109.14627].
  27. [27]M. Zaizen and H. Nagakura, Simple method for determining asymptotic states of fast neutrino-flavor conversion, Phys. Rev. D 107 (2023) 103022 [2211.09343].DOI
  28. [28]H. Nagakura and M. Zaizen, Connecting small-scale to large-scale structures of fast neutrino-flavor conversion, Phys. Rev. D 107 (2023) 063033 [2211.01398].
  29. [29]Z. Xiong, M.-R. Wu, M. George, C.-Y. Lin, N. K. Largani, T. Fischer and G. Martínez-Pinedo, Fast neutrino flavor conversions in a supernova: Emergence, evolution, and effects, Phys. Rev. D 109 (2024) 123008 [2402.19525].DOI
  30. [30]D. F. G. Fiorillo and G. G. Raffelt, Theory of neutrino slow flavor evolution. Part I. Homogeneous medium, JHEP 04 (2025) 146 [2412.02747].DOI
  31. [31]D. F. G. Fiorillo and G. G. Raffelt, Theory of neutrino slow flavor evolution. Part II. Space-time evolution of linear instabilities, JHEP 06 (2025) 146 [2501.16423].arxiv.org/abs/2501.16423
  32. [32]V. A. Kostelecky and S. Samuel, Self-maintained coherent oscillations in dense neutrino gases, Phys. Rev. D 52 (1995) 621 [hep-ph/9506262].DOI
  33. [33]S. Samuel, Bimodal coherence in dense selfinteracting neutrino gases, Phys. Rev. D 53 (1996) 5382 [hep-ph/9604341].
  34. [34]S. Pastor and G. Raffelt, Flavor oscillations in the supernova hot bubble region: Nonlinear effects of the neutrino background, Phys. Rev. Lett. 89 (2002) 191101 [astro-ph/0207281].DOI
  35. [35]H. Duan, G. M. Fuller and Y.-Z. Qian, Collective neutrino flavor transformation in supernovae, Phys. Rev. D 74 (2006) 123004 [astro-ph/0511275].DOI
  36. [36]H. Duan, G. M. Fuller, J. Carlson and Y.-Z. Qian, Simulation of Coherent Non-Linear Neutrino Flavor Transformation in the Supernova Environment. 1. Correlated Neutrino Trajectories, Phys. Rev. D 74 (2006) 105014 [astro-ph/0606616].
  37. [37]H. Duan, G. M. Fuller, J. Carlson and Y.-Z. Qian, Coherent Development of Flavor in the Supernova Environment, Phys. Rev. Lett. 97 (2006) 241101 [astro-ph/0608050].DOI
  38. [38]S. Hannestad, G. G. Raffelt, G. Sigl and Y. Y. Y. Wong, Self-induced conversion in dense neutrino gases: Pendulum in flavour space, Phys. Rev. D 74 (2006) 105010 [astro-ph/0608695]. [Erratum: Phys.Rev.D 76, 029901 (2007)].arxiv.org/abs/astro-ph/0608695
  39. [39]B. Dasgupta and A. Dighe, Collective three-flavor oscillations of supernova neutrinos, Phys. Rev. D 77 (2008) 103002 [0712.3798].
  40. [40]H. Duan, G. M. Fuller and Y.-Z. Qian, Stepwise spectral swapping with three neutrino flavors, Phys. Rev. D 77 (2008) 085016 [0801.1363].DOI
  41. [41]A. Friedland, Self-refraction of supernova neutrinos: mixed spectra and three-flavor instabilities, Phys. Rev. Lett. 104 (2010) 191102 [1001.0996].DOI
  42. [42]H. Duan and A. Friedland, Self-induced suppression of collective neutrino oscillations in a supernova, Phys. Rev. Lett. 106 (2011) 091101 [1006.2359].pubmed.ncbi.nlm.nih.gov/21405613
  43. [43]I. Padilla-Gay, H.-H. Chen, S. Abbar, M.-R. Wu and Z. Xiong, Flavor equilibration of supernova neutrinos: Exploring the dynamics of slow modes, Phys. Rev. D 112 (2025) 043039 [2505.11588].
