Answers are grounded in this paper and cite their sources.
Theoretical and observational study explores runaway and hypervelocity stars, probing their formation processes and implications for stellar evolution.
Runaway stars were originally identified among young, massive O/B-type stars in the Galactic halo, but are now known to occur across nearly all stellar populations. These stars are kinematically distinct from their local stellar populations, exhibiting significantly higher velocities. A rare subset, called hypervelocity stars, are unbound from our Galaxy. Runaway stars are accelerated to velocities exceeding the local velocity dispersion through binary and multiple-star interactions. The processes invoked to explain their origin include supernovae of Type I and Type II, binary-single and binary-binary interactions in clusters, and binary-black hole interactions. As these processes are expected to generate stars that occupy the high-velocity tails of the local Galactic velocity distribution, runaway and hypervelocity stars serve as probes of these interactions. Main-sequence runaway stars trace star formation in young open clusters and challenge known ejection mechanisms in the most extreme kinematic cases. They are mainly formed either as companions to massive stars which undergo core-collapse supernovae or through dynamical encounters in open clusters. On the other hand, evolved runaway stars such as hot subdwarfs and white dwarfs may be formed either as the donor to a white dwarf which completely detonates in a thermonuclear supernova, or as the accreting white dwarf that survives a partial deflagration supernova. These evolved stellar populations are therefore associated with Type Ia and Type Iax (partial deflagration) supernovae, and the known populations of hypervelocity stars allow us to place stringent constraints on their progenitor systems. The observed population of runaways now includes rare objects like D6 stars, LP 40-365 type stars, hot subdwarfs, blue stragglers, and neutron stars. The formation of runaway stars is thus intricately linked to multiple phases of stellar evolution. The aim of this thesis is to probe pressing questions in this field, irrespective of the stellar population and formation. Some questions are more theoretical in nature, such as “why are runaway white dwarfs inflated?”. For such questions I use outputs from three-dimensional simulations of detonations in white dwarf binaries to evolve runaway star models in the one-dimensional stellar evolution code MESA. This approach has so far been relatively successful and has allowed us to constrain the timescales at which hypervelocity stars inflate. I show how these timescales vary as a function of the input physics in chapters 2, 3, 6, and 7. Other questions are more observational in nature, such as “if hot subdwarfs are involved in specific thermonuclear supernovae, where are the expected hypervelocity remnants?”. For this purpose I have relied on the public data release of the Gaia space telescope to gather candidates which were then followed up with multiple telescopes. Since it started operating, Gaia has collected astrometric data for over two billion stars, making it the perfect instrument for such a task. Although the follow-up observations have not yet found a definitive hot subdwarf runaway, each fast hot subdwarf observed provides an implicit constraint on the contribution of hot subdwarf–white dwarf binaries to the Type Ia supernova population. Furthermore, as demonstrated in this thesis, follow-up observations can occasionally result in the rediscovery of a different runaway population, posing a new problem to be solved. The discovery of a new LP 40-365 type star has led to a question which has yet to be answered: “why does the chemical composition of runaways thought to have been produced by Type Iax supernovae appear to be dominated by neon rather than oxygen and carbon?”. In chapter 5, I explain our discovery and outline the theoretical considerations that motivate and constrain further investigation of this open problem. Large datasets often exceed the capacity for exhaustive analysis, requiring statistical methods to identify viable candidates. In searches of runaway objects, most candidates are rejected as false positives. Nevertheless, rare objects can be recovered from noisy datasets through targeted selection and validation. In chapter 4, I discuss the properties of one such object: a tidally-locked, ancient binary, which was discovered by us. As an overarching theme, this thesis talks about objects which were found serendipitously by follow-up spectroscopic observations of flagged hypervelocity star candidates in the Gaia data release, suggesting that more exciting objects await us in the future.
No takes yet. Share an insight, caveat, or question.
Aakash Bhat (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: