[Paper Review] The advanced stages of stellar evolution: impact of mass loss, rotation, and link with B[e] stars

C. Georgy, Hideyuki Saio+2 more|arXiv (Cornell University)|Oct 24, 2016

Stellar, planetary, and galactic studies3 citations

TL;DR

This paper investigates how mass loss and rotation shape the late evolution of massive stars, revealing that some stars undergo a blue-red-blue evolution in the Hertzsprung-Russell diagram. The key finding is that Blue Supergiants (BSGs) may consist of two distinct populations: those on their first HRD crossing and others evolving back to the blue after a Red Supergiant phase, with observable differences in mass, surface gravity, and pulsational properties.

ABSTRACT

In this paper, we discuss some consequences of rotation and mass loss on the evolved stages of massive star evolution. The physical reasons of the time evolution of the surface velocity are explained, and then we show how the late-time evolution of massive stars are impacted in combination with the effects of mass loss. The most interesting result is that in some cases, a massive star can have a blue-red-blue evolution, opening the possibility that Blue Supergiants are composed by two distinct populations of stars: one just leaving the main sequence and crossing the HRD for the first time, and the other one evolving back to the blue side of the HRD after a Red Supergiant phase. We discuss a few possible observational tests that can allow to distinguish these two populations, and how supergiant B[e] stars fit in this context.

Motivation & Objective

• To understand the impact of mass loss and rotation on the late-stage evolution of massive stars.
• To investigate the physical mechanisms driving surface velocity evolution post-main sequence.
• To examine the possibility that Blue Supergiants (BSGs) represent two distinct evolutionary populations: first and second HRD crossings.
• To assess observational diagnostics that can distinguish between first- and second-crossing BSGs.
• To explore the connection between evolved massive stars and the origin of supergiant B[e] stars (sgB[e])

Proposed method

• Stellar evolution models for initial masses between 12 and 40 M⊙ at solar metallicity were used, based on the Geneva stellar evolution code (Ekström et al. 2012).
• Evolution of surface rotational velocity was modeled by tracking angular momentum conservation, internal redistribution via instabilities, and loss via stellar winds.
• Mass loss rates during the Red Supergiant (RSG) phase were modeled as 10⁻⁶ to 10⁻⁵ M⊙ yr⁻¹ over 10⁵ to 10⁶ years, significantly altering stellar mass and structure.
• The luminosity-to-mass ratio (L/M) was computed to assess conditions favoring pulsations, particularly the onset of strange modes when L/M > 10⁴ L⊙/M⊙.
• Chemical abundance evolution, especially the ¹²C/¹³C ratio, was tracked to identify signatures of CNO-cycle burning exposed by mass loss.
• The Flux-weighted Gravity–Luminosity Relation (FGLR) was used to test model consistency with observed BSGs, comparing first- and second-crossing tracks.

Experimental results

Research questions

• RQ1 Can the observed surface velocity evolution in massive stars be explained by radius expansion, internal angular momentum transport, and wind-driven loss?
• RQ2 Under what conditions does a massive star exhibit a blue-red-blue evolution in the Hertzsprung-Russell diagram?
• RQ3 What observable differences distinguish BSGs on their first HRD crossing from those on their second crossing after a Red Supergiant phase?
• RQ4 How does mass loss during the RSG phase affect the L/M ratio and the potential for pulsational instability in BSGs?
• RQ5 Can the properties of supergiant B[e] stars (sgB[e]) be explained by single-star evolution, or is a binary origin more plausible?

Key findings

• Stars with initial masses ≥15 M⊙ exhibit strong surface velocity decline after the main sequence due to combined effects of radius expansion and wind-driven angular momentum loss.
• The ¹²C/¹³C ratio in evolved models shows rapid changes at the end of RSG phases, indicating exposure of CNO-burned material due to mass loss.
• Second-crossing BSGs have significantly lower surface gravity than first-crossing BSGs at the same effective temperature due to mass loss during the RSG phase.
• The L/M ratio in second-crossing BSGs exceeds 10⁴ L⊙/M⊙, creating conditions favorable for the development of strange pulsation modes.
• Observational detection of α Cygni variables—pulsating BSGs—supports the existence of blue-red-blue evolution in stars of 15–25 M⊙.
• Stellar models reproduce the observed FGLR only if most BSGs are either on their first crossing or have lost little mass during the RSG phase, highlighting the critical role of mass loss in enabling blueward evolution.