Centrifugally driven mass-loss and outbursts of massive stars

Abstract

Rotation and mass-loss are crucially interlinked properties of massive stars, strongly affecting their evolution and ultimate fate. Massive stars rotating near their break-up limit shed mass centrifugally, creating Be stars with circumstellar discs and possibly driving outbursts. Using the mesa stellar evolution code, we examine the effects of efficient angular momentum transport on the main-sequence and post-main-sequence rotational evolution of massive stars. In rapid rotators, angular momentum transported from the contracting core to the expanding envelope can spin-up the surface layers past the break-up rate, particularly for stars near (or beyond) the end of the main-sequence and in low-metallicity environments. We also demonstrate that centrifugal instabilities could arise in rapidly rotating massive stars, potentially triggering the S Doradus outbursts observed in luminous blue variable stars. Prior mass accretion from a binary companion increases both the likelihood and the intensity of centrifugal mass-loss. We discuss implications for massive stellar evolution, Be stars, and luminous blue variables.

Progenitors of Type IIb Supernovae. II. Observable Properties

“…A key question about SE SNe is the mechanisms that drive the removal of the envelopes of their progenitor stars. In particular, the debate centers around the relative roles, if any, of stellar winds (e.g., Woosley et al 1993;Georgy et al 2012;Groh et al 2013b), stellar rotation (e.g., Georgy et al 2012;Groh et al 2013a,b;Zhao & Fuller 2020), binary interactions (e.g., Podsiadlowski et al 1992;Yoon et al 2010Yoon et al , 2017Soker 2017;Lohev et al 2019), and, nuclear burning instabilities (e.g., Arnett & Meakin 2011;Strotjohann et al 2015). There has been growing support for binary interactions as dominant due to several independent lines of evidence, including weaker stellar winds (Smith 2014) and higher binary fractions (Sana et al 2012;Moe & Di Stefano 2017) than previously estimated.…”

Wind-envelope interaction as the origin of the slow cyclic brightness variations of luminous blue variables

“…Uncertainties also involve the treatment of convection, turbulent linebroadening, clumping, and rotation on the structure and the force balance in such low-density envelopes. Especially fast rotation (Zhao & Fuller 2020) and turbulent convection (Schultz et al 2020) could contribute to reaching the local Eddington limit in stellar envelopes. The outward-directed centrifugal force or the higher radiative force from the opacity enhancement due to turbulence can contribute to accelerating the outflow to transonic velocities and in this way lower the threshold mass-loss rate to trigger the instability (potentially compensating for the artificial increase in mass-loss rate in our proof-of-concept evolutionary model).…”

Stellar winds and metal enrichment from fast-rotating Population III stars

“…Regardless of rotation, some other effects may be responsible for mass loss episodes during the evolution of massive stars, rather independent of metallicity, and thus occuring even in Pop III or very metal poor stars. An example may be the mechanism responsible for the strong mass loss experienced by Luminous Blue Variables (Smith & Owocki 2006;Zhao & Fuller 2020;Grassitelli et al 2021). Pulsation, either induced when the star is in a red supergiant stage (Yoon & Cantiello 2010), or in a blue supergiant phase (Saio et al 2013), may be linked to violent mass loss episodes.…”

“…To evaluate whether Pop III winds can make a difference in primordial metal enrichment, we further con- sider still more extreme cases where winds go deeper into the He core. Such strong mass loss may occur from pre-SN pulsations and instabilities (Smith & Owocki 2006;Woosley et al 2007;Yoon & Cantiello 2010;Fuller 2017;Fuller & Ro 2018;Leung & Fuller 2020;Zhao & Fuller 2020;Wu & Fuller 2021;Grassitelli et al 2021). We adopt as the maximum amount that can be lost by winds the mass above the incipient stellar remnant.…”

“…We set the wind timescale to the duration of the postmain sequence (MS) phase as an upper limit, as most winddriving mechanisms tend to occur during post-MS (Smith & Owocki 2006;Woosley et al 2007;Yoon & Cantiello 2010;Fuller 2017;Zhao & Fuller 2020;Wu & Fuller 2021;Grassitelli et al 2021). The wind velocity in turn is estimated Table 1.…”