  44. [44]D. F. G. Fiorillo and G. G. Raffelt, Theory of neutrino fast flavor evolution. Part I. Linear response theory and stability conditions, JHEP 08 (2024) 225 [2406.06708].arxiv.org/abs/2406.06708
  45. [45]D. F. G. Fiorillo and G. G. Raffelt, Theory of neutrino fast flavor evolution. Part II. Solutions at the edge of instability, JHEP 12 (2024) 205 [2409.17232].
  46. [46]N. F. Bell, A. A. Rawlinson and R. F. Sawyer, Speedup through entanglement: Many body effects in neutrino processes, Phys. Lett. B573 (2003) 86 [hep-ph/0304082].DOI
  47. [47]A. Friedland and C. Lunardini, Do many particle neutrino interactions cause a novel coherent effect?, JHEP 10 (2003) 043 [hep-ph/0307140].DOI
  48. [48]A. Friedland, B. H. J. McKellar and I. Okuniewicz, Construction and analysis of a simplified many-body neutrino model, Phys. Rev. D73 (2006) 093002 [hep-ph/0602016].DOI
  49. [49]A. Mezzacappa, E. Endeve, O. E. Bronson Messer and S. W. Bruenn, Physical, numerical, and computational challenges of modeling neutrino transport in core-collapse supernovae, Liv. Rev. Comput. Astrophys. 6 (2020) 4 [2010.09013].arxiv.org/abs/2010.09013
  50. [50]Y.-Z. Qian, G. M. Fuller, G. J. Mathews, R. Mayle, J. R. Wilson and S. E. Woosley, A connection between flavor mixing of cosmologically significant neutrinos and heavy element nucleosynthesis in supernovae, Phys. Rev. Lett. 71 (1993) 1965.pubmed.ncbi.nlm.nih.gov/10054549
  51. [51]J. T. Pantaleone, Neutrino flavor evolution near a supernova’s core, Phys. Lett. B 342 (1995) 250 [astro-ph/9405008].
  52. [52]H. Duan, A. Friedland, G. McLaughlin and R. Surman, The influence of collective neutrino oscillations on a supernova r-process, J. Phys. G 38 (2011) 035201 [1012.0532].DOI
  53. [53]C. J. Horowitz and G. Li, Nucleosynthesis in Supernovae, Phys. Rev. Lett. 82 (1999) 5198 [astro-ph/9904171].DOI
  54. [54]C. J. Horowitz and G. Li, Charge conjugation violating neutrino interactions in supernovae, Phys. Rev. D 61 (2000) 063002 [astro-ph/9908219].DOI
  55. [55]C. J. Horowitz, Weak magnetism for anti-neutrinos in supernovae, Phys. Rev. D 65 (2002) 043001 [astro-ph/0109209].
  56. [56]This subdominant but important effect boosts the νₑ cross sections on nucleons compared to those of ν̄ₑ.
  57. [57]M. Arnould, S. Goriely and K. Takahashi, The r-process of stellar nucleosynthesis: Astrophysics and nuclear physics achievements and mysteries, Phys. Rept. 450 (2007) 97 [0705.4512].DOI
  58. [58]D. Kasen, B. Metzger, J. Barnes, E. Quataert and E. Ramirez-Ruiz, Origin of the heavy elements in binary neutron-star mergers from a gravitational wave event, Nature 551 (2017) 80 [1710.05463].DOI
  59. [59]F.-K. Thielemann and J. J. Cowan, The r-process: history, required conditions, astrophysical sites, and observations, European Physical Journal A 62 (2026) 105 [2601.17246].
