The full $\mathit{Kepler}$ phase curve of the eclipsing hot white dwarf binary system KOI-964

TL;DR

This paper presents a comprehensive analysis of the KOI-964 binary system using Kepler photometry and radial velocity data, revealing detailed stellar and orbital properties, variability signals, and phase curve components, with some discrepancies in expected ellipsoidal distortion.

Contribution

It provides the first full phase curve analysis of KOI-964, combining photometry and spectroscopy to refine system parameters and explore stellar variability and tidal interactions.

Findings

Refined stellar and orbital parameters of KOI-964.

Detected stellar variability with a period of 0.620276 days.

Discrepancy between observed and predicted ellipsoidal distortion amplitudes.

Abstract

We analyze the full $\mathit{Kepler}$ phase curve of KOI-964, a binary system consisting of a hot white dwarf on an eclipsing orbit around an A-type host star. Using all 18 quarters of long-cadence photometry, we carry out a joint light curve fit and obtain improved phase curve amplitudes, occultation depths, orbital parameters, and transit ephemeris over the previous results of Carter et al. 2011. A periodogram of the residuals from the phase curve fit reveals an additional stellar variability signal from the host star with a characteristic period of $0.620276\pm0.000011$ days and a full amplitude of $24\pm2$ ppm. We also present new Keck/HIRES radial velocity observations which we use to measure the orbit and obtain a mass ratio of $q=0.106\pm0.012$. Combining this measurement with the results of a stellar isochrone analysis, we find that the masses of the host star and white dwarf companion are $2.23\pm0.12\,M_{\odot}$ and $0.236^{+0.028}_{-0.027}\,M_{\odot}$, respectively. The effective temperatures of the two components are $9940^{+260}_{-230}$ K and $15080\pm400$ K, respectively, and we determine the age of the system to be $0.21^{+0.11}_{-0.08}$ Gyr. We use the measured system properties to compute predicted phase curve amplitudes and find that while the measured Doppler boosting and mutual illumination components agree well with theory, the ellipsoidal distortion amplitude is significantly underestimated. We detail possible explanations for this discrepancy, including interactions between the dynamical tide of the host star and the tidal bulge and possible non-synchronous rotation of the host star.