Population Synthesis of Evolved Massive Binary Stars in the Magellanic Clouds

Abstract

Massive stars are crucial building blocks of the Universe. They produce heavy elements, drive the evolution of galaxies, and are the origin of spectacular events such as supernovae and gravitational-wave coalescences. While the evolution of massive single stars alone is complicated, observations have shown that most of them are part of close binary systems, in which the two stars sooner or later interact by mass transfer between them, leading to an even more complex evolution that is only partially understood. <br /> Among the many phases of binary evolution, we study binaries in which one component has become a stellar remnant, as this is the first long-lived phase after the well-understood phase with two stars. To study these, we focus not on individual systems, but on their entire population. We use two methods that assume two different physical conditions for stable mass transfer. The first method, based on detailed binary models in which the stellar structure equations derived from first principles are solved, uses an energy criterion, and the second method, rapid population synthesis, which models the stars and their evolution in a simplified way, uses a thermodynamic criterion. To use the second method, we need to make some preparations and derive new and more flexible descriptions of binary physics. These are rotation, mass transfer on the nuclear timescale, and accretion. <br /> Rotation is a ubiquitous phenomenon in stellar evolution. To predict the spin of stellar remnants, we need to know how the angular momentum is distributed inside a star. We compare the outcome of different model assumptions with the star LB-1, which has recently lost its envelope, providing a direct view into the stellar interior, and find a strong preference for magnetic angular momentum transport. <br /> Massive binary stars prefer close orbits, which makes the interaction phase more complex. This is a challenge for the rapid method, so we use dense grids of detailed models to derive recipes for the outcome of their interaction. We find that for fixed initial masses for close systems there is a correlation between the final mass and the orbital periods, and an anti-correlation between the duration of the mass transfer and the orbital period. <br /> Finally, we need to understand the conditions under which binary interactions are stable and lead to the systems of interest. It has long been known that accreting stars can expand, and so we derive conditions for the intensity of the expansion. Assuming that this effect can lead to a massive loss of angular momentum from the binary, we derive which orbital configurations can stably interact and lead to a close binary containing a black hole or neutron star. <br /> Using the above results, we derive synthetic populations of massive stars with remnant companions. Both methods are in good agreement with observations and predict a large number of yet unobserved massive stars with black hole companions, which can be identified either as binaries with large radial-velocity variations or as emission-line stars. We also predict that there are significant and testable differences between the two models.