  60. [60]M. Arnould and S. Goriely, The p-process of stellar nucleosynthesis: astrophysics and nuclear physics status, Phys. Rept. 384 (2003) 1.DOI
  61. [61]T. Rauscher, N. Dauphas, I. Dillmann, C. Fröhlich, Z. Fülöp and G. Gyürky, Constraining the astrophysical origin of the p-nuclei through nuclear physics and meteoritic data, Reports on Progress in Physics 76 (2013) 066201 [1303.2666].
  62. [62]C. Frohlich, G. Martinez-Pinedo, M. Liebendorfer, F. K. Thielemann, E. Bravo, W. R. Hix, K. Langanke and N. T. Zinner, Neutrino-induced nucleosynthesis of a>64 nuclei: the νp-process, Phys. Rev. Lett. 96 (2006) 142502 [astro-ph/0511376].
  63. [63]J. Pruet, R. D. Hoffman, S. E. Woosley, H. T. Janka and R. Buras, Nucleosynthesis in early supernova winds. 2. the role of neutrinos, Astrophys. J. 644 (2006) 1028 [astro-ph/0511194].
  64. [64]S. Wanajo, The ν-process in neutrino-driven winds, Astrophys. J. 647 (2006) 1323 [astro-ph/0602488].
  65. [65]H. Schatz, L. Bildsten, A. Cumming and M. Wiescher, The Rapid Proton Process Ashes from Stable Nuclear Burning on an Accreting Neutron Star, Astrophys. J. 524 (1999) 1014 [astro-ph/9905274].DOI
  66. [66]J. Ehring, S. Abbar, H.-T. Janka, G. Raffelt and I. Tamborra, Fast Neutrino Flavor Conversions Can Help and Hinder Neutrino-Driven Explosions, Phys. Rev. Lett. 131 (2023) 061401 [2305.11207].DOI
  67. [67]J. Ehring, S. Abbar, H.-T. Janka, G. Raffelt and I. Tamborra, Fast neutrino flavor conversion in core-collapse supernovae: A parametric study in 1D models, Phys. Rev. D 107 (2023) 103034 [2301.11938].DOI
  68. [68]H. Nagakura, Roles of Fast Neutrino-Flavor Conversion on the Neutrino-Heating Mechanism of Core-Collapse Supernovae, Phys. Rev. Lett. 130 (2023) 211401 [2301.10785].DOI
  69. [69]T. Wang and A. Burrows, The Effect of the Fast-Flavor Instability on Core-Collapse Supernova Models, 2503.04896.
  70. [70]T. Wang and A. Burrows, The Effect of the Fast-flavor Instability on Core-Collapse Supernova Models. II. Quasi-equipartition and the Impact of Various Angular Reconstruction Methods, Astrophys. J. 997 (2026) 325 [2511.20767].arxiv.org/abs/2511.20767
  71. [71]K. Mori, T. Takiwaki, K. Kotake and S. Horiuchi, Three-dimensional core-collapse supernova models with phenomenological treatment of neutrino flavor conversions, Publ. Astron. Soc. Jpn. 77 (2025) L9 [2501.15256].arxiv.org/abs/2501.15256
  72. [72]M. Gogilashvili and I. Tamborra, Flavor Conversion Enhances or Suppresses Supernova Explodability Independent of the Progenitor Mass, 2605.18972.DOI
  73. [73]M. Gogilashvili and I. Tamborra, Neutrino Flavor Conversion Shapes the Rate of Failed Core-collapse Supernovae, 2605.16504.DOI
  74. [74]R. Akaho, H. Nagakura, W. Iwakami, S. Furusawa, A. Harada, H. Okawa, H. Matsufuru, K. Sumiyoshi and S. Yamada, Bifurcated Impact of Neutrino Fast Flavor Conversion on Core-Collapse Supernovae Informed by Multiangle Neutrino Radiation Hydrodynamics, Phys. Rev. Lett. 136 (2026) 191002 [2601.08269].DOI
  75. [75]Z. Xiong, A. Sieverding, M. Sen and Y.-Z. Qian, Potential Impact of Fast Flavor Oscillations on Neutrino-driven Winds and Their Nucleosynthesis, Astrophys. J. 900 (2020) 144 [2006.11414].DOI
  76. [76]G. Martinez-Pinedo, B. Ziebarth, T. Fischer and K. Langanke, Effect of collective neutrino flavor oscillations on νp-process nucleosynthesis, Eur. Phys. J. A 47 (2011) 98 [1105.5304].
