Production of radioactive $^{22}$Na in core-collapse supernovae: the Ne-E(L) component in presolar grains and its possible consequences on supernova observations

TL;DR

The study addresses the Ne-E(L) signature observed in presolar graphites by exploring late H-ingestion in the He shell of core-collapse supernovae as a mechanism to synthesize $^{22}$Na. Through twelve 1D CCSN models, it shows that $^{22}$Na production in the He shell can reach yields up to $3\times10^{-3}\,M_\odot$, with a minimum grain-relevant abundance around $10^{-4}$ in a C-rich mixture, making it plausible to explain Ne-E(L). The work further demonstrates that such $^{22}$Na enhancements can power CCSN light curves at 500–1500 days and produce a detectable $^{22}$Na gamma-ray line at $1274.53$ keV with current or future observatories, while also assessing potential contributions to Galactic positrons. The results connect presolar grain isotopic signatures with observable late-time CCSN phenomena and highlight significant modeling and nuclear-rate uncertainties that require 3D hydrodynamics and further experiments to refine the scenario.

Abstract

Presolar graphite grains carry the isotopic signatures of their parent stars. A significant fraction of presolar graphites shows isotopic abundance anomalies relative to solar for elements such as O, Si, Mg and Ca, which are compatible with nucleosynthesis in core-collapse supernovae (CCSNe). Therefore, they must have condensed from CCSN ejecta before the formation of the Sun. Their most puzzling abundance signature is the $^{22}$Ne-enriched component Ne-E(L), interpreted as the effect of the radioactive decay of $^{22}$Na ($T_{1/2}$ = 2.6 years). Previous works have shown that if H is ingested into the He shell and not fully destroyed before the explosion, the CCSN shock in the He shell material produces large $^{22}$Na amounts. Here we focus on such CCSN models, showing a radioactive $^{26}$Al production compatible with grains measurements, and analyze the conditions of $^{22}$Na nucleosynthesis. In these models, $^{22}$Na is mostly made in the He shell, with a total ejected mass varying between 2.6$\times$10$^{-3}$ M$_{\odot}$ and 1.9$\times$10$^{-6}$ M$_{\odot}$. We show that such $^{22}$Na may already impact the CCSN light curve 500 days after the explosion, and at later stages it can be the main source powering the CCSN light curve for up to a few years before the $^{44}$Ti decay becomes dominant. Based on the CCSN yields above, the 1274.53 keV $γ$-ray flux due to $^{22}$Na decay could be observable for years after the first CCSN light is detected, depending on the distance. This makes CCSNe possible sites to detect a $^{22}$Na $γ$-ray signature consistently with the Ne-E(L) component found in presolar graphites. Finally, we discuss the potential contribution from the $^{22}$Na decay to the galactic positron annihilation rate.

Figures (4)

• Figure 1: Top left panel: isotopic abundances (in mass fraction) in the He-shell ejecta of the CCSN models 25T-H and 25T-H10 (thick and thin lines, respectively). The horizontal red colored band highlights the $^{22}$Na production range compatible with the Ne-E(L) component (see text for details); Top right panel: As the top left panel, but for the 25av-H and 25av-H10 (thick and thin lines, respectively). Bottom left panel: Total $^{22}$Na yields in solar masses (continuous lines) and from the He-shell ejecta only (dashed lines) for all the 25T set models with respect to the available amount of hydrogen at the onset of the CCSN explosion in the He shell. For comparison, the same data are provided for $^{26}$Al and $^{60}$Fe. The yields of the models consistent with the Ne-E(L) component are highlighted with large red circles. The vertical magenta colored area highlights the production variation between the 25T-H and the 25T-H10 models. In this area, the total $^{22}$Na yields are always dominated by the He-shell ejecta. Bottom right panel: As the bottom left panel, but for the 25av set of CCSN models.
• Figure 2: Integrated nucleosynthesis fluxes (log$_{10}([\delta$Y$_{\rm i}$/$\delta$t]$_{\rm j})$, showing the cumulative variation of the abundance Y$_{\rm i}$ = X$_{\rm i}$/A$_{\rm i}$ due to the reaction j) at 47 seconds after the peak explosion temperature at mass coordinates 6.855 M$_{\odot}$ (top left, CCSN shock peak temperature T$_{\rm peak}$ = 1.1 GK), 6.908 M$_{\odot}$ (top right, T$_{\rm peak}$ = 0.8 GK), 7.038 M$_{\odot}$ (bottom left, T$_{\rm peak}$ = 0.55 GK), 7.333 M$_{\odot}$ (bottom right, T$_{\rm peak}$ = 0.33 GK) of the model 25T-H, in the regions of interest for the $^{22}$Na production. These mass coordinates correspond to different production efficiency of $^{22}$Na (see Figure \ref{['fig:na22_yields']}, upper left panel). The arrow color scale on the right correspond to the flux strength. Heavy-lined boxes correspond to the stable isotopes.
• Figure 3: Top panel: The UVOIR pseudo-bolometric light curve is shown between 500 and 5000 days after the explosion timmes:96. The relative contribution from the decay of $^{56}$Co, $^{57}$Co, $^{44}$Ti, $^{60}$Co and $^{22}$Na are considered using previous default abundances. In particular, the partial $^{22}$Na curve obtained using the default value of 2$\times$10$^{-6}$ M$_{\odot}$ from timmes:96 is also shown (green continuous line). For the range of $^{22}$Na production obtained in the 25T-H (highest green dashed line), 25T-H5 (green dotted line) and 25T-H10 models (lowest green dashed line), the corresponding total pseudo-bolometric light curves are shown as a higher black dashed line, black dotted line and lower black dashed line respectively. Their variation due to the $^{22}$Na production range is highlighted as a black striped area. Bottom panel: Same as the top panel, but considering the $^{22}$Na production of the 25av-H, 25av-H5 and 25av-H10 models.
• Figure 4: Upper Panel: The predicted 1274.53 keV $\gamma$-ray flux due to $^{22}$Na decay is calculated for the SN1987A remnant from the first possible detection of the CCSN explosion until the present day, if we consider the yields from the models 25T-H (higher green dashed line), 25T-H5 (green continous line) and 25T-H10 (lower green dashed line) and the models 25av-H (higher magenta dashed line), 25av-H5 (magenta continous line) and 25av-H10 (lower magenta dashed line). The area between the two sets of curves are colored accordingly. A distance of 51.4 Kpc is adopted panagia:99. The flux obtained using the default $^{22}$Na abundance from timmes:96 is reported as a reference (black continuous line). Note that the curves for 25T-H10 and 25av-H10 are not visible, hidden behind the default (black) curve (25T-H10) and the 25av-H curve (25av-H10). Horizontal lines are the detection limit for COMPTEL iyudin:10 and INTEGRAL siegert:18 during their operation time. Lower Panel: The same as in the Upper Panel, but for Cas A, at 3.4 Kpc. In this case, COMPTEL and INTEGRAL detection limits appear just like data points (within the red circle) along the longer timeline.