FRIENDS OF HOT JUPITERS. I. A RADIAL VELOCITY SEARCH FOR MASSIVE, LONG-PERIOD COMPANIONS TO CLOSE-IN GAS GIANT PLANETS

In this paper we search for distant massive companions to known transiting hot Jupiters that may have influenced the dynamical evolution of these systems. We present new radial velocity observations for a sample of 51 hot Jupiters obtained using the Keck HIRES instrument, and use these observations to search for long-term radial velocity accelerations. We find new, statistically significant accelerations in seven systems, including: HAT-P-10, HAT-P-20, HAT-P-22, HAT-P-29, HAT-P-32, WASP-10, and XO-2. We combine our radial velocity fits with Keck NIRC2 AO imaging data to place constraints on the allowed masses and orbital periods of the companions. 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 short-period hot Jupiters in these systems, making them candidates for influencing the orbital evolution of the inner hot Jupiters. They also appear to occur preferentially in systems with more metal-rich host stars, and with typical orbital separations that are larger than those of multi-planet systems without hot Jupiters. We estimate a total occurrence rate of 55% +11% / -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 in hot Jupiter systems 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 hot Jupiters included in our survey.