  77. [77]H. Sasaki, T. Kajino, T. Takiwaki, T. Hayakawa, A. B. Balantekin and Y. Pehlivan, Possible effects of collective neutrino oscillations in three-flavor multangle simulations of supernova νp processes, Phys. Rev. D 96 (2017) 043013 [1707.09111].
  78. [78]A. B. Balantekin, M. J. Cervia, A. V. Patwardhan, R. Surman and X. Wang, Collective Neutrino Oscillations and Heavy-element Nucleosynthesis in Supernovae: Exploring Potential Effects of Many-body Neutrino Correlations, Astrophys. J. 967 (2024) 146DOI
  79. [79]S. Wanajo, H.-T. Janka and S. Kubono, Uncertainties in the ν p-process: supernova dynamics versus nuclear physics, Astrophys. J. 729 (2011) 46 [1004.4487].DOI
  80. [80]A. Friedland, P. Mukhopadhyay and A. V. Patwardhan, Successful νp-process in neutrino-driven outflows in core-collapse supernovae, JCAP 02 (2025) 005 [2312.03208].DOI
  81. [81]A. Friedland, D. J. Li, G. Lucente, I. Padilla-Gay and A. V. Patwardhan, νp-process in core-collapse supernovae: imprint of general relativistic effects, JCAP 02 (2026) 067 [2508.02055].arxiv.org/abs/2508.02055
  82. [82]A. Friedland and P. Mukhopadhyay, Near-critical supernova outflows and their neutrino signatures, Phys. Lett. B 834 (2022) 137403 [2009.10059].DOI
  83. [83]Garching core-collapse supernova research archive, https://wwwmpa.mpa-garching.mpg.de/ccsnarchive/.
  84. [84]D. F. G. Fiorillo, M. Heinlein, H.-T. Janka, G. Raffelt, E. Vitagliano and R. Bollig, Supernova simulations confront SN 1987A neutrinos, Phys. Rev. D 108 (2023) 083040 [2308.01403].DOI
  85. [85]G. Lucente, M. Heinlein, H.-T. Janka and A. Mirizzi, Simple fits for the neutrino luminosities from protoneutron star cooling, Phys. Rev. D 110 (2024) 063023 [2405.00769].DOI
  86. [86]T. Sukhbold, T. Ertl, S. E. Woosley, J. M. Brown and H. T. Janka, Core-Collapse Supernovae from 9 to 120 Solar Masses Based on Neutrino-powered Explosions, Astrophys. J. 821 (2016) 38 [1510.04643].DOI
  87. [87]H.-T. Janka and W. Hillebrandt, Neutrino emission from type II supernovae : an analysis of the spectra., Astron. Astrophys. 224 (1989) 49.
  88. [88]H.-T. Janka and W. Hillebrandt, Monte Carlo simulations of neutrino transport in type II supernovae., Astron. Astrophys. Suppl. Ser. 78 (1989) 375.
  89. [89]M. T. Keil, G. G. Raffelt and H.-T. Janka, Monte Carlo study of supernova neutrino spectra formation, Astrophys. J. 590 (2003) 971 [astro-ph/0208035].DOI
  90. [90]This prescription should be viewed as a representative, approximate limiting case, since flavor equilibration is constrained by lepton-flavor number conservation [110, 111]. However, the resulting asymptotic value of Yₑ is likely to vary within a narrow range around ∼ 0.6. A further quantitative investigation will follow in future work [101].
