Giant Planets in Open Clusters S.N. Quinn 1 , 3 , R.J. White 1 , D.W. - PDF document

Twenty years of giant exoplanets - Proceedings of the Haute Provence Observatory Colloquium, 5-9 October 2015 Edited by I. Boisse, O. Demangeon, F. Bouchy & L. Arnold Giant Planets in Open Clusters S.N. Quinn 1 , 3 , R.J. White 1 , D.W. Latham 2 Talk given at OHP-2015 Colloquium 1 Department of Physics & Astronomy, Georgia State University, 25 Park Place NE Suite 605, Atlanta, GA 30316, USA ( quinn@astro.gsu.edu ) 2 Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138, USA 3 National Science Foundation Graduate Research Fellow Abstract Two decades after the discovery of 51 Peg b, more than 200 hot Jupiters have now been con- firmed, but the details of their inward migration remain uncertain. While it is widely accepted that short period giant planets could not have formed in situ , several di ff erent mechanisms (e.g., Type II migration, planet-planet scattering, Kozai-Lidov cycles) may contribute to shrinking planetary or- bits, and the relative importance of each is not well-constrained. Migration through the gas disk is expected to preserve circular, coplanar orbits and must occur quickly (within ∼ 10 Myr), whereas multi-body processes should initially excite eccentricities and inclinations and may take hundreds of millions of years. Subsequent evolution of the system (e.g., orbital circularization and inclina- tion damping via tidal interaction with the host star) may obscure these di ff erences, so observing hot Jupiters soon after migration occurs can constrain the importance of each mechanism. Fortunately, the well-characterized stars in young and adolescent open clusters (with known ages and composi- tions) provide natural laboratories for such studies, and recent surveys have begun to take advantage of this opportunity. We present a review of the discoveries in this emerging realm of exoplanet sci- ence, discuss the constraints they provide for giant planet formation and migration, and reflect on the future direction of the field. 1 Introduction Open clusters have long provided key observational constraints in the field of stellar astrophysics due to their unique properties. Having formed at the same time (and the same distance), the ensemble properties of the stars in clusters allow one to determine their ages more precisely than is possible for typical field stars. Moreover, because they formed from the same cloud, the stars in a cluster have approximately the same metal abundance, so di ff erences between stars within a cluster arise primarily as a function of stellar mass. As a result, open clusters enable direct observation of billions of years of stellar evolution (e.g., structure, rotation, activity) across a range of stellar mass. This allows us to conduct astrophysical experiments under conditions controlled for age, mass, composition, and even a known dynamical environment — such as stellar binary fraction and space density. With the discoveries of thousands of exoplanets over the past two decades 1 in a wide range of environments — including hot Jupiters, planets orbiting early-type stars, M dwarfs, giant stars, white dwarfs, in binary systems, and even circumbinary planets — it has become clear that planets are an expected byproduct of star formation. Given that planets are common (e.g., Fressin et al. 2013; Mayor et al. 2011) and most stars form in a clustered environment (e.g., Lada & Lada 2003; Bressert et al. 2010), it is reasonable to expect that planets exist in open clusters. This is an exciting prospect, because that means we can begin to use clusters to study planetary evolution in much the same way that they have been used to study stellar evolution. Planets found in clusters will immediately be among those with the best determined ages, and because open clusters dissipate into the field over time, they tend to be younger than field stars — many have ages < 1 Gyr. It is particularly interesting to note that these ages are similar to the expected timescales for many of the drastic changes that can occur in planetary systems — e.g., 1 http: // www.exoplanets.org 29

Twenty years of giant exoplanets - Proceedings of the Haute Provence Observatory Colloquium, 5-9 October 2015 Edited by I. Boisse, O. Demangeon, F. Bouchy & L. Arnold planetary migration, tidal interaction between short-period planets and their host stars, disruptive encounters with passing stars — as well as physical changes to the planets themselves — such as the convergence of cooling tracks for di ff erent giant planet and brown dwarf formation models, or atmospheric escape for short-period small planets with gaseous envelopes. We can therefore use the properties of young cluster planets to gain insight into the above processes and the mechanisms that drive them. For example, while hot Jupiters likely formed beyond the ice line and then migrated to within a fraction of an AU of their host stars, the primary channel by which they migrate is unknown. There are many mechanisms that could cause a giant planet to migrate, and two leading ideas are Type II migration (Goldreich & Tremaine 1980, through a gas disk) and high eccentricity migration (HEM) via multi-body interactions, such as planet-planet scattering (Rasio & Ford 1996) or Kozai-Lidov cycles (e.g., Fabrycky & Tremaine 2007). Type II migration must occur while the gas disk is present ( ∼ 10 Myr) and is expected to preserve near-circular orbits well-aligned with the stellar spin axis. Conversely, if HEM is the primary driver of migration, it may take hundreds of millions of years and most hot Jupiters should initially possess non-zero eccentricity and orbital inclination (before subsequent tidal circularization or realignment can occur). Therefore, one way to distinguish between migration mechanisms is to observe systems during, or shortly after, the migration process. If the hot Jupiter occurrence rate is low for stars that are tens to hundreds of millions of years old, we can rule out disk migration as the dominant mechanism. Similarly, if most young planets have eccentric, inclined orbits, HEM has likely played a large role. Until recently, searches for hot Jupiters in clusters were unsuccessful, despite numerous e ff orts with su ffi cient sensitivity, employing both radial velocities (Paulson et al. 2004; Pasquini et al. 2012) and transit photometry (e.g., Bramich et al. 2005; Mochejska et al. 2006, 2008; Rosvick & Robb 2006; Montalto et al. 2007; Pepper et al. 2008; Miller et al. 2008; Hartman et al. 2009). While these results initially seem to be in opposition to the ubiquitous presence of planets, hot Jupiters are relatively more rare (orbiting ∼ 1% of Sun-like stars; Wright et al. 2012; Mayor et al. 2011), and van Saders & Gaudi (2011) calculate that all combined transit surveys in clusters simply had not surveyed enough stars to guarantee a detection. In this paper, we review the more recent searches for — and discoveries of — planets in clusters and discuss the conclusions that can be drawn from these surveys. We particularly highlight giant planet migration as one topic to which cluster planets can contribute new constraints, and we briefly consider potential future exoplanet studies to be carried out in clusters. 200 400 300 150 Radial Velocity (m s -1 ) Radial Velocity (m s -1 ) Radial Velocity (m s -1 ) 200 100 200 0 50 100 0 -200 -50 0 -400 45 55 30 O-C O-C O-C 0 0 0 0 0 0 -45 -55 -30 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Orbital Phase Orbital Phase Orbital Phase Figure 1: The first two hot Jupiters in an open cluster, Pr0201 b and Pr0211 b (left and middle Quinn et al. 2012), and a Hyades hot Jupiter (right Quinn et al. 2014). The top panels show the radial velocities phased to the orbital periods of 4 . 4264, 2 . 1415, and 6 . 09708 days, and the bottom panels show the residuals to the best fit orbit. 2 Two ‘b’s in the Beehive and the new field of cluster planets While massive companions had been found previously in open clusters (Sato et al. 2007; Lovis & Mayor 2007), these planets orbit evolved stars of intermediate mass, so they cannot constrain planet occurrence around Sun-like stars in clusters, and their long periods do not provide insight into migration. The discovery of two hot Jupiters in the Praesepe open cluster (also called the Beehive; Quinn et al. 2012, see Fig.1) provided the first evidence that giant planets do form and migrate in dense stellar environments and suggested that they do so at a rate similar to 30