Eclipse timing of binary stars to search for extra-solar planets
Over the past seven years, the Kepler space telescope has discovered several thousand extra-solar planets and planet candidates by observing their periodic transits in front of their host stars. About a dozen of these transiting exoplanets have been detected in wide orbits around a central stellar binary. They are called “circumbinary” or “P-type” planets (“P” for planet). Most of their host binaries also show mutual eclipses, therefore allowing precise radius estimates of both the stars and their planets. Planets can also exist in close orbits around single stars with an outer stellar companion on a wide orbit. Many of the known exoplanets actually belong to this class of “S-type” planets (“S” for satellite), but none of their wide-orbit stellar companions are eclipsers. In other words, no S-type exoplanet has been found in an eclipsing stellar binary. In fact, S-type planets in eclipsing binaries can be expected to be rare from both a formation and an orbital stability point of view. Hence, the discovery of an S-type planet would have major effects on our understanding of planet formation and evolution. The detection of almost 3,000 close eclipsing binaries with Kepler, ~200 of which have orbital periods between 30d and 100d, suggests that a dedicated search could be successful. At the end of this thesis, the PhD student will have obtained an expertise in computational data analysis, in exoplanetary physics, and stellar light curve analyses. The student will be well-prepared for follow-up studies involving observations from near-future exoplanet space missions such as TESS, CHEOPS, or PLATO.
The PhD student will download the freely accessible raw Kepler light curves of the ~3,000 eclipsing stellar binaries from the MAST online repository. A first exercise will be to detrend these light curves, to remove non-physical artifacts, and to recover the previously published system parameters using a stellar binary transit model. Various routes can be tested towards an S-type exoplanet. One approach invokes the well-established search for the planet’s own photometric transit in the light curve. A different approach is based on the search for eclipse timing variations (ETVs) and eclipse duration variations (EDVs) of the stars, both of which can be caused by the presence of a planet around either star. This latter approach has only been identified very recently using simulations (Oshagh, Heller, Dreizler 2017), but a combined ETV-EDV-based exoplanet search in real Kepler data has not been performed.
The candidate will need a Master of Science with a specialization in physics, mathematics, computer sciences, or a related field. The project requires programming skills or the proven potential to obtain them quickly. A basic understanding of stellar and planetary physics is desired and will be helpful. Data analysis and computer simulations will form an integral part of this thesis.