The role of rotation on the yields of the two γ-ray emitters 26Al and 60Fe ejected by massive stars

We show that the observed 60Fe/26Al flux ratio provided by the SPectrometer on INTEGRAL satellite (0.24 +- 0.04) can be reproduced only if rotation is taken into account in the computation of the stellar models. Predictions from non-rotating stellar models yield to a significantly lower ratio (0.062), which is incompatible with the observed value. The adopted models and the associated yields are based on a combination of models already published by Limongi & Chieffi (2018) complemented by additional ones fully consistent with the original grid, allowing a finer resolution in the initial rotational velocity distribution.

💡 Research Summary

The paper investigates how stellar rotation influences the nucleosynthetic yields of the two γ‑ray emitting isotopes 26Al and 60Fe produced by massive stars, and whether rotating models can reproduce the observed Galactic 60Fe/26Al flux ratio measured by the SPI instrument on INTEGRAL (0.24 ± 0.04). The authors build upon the Limongi & Chieffi (2018) grid of solar‑metallicity massive‑star models, extending it with additional calculations at intermediate initial equatorial rotation velocities (0, 50, 100, 150, 200, 250, 300 km s⁻¹) in steps of 50 km s⁻¹. The stellar evolution is computed with the FRANEC code, using the same input physics (nuclear network of 335 isotopes, mass‑loss prescriptions of Vink, de Jager, Nugis & Lamers, and an Eddington‑luminosity correction). Explosions are simulated with the HYperion code, enforcing an ejection of 0.07 M⊙ of 56Ni; stars with initial mass ≤ 25 M⊙ explode, while more massive stars are assumed to collapse completely, contributing only via their winds.

The authors first review the production channels of 26Al: central H‑burning (with rotation‑induced mixing enlarging the convective core), C‑shell burning, and explosive C/Ne burning. Rotation increases the convective core size, enhances mixing, and boosts mass loss, all of which raise the total 26Al yield. Their Figure 1 shows a nearly linear increase of 26Al yield with initial rotation speed for a range of masses.

For 60Fe, production occurs mainly in He‑ and C‑convective shells and in the explosion. In stars below 25 M⊙ the C‑shell dominates, while in more massive stars the He‑shell is the main source. Rotation modifies the internal structure so that rotating models behave like more massive non‑rotating ones: the CO core becomes larger, the residual 12C after He‑burning is reduced, and the C‑shells become more extended. Consequently, 60Fe yields generally increase with rotation, but the trend is non‑monotonic. At low rotation (≈ 50 km s⁻¹) the He‑shell is deeper and hotter, enhancing 60Fe, whereas at higher rotation the stronger wind strips the He‑shell, reducing the yield. This complex behavior is illustrated in Figure 2.

To compare with observations, the authors integrate the stellar yields over a Salpeter IMF (x = 1.35) and over a distribution of initial rotation velocities. Two rotation‑velocity distributions are considered: (i) a Gaussian fit to projected velocities of unevolved B‑type stars in NGC 3293 and NGC 4755 (peak v = 250 km s⁻¹, σ = 110 km s⁻¹) from Dufton et al. (2006), and (ii) an empirical distribution derived by Prantzos et al. (2018) that reproduces Galactic chemical evolution trends (65 % at 0 km s⁻¹, 30 % at 150 km s⁻¹, 5 % at 300 km s⁻¹). Performing a second integration over these velocity distributions yields a theoretical 60Fe/26Al ratio of 0.24 (Gaussian) and 0.26 (empirical), both comfortably within the SPI measurement. By contrast, the non‑rotating models predict a ratio of only 0.062, far below the observed value. The same low ratio is obtained from the earlier Limongi & Chieffi (2006) non‑rotating grid when the same explodability criteria are applied.

The conclusions emphasize that rotation is a crucial ingredient for reproducing the Galactic 60Fe/26Al γ‑ray flux ratio. Rotation not only boosts the production of 26Al via larger convective cores and stronger winds, but also modifies the He‑ and C‑shell structure in ways that can increase 60Fe yields, albeit in a mass‑ and velocity‑dependent manner. The authors argue that any realistic Galactic chemical evolution model must incorporate a realistic distribution of stellar rotation rates; otherwise, predictions for γ‑ray line fluxes and for the abundances of other isotopes will be systematically biased. This work thus provides the first comprehensive, rotation‑inclusive set of 26Al and 60Fe yields for massive stars and demonstrates their necessity for matching high‑precision γ‑ray observations.