Abstract

# Introduction

Stars are the fundamental observable in astrophysics. For this reason, much has been done to advance our understanding of stellar properties. Such work has many applications: beyond a simple desire to better understand how stars form, live, and evolve, stars are critical to understanding everything from our own Sun to the structure of galaxies. At this particularly exciting time, as extrasolar planets turn up everywhere we look, it is important to realize that our characterization of these planets is only as good as our understanding of their host stars. For this reason alone, any technique that allows us to characterize large numbers of stars both quickly and accurately is important.

Recently, asteroseismology has emerged as such a tool. Many stars with convective envelopes exhibit solar-like oscillations, which are a powerful way to probe stellar interiors. The recent Kepler mission was particularly well-suited to detect these, because it observed the same targets for a long period of time with a regular cadence. In particular, red giant stars with solar-like oscillations pulsate more slowly than our Sun due to their large size, and therefore appear in Kepler long-cadence ($$\sim 30$$ min) data.

Observing solar-like oscillations in red giants can tell us many things with relative ease: density, surface gravity, mass, radius, and even evolutionary stage. However, it is not always clear why some red giants willingly broadcast their inner secrets while others do not oscillate, or oscillate with lower amplitudes than we might expect. Given the power of asteroseismology coupled with our limited understanding of why it works when it does, it is imperative to study solar-like oscillators in environments that are observationally accessible using techniques independent of asteroseismology.

One such accessible environment is eclipsing binary stars. Gaulme et al. (2013) and Gaulme et al. (2014) identified some 20 red giants in eclipsing binaries (hereafter RG/EBs) in the Kepler field, most of which show solar-like oscillations but some of which do not. They also identified several triple systems consisting of two main sequence stars in an eclipsing binary that show transit timing variations due to the orbit of an oscillating red giant. The purpose of this work is to use the many tools available for modeling eclipsing binaries to piece together a “life story” of several RG/EBs, thereby giving more context to the results from asteroseismology and enabling us to understand what makes evolved stars oscillate in some situations but not others.

In this thesis proposal, we first review relevant literature in Section \ref{background}. We then present the main science questions to be addressed in Section \ref{questions}. Sections \ref{data} and \ref{method} outline the proposed project, while Section \ref{results} describes anticipated results. Finally, we summarize the project and conclude in Section \ref{conclusion}, and present the proposed timeline in Section \ref{timeline}.

# Background

\label{background}

## Red Giant Asteroseismology

\label{seismo} Stars with convective outer layers potentially exhibit solar-like oscillations. These oscillations depend on the physical processes in their interiors. Two kinds of waves exist: acoustic pressure modes, or p-modes, where a pressure gradient is the restoring force, and internal gravity waves, or g-modes, where gravity is the restoring force. The oscillations are driven stochastically and damped intrinsically by turbulence in the star’s subsurface convection zone (Houdek et al., 1999). Therefore, near-surface convection is a necessary condition for persistent solar-like oscillations.

P-mode oscillations appear as spikes in an amplitude spectrum of a light curve that is sampled both frequently enough and for a sufficiently long duration. Figure \ref{fig:rainbowmodes} shows oscillation spectra for stars with similar mass at different stages of stellar evolution. Each redder color corresponds to roughly an order of magnitude decrease in surface gravity.