  91. [91]J. Lippuner and L. F. Roberts, SkyNet: A modular nuclear reaction network library, Astrophys. J. Suppl. 233 (2017) 18 [1706.06198].DOI
  92. [92]‘SkyNet’ source code available at: https://bitbucket.org/jlippuner/skynet/src/master/.
  93. [93]M. Beard, S. M. Austin and R. Cyburt, Enhancement of the Triple Alpha Rate in a Hot Dense Medium, Phys. Rev. Lett. 119 (2017) 112701 [1708.07204].DOI
  94. [94]S. Jin, L. F. Roberts, S. M. Austin and H. Schatz, Enhanced triple-α reaction reduces proton-rich nucleosynthesis in supernovae, Nature 588 (2020) 57.DOI
  95. [99]Studies prior to Refs. [80, 81] used insufficient 92Nb production to argue against νp-process viability [61, 112, 113].
  96. [100]K. Lodders, Solar System Abundances and Condensation Temperatures of the Elements, ApJ 591 (2003) 1220.DOI
  97. [101]A. Friedland, D. Li, G. Lucente, P. Mukhopadhyay, I. Padilla-Gay and A. Patwardhan, in preparation.
  98. [102]Oscillations at these distances might arise from the Mikheyev-Smirnov-Wolfenstein (MSW) effect, a manifestation of the coherent forward scattering of neutrinos with the electrons in the medium [114, 115]. This becomes only relevant when the outflow reaches densities of about 10^3 g/cm3 at distances of thousands of kilometers from the PNS. These densities are reached, however, at late times t ≳ 6 s [82] i.e. after the optimal window of the νp-process.
  99. [103]H.-T. Janka, Long-Term Multidimensional Models of Core-Collapse Supernovae: Progress and Challenges, Ann. Rev. Nucl. Part. Sci. 75 (2025) 425 [2502.14836].arxiv.org/abs/2502.14836
  100. [104]T. Fischer, G. Martinez-Pinedo, M. Hempel and M. Liebendorfer, Neutrino spectra evolution during proto-neutron star deleptonization, Phys. Rev. D 85 (2012) 083003 [1112.3842].DOI
  101. [105]K. Nakazato, K. Sumiyoshi, H. Suzuki, T. Totani, H. Umeda and S. Yamada, Supernova Neutrino Light Curves and Spectra for Progenitor Stars: From Core Collapse to Proto-neutron Star Cooling, Astrophys. J. Suppl. 205 (2013) 2 [1210.6841].DOI
  102. [106]T. Fischer, G. Guo, K. Langanke, G. Martinez-Pinedo, Y.-Z. Qian and M.-R. Wu, Neutrinos and nucleosynthesis of elements, Prog. Part. Nucl. Phys. 137 (2024) 104107 [2308.03962].DOI
  103. [107]L. Choi, A. Burrows and D. Vartanyan, Predicted neutrino signal features of core-collapse supernovae, Phys. Rev. D 111 (2025) 123038 [2503.07531].
  104. [108]B. Müller, H.-T. Janka and A. Marek, A New Multi-dimensional General Relativistic Neutrino Hydrodynamics Code for Core-collapse Supernovae. II. Relativistic Explosion Models of Core-collapse Supernovae, Astrophys. J. 756 (2012) 84 [1202.0815].DOI
  105. [109]R. Buras, M. Rampp, H. T. Janka and K. Kifonidis, Two-dimensional hydrodynamic core-collapse supernova simulations with spectral neutrino transport. 1. Numerical method and results for a 15 solar mass star, Astron. Astrophys. 447 (2006) 1049 [astro-ph/0507135].
  106. [116]Y. Z. Qian and S. E. Woosley, Nucleosynthesis in neutrino driven winds: 1. The Physical conditions, Astrophys. J. 471 (1996) 331 [astro-ph/9611094].
