Characterizing the O’Connell Effect in Kepler Eclipsing Binaries

The O'Connell effect – the presence of unequal maxima in eclipsing binary light curves – is a poorly understood phenomenon that has been recognized for over a century. Several ideas have been proposed to explain it, including chromospheric spots, effects of mass transfer, or circumstellar material, but the exact cause of the effect nevertheless remains unresolved. The Kepler mission observed nearly 3,000 eclipsing binaries, of which my analysis shows that 201 show a significant O'Connell effect.

I have found that the systems predominantly have periods under 1 day, range in spectral type from early-M to mid-A, and largely fall on the main sequence. I found some interesting trends, such that there are few β Lyrae systems present in the sample and that systems in the sample tend to have the deeper eclipses than non-sample systems with similar-looking light curves. I found that the difference in maxima is weakly correlated with the depth of primary eclipse, but no other correlations exist with the maxima difference. This last result is surprising because previous research on spots would imply a positive correlation between maxima difference and both temperature and orbital period.

I have found a great variety of systems in the sample, and some classes of systems demand special attention. Many systems have light curves that are temporally stable, but many other systems have light curves that vary considerably. These systems are generally cooler than other systems in the sample and often display flares, pointing to very active stars and suggesting that the evolution of chromospheric spots is the cause of these variations. There are also a few rare systems that display an asymmetry in the minimum, and it is notable that all such systems display total eclipses. Testing indicates that the presence of this asymmetry is strongly dependent on the orbital inclination and I postulate that this asymmetry can only be detected in totally eclipsing systems. These systems are generally hotter than most others in the sample.

This study lays the foundation for a future study investigating the physical causes of the O’Connell effect. This impacts our theories on stellar evolution given that a plausible explanation of the phenomenon is mass transfer, a process believed to occur in all close binaries. Studying if spots cause the O’Connell effect gives insight into what stars exhibit spots, and such knowledge is vital for photometrically precise experiments such as detecting exoplanet transits.