The formation mechanisms of classical emission line stars

Abstract

Classical emission line stars are known fast rotators hosting a decretion disc in which emission lines are formed. Since their discovery in 1866, the formation mechanism of this class of stars has proved difficult to identify. A robust understanding of how emission line stars ultimately came to be would constrain both the stars’ previous and future evolution. Classical emission line stars make up a significant proportion of massive stars, with recent observations showing that around one third of massive stars display emission lines, thus knowledge of these stars is important for the study of massive stars in general. <br /> To explain the rapid rotation, two formation channels exist; single and binary star evolution respectively. The single star formation channel is whereby a star with a given amount of seed angular momentum undergoes structural changes during its evolution that cause the centrifugal force at the equator to approach the gravitational force, causing the star to effectively spin up. A fast rotating star is formed through binary evolution via mass-transfer, whereby an accreting star gains angular momentum as well as mass, and thus can attain large rotational velocities. <br /> This thesis investigates the contributions of both the single and binary star evolution channels to the observed population of emission line stars. Numerical models of single rotating stars were used to predict the rotational velocities of a stellar population. The failure of the models to explain the large number of emission line stars found in open clusters suggests that binary evolution to be the dominant formation mechanism. As the outcome of binary star evolution is sensitive to uncertain physics, a simple and flexible analytic model of binary evolution was developed. Comparison of the model with an observed population found a good match between the two, but only when the model contains certain specific assumptions, which may or may not be realised in nature, such as very inefficient mass-transfer. <br /> Both formation mechanisms suffer from distinct uncertainties. For the single star channel, stellar winds, which govern rotational evolution, are affected by rapid rotation in ways which are often ignored. A self-consistent description of the wind of a fast rotator revealed however the effects to be minimal and the spin evolution was not expected to differ significantly from previous models. In the binary channel, mass-transfer efficiency was constrained using stripped-star binaries. It was found that around half of the mass removed from the donor star is accreted by its companion, challenging the validity of the assumptions required for the binary channel to dominate. <br /> Based on the theoretical arguments set out in each of the four chapters, this thesis cannot fully endorse one formation channel over the other, with the most probable situation being that both channels co-exist to produce emission line stars.