  107. [117]K. Otsuki, H. Tagoshi, K. Kajino and S.-y. Wanajo, General relativistic effects on neutrino driven wind from young, hot neutron star and the r process nucleosynthesis, Astrophys. J. 533 (2000) 424 [astro-ph/9911164].DOI
  108. [118]R. H. Cyburt, A. M. Amthor, R. Ferguson, Z. Meisel, K. Smith, S. Warren, A. Heger, R. D. Hoffman, T. Rauscher, A. Sakharuk, H. Schatz, F. K. Thielemann and M. Wiescher, The JINA REACLIB Database: Its Recent Updates and Impact on Type-I X-ray Bursts, Astrophys. J. Supplement Series 189 (2010) 240.DOI
  109. [119]A. Burrows and T. A. Thompson, Neutrino - matter interaction rates in supernovae: The Essential microphysics of core collapse, astro-ph/0211404.DOI
  110. [120]M. Rampp and H.-T. Janka, Radiation hydrodynamics with neutrinos: Variable Eddington factor method for core collapse supernova simulations, Astron. Astrophys. 396 (2002) 361 [astro-ph/0203101].DOI
  111. [121]H.-T. Janka, Explosion Mechanisms of Core-Collapse Supernovae, Ann. Rev. Nucl. Part. Sci. 62 (2012) 407 [1206.2503].DOI
  112. [122]R. Bollig, H. T. Janka, A. Lohs, G. Martinez-Pinedo, C. J. W. and T. Melson, Muon Creation in Supernova Matter Facilitates Neutrino-driven Explosions, Phys. Rev. Lett. 119 (2017) 242702 [1706.04630].DOI
  113. [123]R. Bollig, N. Yadav, D. Kresse, H. T. Janka, B. Müller and A. Heger, Self-consistent 3D Supernova Models From −7 Minutes to +7 s: A 1-bethe Explosion of a ∼19 M⊙ Progenitor, Astrophys. J. 915 (2021) 28 [2010.10506].DOI
  114. [124]A. W. Steiner, M. Hempel and T. Fischer, Core-collapse Supernova Equations of State Based on Neutron Star Observations, Astrophys. J. 774 (2013) 17 [1207.2184].DOI
  115. [125]M. Hempel and J. Schaffner-Bielich, A statistical model for a complete supernova equation of state, Nuclear Physics A 837 (2010) 210 [0911.4073].DOI
  116. [126]L. F. Roberts, G. Shen, V. Cirigliano, J. A. Pons, S. Reddy and S. E. Woosley, Proto-Neutron Star Cooling with Convection: The Effect of the Symmetry Energy, Phys. Rev. Lett. 108 (2012) 061103 [1112.0335].DOI
  117. [127]A. Pascal, J. Novak and M. Oertel, Proto-neutron star evolution with improved charged-current neutrino–nucleon interactions, Mon. Not. Roy. Astron. Soc. 511 (2022) 356 [2201.01955].DOI
  118. [128]J. Bliss, A. Arcones and Y.-Z. Qian, Production of Mo and Ru isotopes in neutrino-driven winds: implications for solar abundances and presolar grains, Astrophys. J. 866 (2018) 105 [1804.03947].
  119. [129]S. E. Woosley, J. R. Wilson, G. J. Mathews, R. D. Hoffman and B. S. Meyer, The r process and neutrino heated supernova ejecta, Astrophys. J. 433 (1994) 229.DOI
  120. [130]T. Iizuka, Y.-J. Lai, W. Akram, Y. Amelin and M. Schönbächler, The initial abundance and distribution of 92Nb in the Solar System, Earth and Planetary Science Letters 439 (2016) 172 [1602.00986].
  121. [131]Y. Hibiya, T. Iizuka, H. Enomoto and T. Hayakawa, Evidence for Enrichment of Niobium-92 in the Outer Protosolar Disk, Astrophys. J. Lett. 942 (2023) L15.DOI

Paper details

vlnc:W_ajqp6vsw
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