Eclipsing binaries and fast rotators in the Kepler sample: Characterization via radial velocity analysis from Calar Alto ∗ ∗ ∗

Context. The Kepler mission has searched for planetary transits in more than two hundred thousand stars by obtaining very accurate photometric data over a long period of time. Among the thousands of detected candidates, the planetary nature of around 15% has been established or validated by different techniques. But additional data are needed to characterize the rest of the candidates and reject other possible configurations. Aims: We started a follow-up program to validate, confirm, and characterize some of the planet candidates. In this paper we present the radial velocity analysis of those that present large variations, which are compatible with being eclipsing binaries. We also study those showing high rotational velocities, which prevents us from reaching the necessary precision to detect planetary-like objects. Methods: We present new radial velocity results for 13 Kepler objects of interest (KOIs) obtained with the CAFE spectrograph at the Calar Alto Observatory and analyze their high-spatial resolution (lucky) images obtained with AstraLux and the Kepler light curves of some interesting cases. Results: We have found five spectroscopic and eclipsing binaries (group A). Among them, the case of KOI-3853 is of particular interest. This system is a new example of the so-called heartbeat stars, showing dynamic tidal distortions in the Kepler light curve. We have also detected duration and depth variations of the eclipse. We suggest possible scenarios to explain such an effect, including the presence of a third substellar body possibly detected in our radial velocity analysis. We also provide upper mass limits to the transiting companions of six other KOIs with high rotational velocities (group B). This property prevents the radial velocity method from achieving the necessary precision to detect planetary-like masses. Finally, we analyze the large radial velocity variations of two other KOIs, which are incompatible with the presence of planetary-mass objects (group C).These objects are likely to be stellar binaries. However, a longer timespan is needed to complete their characterization. Based on observations collected at the German-Spanish Astronomical Center, Calar Alto, jointly operated by the Max-Planck Institut fur Astronomie (Heidelberg) and the Instituto de Astrofísica de Andalucía (IAA-CSIC, Granada).Appendix A is available in electronic form at http://www.aanda.org