No Arabic abstract
Stellar companions can influence the formation and evolution of planetary systems, but there are currently few observational constraints on the properties of planet-hosting binary star systems. We search for stellar companions around 77 transiting hot Jupiter systems to explore the statistical properties of this population of companions as compared to field stars of similar spectral type. After correcting for survey incompleteness, we find that $47%pm7%$ of hot Jupiter systems have stellar companions with semi-major axes between 50-2000 AU. This is 2.9 times larger than the field star companion fraction in this separation range, with a significance of $4.4sigma$. In the 1-50AU range, only $3.9^{+4.5}_{-2.0}%$ of hot Jupiters host stellar companions compared to the field star value of $16.4%pm0.7%$, which is a $2.7sigma$ difference. We find that the distribution of mass ratios for stellar companions to hot Jupiter systems peaks at small values and therefore differs from that of field star binaries which tend to be uniformly distributed across all mass ratios. We conclude that either wide separation stellar binaries are more favorable sites for gas giant planet formation at all separations, or that the presence of stellar companions preferentially causes the inward migration of gas giant planets that formed farther out in the disk via dynamical processes such as Kozai-Lidov oscillations. We determine that less than 20% of hot Jupiters have stellar companions capable of inducing Kozai-Lidov oscillations assuming initial semi-major axes between 1-5 AU, implying that the enhanced companion occurrence is likely correlated with environments where gas giants can form efficiently.
Surveys of nearby field stars indicate that stellar binaries are common, yet little is known about the effects that these companions may have on planet formation and evolution. The Friends of Hot Jupiters project uses three complementary techniques to search for stellar companions to known planet-hosting stars: radial velocity monitoring, adaptive optics imaging, and near-infrared spectroscopy. In this paper, we examine high-resolution K band infrared spectra of fifty stars hosting gas giant planets on short-period orbits. We use spectral fitting to search for blended lines due to the presence of cool stellar companions in the spectra of our target stars, where we are sensitive to companions with temperatures between 3500-5000 K and projected separations less than 100 AU in most systems. We identify eight systems with candidate low-mass companions, including one companion that was independently detected in our AO imaging survey. For systems with radial velocity accelerations, a spectroscopic non-detection rules out scenarios involving a stellar companion in a high inclination orbit. We use these data to place an upper limit on the stellar binary fraction at small projected separations, and show that the observed population of candidate companions is consistent with that of field stars and also with the population of wide-separation companions detected in our previous AO survey. We find no evidence that spectroscopic stellar companions are preferentially located in systems with short-period gas giant planets on eccentric and/or misaligned orbits.
The existence of hot Jupiters has challenged theories of planetary formation since the first extrasolar planets were detected. Giant planets are generally believed to form far from their host stars, where volatile materials like water exist in their solid phase, making it easier for giant planet cores to accumulate. Several mechanisms have been proposed to explain how giant planets can migrate inward from their birth sites to short-period orbits. One such mechanism, called Kozai-Lidov migration, requires the presence of distant companions in orbits inclined by more than $sim40$ degrees with respect to the plane of the hot Jupiters orbit. The high occurrence rate of wide companions in hot Jupiter systems lends support to this theory for migration. However, the exact orbital inclinations of these detected planetary and stellar companions is not known, so it is not clear whether the mutual inclination of these companions is large enough for the Kozai-Lidov process to operate. This paper shows that in systems orbiting cool stars with convective outer layers, the orbits of most wide planetary companions to hot Jupiters must be well aligned with the orbits of the hot Jupiters and the spins of the host stars. For a variety of possible distributions for the inclination of the companion, the width of the distribution must be less than $sim20$ degrees to recreate the observations with good fidelity. As a result, the companion orbits are likely well-aligned with those of the hot Jupiters, and the Kozai-Lidov mechanism does not enforce migration in these systems.
In this paper we search for distant massive companions to known transiting gas giant planets that may have influenced the dynamical evolution of these systems. We present new radial velocity observations for a sample of 51 planets obtained using the Keck HIRES instrument, and find statistically significant accelerations in fifteen systems. Six of these systems have no previously reported accelerations in the published literature: HAT-P-10, HAT-P-22, HAT-P-29, HAT-P-32, WASP-10, and XO-2. We combine our radial velocity fits with Keck NIRC2 adaptive optics (AO) imaging data to place constraints on the allowed masses and orbital periods of the companions responsible for the detected accelerations. The estimated masses of the companions range between 1-500 M_Jup, with orbital semi-major axes typically between 1-75 AU. A significant majority of the companions detected by our survey are constrained to have minimum masses comparable to or larger than those of the transiting planets in these systems, making them candidates for influencing the orbital evolution of the inner gas giant. We estimate a total occurrence rate of 51 +/- 10% for companions with masses between 1-13 M_Jup and orbital semi-major axes between 1-20 AU in our sample. We find no statistically significant difference between the frequency of companions to transiting planets with misaligned or eccentric orbits and those with well-aligned, circular orbits. We combine our expanded sample of radial velocity measurements with constraints from transit and secondary eclipse observations to provide improved measurements of the physical and orbital characteristics of all of the planets included in our survey.
Multi-star systems are common, yet little is known about a stellar companions influence on the formation and evolution of planetary systems. For instance, stellar companions may have facilitated the inward migration of hot Jupiters towards to their present day positions. Many observed short period gas giant planets also have orbits that are misaligned with respect to their stars spin axis, which has also been attributed to the presence of a massive outer companion on a non-coplanar orbit. We present the results of a multi-band direct imaging survey using Keck NIRC2 to measure the fraction of short period gas giant planets found in multi-star systems. Over three years, we completed a survey of 50 targets (Friends of Hot Jupiters) with 27 targets showing some signature of multi-body interaction (misaligned or eccentric orbits) and 23 targets in a control sample (well-aligned and circular orbits). We report the masses, projected separations, and confirmed common proper motion for the 19 stellar companions found around 17 stars. Correcting for survey incompleteness, we report companion fractions of $48%pm9%$, $47%pm12%$, and $51%pm13%$ in our total, misaligned/eccentric, and control samples, respectively. This total stellar companion fraction is $2.8,sigma$ larger than the fraction of field stars with companions approximately $50-2000,$AU. We observe no correlation between misaligned/eccentric hot Jupiter systems and the incidence of stellar companions. Combining this result with our previous radial velocity survey, we determine that $72% pm 16%$ of hot Jupiters are part of multi-planet and/or multi-star systems.
The Jovian-sized object WD~1856~b transits a white dwarf (WD) in a compact $1.4$-day orbit. Unlikely to have endured stellar evolution in its current orbit, WD~1856~b is thought to have migrated from much wider separations. Because the WD is old, and a member of a well-characterized hierarchical multiple, the well-known Kozai mechanism provides an effective migration channel for WD~1856~b. Moreover, the lack of tides in the star allows us to directly connect the current semi-major axis to the pre-migration one, from which we can infer the initial conditions of the system. By further demanding that successful migrators survive all previous phases of stellar evolution, we are able to constrain the mass of WD~1856~b to be $simeq0.7-3M_{rm J}$ and its main sequence semi-major axis to be $simeq 2-2.5$ au. These properties imply that WD~1856~b was born a typical gas giant. We further estimate the occurrence rate of Kozai-migrated planets around WDs to be ${cal O}(10^{-3}{-}10^{-4})$, suggesting that WD~1856~b is the only one in the {it TESS} sample, but implying ${cal O}(10^2)$ future detections by LSST. In a sense, WD~1856~b was an ordinary Jovian planet that underwent an extraordinary dynamical history.