Highly eccentric but non-eclipsing binary systems have been observed by the Kepler mission, so far dubbed as 'Heartbeat Stars' due to the odd resemblance of their light curves to a classic medical EKG monitor. In these cases the periodic brightening or dimming variability is caused not by massive stellar internal organs but instead from tidally induced distortions caused by the companion star as it makes its close pericenter passage; these 'eccentricity pulses' have been observed to cause a wide range of effects on the light curve. In cases of dimming these distortions are often not in the same magnitude league as a true eclipse or transit, allowing them to be missed by initial pipeline transit searches. They can sometimes resemble types of stellar variation or pulsation, but their periodic nature sets them apart.
Literature from
Dong, et al and
Welsh, et al explored an initial set of benchmark systems (including the amazing target KOI-54); more were identified in the paper '
A Class of Eccentric Binaries with Dynamic Tidal Distortions Discovered with Kepler' by Thompson, et al. Hot stars in the
A and
F class ranges seem to lend themselves to betraying these signals and make up a majority of the identified group. A crowd sourced effort at PlanetHunters, where manual search methods were already under way for exoplanets lost in the noise, quickly began to
identify additional eccentric binary candidates.
Not all Heartbeat stars encountered so far fit the
A to
F class model however, with a small subset of new candidates being found in the
G and
K type range. Three such departures of interest I've highlighted below are found associated with candidate targets KIC 8210370, 8095275 and 7431665; all listed as Type
K Giants with radius far exceeding that of Sol and effective temperatures below many of their previous Kepler listed HB counterparts.
KIC 8210370
A Type K with radius 10.7x Sol and a period of ~154 days:
"Cannot see any secondaries but find it unlikely given the stellar
parameters that these are going to be similar sized objects. It may
well be an EB, but I think it could instead be a Heartbeat Binary.
We've seen a few now. The properties of Giant stars make them very
suitable for observable heartbeat effects, which probably explains why
we have a not-insignificant number given how Giants are a relatively
small subset of Kepler targets." -PlanetSam
Source
KIC 8095275
A Type K Giant with radius 6.8x Sol and a period of ~45 days:
"The spike after the eclipse is now very apparent. There doesn't seem
to be a corresponding one before the minima. The waveform looks very
much like a heartbeat binary - tidally excited pulsations caused by a
slightly eccentrically orbiting body. But the orbit doesn't look
eccentric, otherwise the minimas would show the phase difference, and
the primary is a Type K giant instead of a Type A.
One possibility is that the period is actually ~23d and there is only
one eclipse, the other being out of line-of-sight, and the orbiting
object indeed causes some pulsation that results in the shoulder at the
egress of the eclipse." -Kianjin
Source
KIC 7431665
A Type K Giant with radius 9.0x Sol and a period of ~281 days
"...there are asymmetrical bulges on either limb, very reminiscent of SPH10004034
or KID 10614012. It is possible that this is actually due to a
heartbeat-like tidal interaction that deforms the giant star, i.e. the
secondary (which is a small red dwarf perhaps) is in an extremely
eccentric orbit that causes this when it is close to periastron." -Kianjin
Source
Further Info:
Heartbeat Stars Presentation by Susan Thompson
