Metal-poor single Wolf-Rayet stars: The interplay of optically thick winds and rotation

TL;DR

This paper demonstrates that single metal-poor O-type stars can become Wolf-Rayet stars through envelope self-stripping when rotation drives optically thick winds. Using MESA with Sabhahit2023-based wind physics and a two-channel activation framework, the authors show that at SMC-like metallicity ($Z\sim0.002$–$0.004$) fast rotation ($\Omega/\Omega_c\approx0.6$) enables WR formation for initial masses as low as $\sim25\,M_\odot$, aligning with observed SMC WRs and alleviating the Humphreys-Davidson overpopulation issue. The results depend non-monotonically on metallicity and rotation, with overshooting playing a degenerate role similar to rotation in enabling thick winds; implications arise for black-hole masses, explosion outcomes, and the formation of close compact-object binaries. The study also shows thick winds can operate down to very low $Z$, potentially explaining WR signatures in extremely metal-poor galaxies like IZw18, and provides a framework to interpret HD-limit occupancy and late-stage stellar evolution in metal-poor environments.

Abstract

The Small Magellanic Cloud (SMC) hosts 12 known Wolf-Rayet (WR) stars, seven of which are apparently single. Their formation is a challenge for current stellar evolution models because line-driven winds are generally assumed to be quenched at a metallicity of Z < 0.004. Here, we present a set of mesa models of single stars with zero-age main sequence masses of 20 - 80 Msun considering different initial rotation speeds (Ω = 0 - 0.7 Ω_c), metallicities (Z = 0.002 - 0.0045), and wind mass-loss models (optically thin and thick winds). We show that if we account for optically thick winds, fast rotating (Ω = 0.6 Ω_c) single metal-poor O-type stars (with M > 20 Msun) shed their envelope and become WR stars even at the low metallicity of the SMC. The luminosity, effective temperature, evolutionary timescale, surface abundance, and rotational velocity of our simulated WR stars are compatible to the WRs observed in the SMC. We speculate that this scenario can also alleviate the excess of giant stars across the Humphreys-Davidson limit. Our results have key implications for black hole masses, (pair instability) supernova explosions, and other observable signatures.

Figures (24)

• Figure 1: Stellar tracks on the HR diagram for initial masses $M=25,\,30,\, 40,\, 60$, and $80$ M$_\odot$ at metallicity $Z=0.0025$. In the left-hand panels, the Vink2001 model with only optically thin winds is implemented. In the right-hand panels, the Sabhahit2023 model is enforced, with the possibility to activate optically thick winds. The upper panels show the case with no rotation, while the lower panels the $\Omega/\Omega_c=0.6$ case. Different line styles represent different wind regimes: dotted for optically thin winds, solid for optically thick winds, dashed for WR-type winds. Colors represent different burning stages, defined as the element whose burning generates most of the energy of the star. Blue is for H-burning through CNO, orange for He-burning through triple $\alpha$, red for carbon burning, and green for oxygen burning. Green circles are observations of single WRs Hainich2015, filled for the hottest, open for the coldest. The lower-right plot shows that the combination of optically thick winds and high initial rotation is able to self-strip stars even at SMC metallicity. Stars in the lower-right plot (Sabhahit2023, $\Omega/\Omega_c=0.6$) avoid the HD limit and transition from main-sequence stars to WRs.
• Figure 2: Velocity distribution of O-type stars in the SMC. The plot shows the normalized observed $\varv \sin i$ distribution (blue) and the reconstructed initial rotational velocity ($\varv_{\rm rot}$, red) for 171 O-type stars. The dataset is a compilation from Ramachandran2019, Rickard2024, and Bestenlehner2025. About $10\%$ of the O-type stars feature $\varv_{\rm rot}>450$ km/s.
• Figure 3: Illustration depicting the possible evolutionary paths of a chemically homogeneous star with initial mass $\gtrsim 20\,\rm M_\odot$ at SMC metallicity. After core H-burning (yellow ball with blue center), the star contracts, $\Gamma_\text{e,switch}$ decreases and $\Gamma_\text{e}$ increases. If $\Gamma_\text{e}>\Gamma_\text{e,switch}$ the star enters the optically thick winds regime (left arrow) and ends up being a WR (channel 1). If $\Gamma_\text{e}<\Gamma_\text{e,switch}$ (right arrow), winds are still optically thin during shell H-burning (yellow ball with blue shell) and the star expands and cools, increasing $\Gamma_\text{e,switch}$. At this stage, there are two possible outcomes: (i) If the star cools enough to cross the bi-stability jump $T_{\rm eff}<25$ kK (right arrow), it may enter the optically thick wind regime and become a WR (channel 2). (ii) If the star starts core He-burning at $T_{\rm eff}>25$ kK (left arrow), $T_{\rm eff}$ rises again, and the star does not cross the bi-stability jump. Optically thick winds are not activated (or are activated too late) and the star ends up being a cool supergiant.
• Figure 4: Stellar tracks for a $25$ M$_\odot$ star. The upper panel shows a rotating star with $\Omega/\Omega_c=0.6$ for three different metallicities $Z=0.002,\,0.003,\,0.004$. The lower panel shows a star with metallicity $Z=0.003$ for three different values of the initial rotation speed $\Omega/\Omega_c=0.5,\,0.6,\,0.7$. The linestyles and color code are the same as in Figure \ref{['fig:new_winds']}. The vertical dotted black line represents the temperature below which we expect the bi-stability jump. This Figure shows the non-monotonic dependence of mass-loss on metallicity and rotation. Low metallicity ($Z=0.002$) favors the activation of optically thick winds through channel 1, because the star is more luminous and compact. Higher metallicity ($Z=0.004$) can also activate optically thick winds through channel 2. Stars in the intermediate metallicity case ($Z=0.003$), instead, fail to activate optically thick winds soon enough and end up as cool supergiant stars. The same trend happens for rotation.
• Figure 5: Time spent on different regions of the HR diagram for stars with mass $M=20,\,30,\,40,\,60$, and $80$ M$_\odot$ at $Z=0.002$ and $\Omega/\Omega_c=0.6$. In the upper panel, the color code represents the quantity $\tau_\text{HRD}$, in yr $\rm dex^{-1}$, which is the time spent to move by $1\,\rm dex$ in the HR diagram. Green circles are WRs in the SMC ,and blue squares are O-type stars in the SMC with $\varv\,\sin i>200$ km/s from Mokiem2006Bouret2013Bouret2021Ramachandran2019Dufton2019Rickard2024Backs2024. In the lower panel, the color code represents different burning stages: core H-burning (blue), He-burning (orange), carbon burning (red), and oxygen burning (green), while the line styles represent different wind regimes: dotted for optically thin winds, solid for optically thick winds, dashed for WR-type winds. Cross markers are spaced by $\sim 5\times 10^4\,\rm yr$ each. After the main sequence, stars spend most of their time at the turn-off point, where they start to peel off their H-rich envelope and move blueward, and in the hot WR phase, at $4.9<\log (T_{\rm eff}/{\rm K})<5.2$.