Production of heavy $α$-elements and $^{44}$Ti in Cas A: comparison to abundances from 1D core-collapse supernova models and evidence for Carbon-Oxygen shell mergers

Abstract

The merger between the carbon (C) and oxygen (O) shells hours to days before the collapse of a massive star significantly changes its nucleosynthesis, which is reflected in the elemental ratios observed in supernova remnants (SNRs). We present a nucleosynthesis study of $^{44}$Ti production in core-collapse supernovae (CCSNe), highlighting large silicon (Si), sulfur (S), calcium (Ca), and, most importantly, argon (Ar) to neon (Ne) ratios as diagnostics for carbon-oxygen (C--O) shell mergers. We compare yields from eight different sets of CCSNe models to observations of Cassiopeia A (Cas A), and show that C--O shell mergers are consistently the models that best match X-ray and infrared observations. These models produce high Ar/Ne ratios ($\gtrsim 0.1$), due to $^{20}$Ne depletion and production of $^{36}$Ar and $^{38}$Ar, while lower ratios are obtained from non-merger cases. Based on the Ar/Ne diagnostic, we compare the range of expected $^{44}$Ti produced by C--O shell mergers, which is up to $\sim 20 - 30 \%$ of the overall $^{44}$Ti, but expected to be located outside the reverse shock. Based on the sets of models considered, the photon flux expected from the $^{44}$Ti synthesized in the C--O shell merger in Cas A is below the $NuSTAR$ and $COSI$ detection limits, compatible with current limits locating most of the $^{44}$Ti interior to the reverse shock, but might be detectable from proposed missions like $ASCENT$. Finally, for the SNR of 1987A, a dominant C--O merger origin of the observed $^{44}$Ti is unlikely based on the observed redshift in its $^{44}$Ti line.

Figures (2)

• Figure 1: Mass ratios of selected elements after the explosion, when all isotopes have decayed back to stability, for the nucleosynthesis yields from Boccioli2025_ExplMattHowRealNdriExplChan (star symbols) and Sukhbold2016_CoreSupe9120SolaMassBase (squares). The x-axis of the rightmost panel shows the amount of $\ce{^44Ti}$ synthesized in the C--O merger and then ejected, calculated as outlined in the caption of Figure \ref{['fig:Ti44_ej_frac']}. Points are color-coded by the fraction of merger completeness (i.e. hoe well mixed the shell is Loddo2026_), where everything below $\sim 0.4-0.5$ (i.e. the cyan points) can be considered to not have experienced a C--O merger. The green dots show mass ratio estimates from Hwang2012_ChanXraySurvEjecCassSupeRemn for two values of the filling factor. Orange crosses are from three different regions (N, SE, W) observed by Vink1996_NewMassEstiPuzzAbunSNRCass. Red points are from Willingale2002_XraySpecImagDoppMappCass, but since the rms uncertainty associated with them is very large, we decided not to show it.
• Figure 2: M$^{\rm ^{44}Ti}_{\rm merger}$ represents the $\ce{^44Ti}$ produced in the merger. We define it as the integrated yield above the Si/Si-O interface in the pre-SN progenitor. M$^{\rm ^{44}Ti}_{\rm ejected}$ is then calculated by considering only the shells above the Si/Si-O interface where the pre-SN abundance of $\ce{^44Ti}$ changes by less than 1 % after the explosion. Grey points are stars without a C--O merger, whereas those with a C--O merger have been color-coded by the mass of the merged shell, and annotated with the model ID convention outlined in Table \ref{['tab:merger_description']} and in Appendix \ref{['app:preSN_models']}.