Direct imaging search for planets around low-mass stars and spectroscopic characterization of young exoplanets

Abstract

Low-mass stars between 0.1-0.6M are the most abundant members our galaxy and may be the most common sites of planet formation, but little is known about the outer architecture of their planetary systems. We have carried out a high-contrast adaptive imaging search for gas giant planets between 1-13 MJup around 122 newly identified young M dwarfs in the solar neighborhood ( < 35 pc). Half of our targets are younger than 145 Myr, and 90% are younger than 580 Myr. After removing 39 resolved stellar binaries, our homogeneous sample of 83 single young M dwarfs makes it the largest imaging search for planets around low-mass stars to date. Our H-and K-band coronagraphic observations with Subaru/HiCIAO and Keck/NIRC2 achieve typical contrasts of 9{13 mag and 12-14 mag at 100, respectively, which corresponds to limiting masses of ~1-10 MJup at 10-30 AU for most of our sample. We discovered four brown dwarfs with masses between 25-60 MJup at projected separations of 4-190 AU. Over 100 candidate planets were discovered, nearly all of which were found to be background stars from follow-up second epoch imaging. Our null detection of planets nevertheless provides strong statistical constraints on the occurrence rate of giant planets around M dwarfs. Assuming circular orbits and a logarithmically-at power law distribution in planet mass and semi-major axis of the form d2N=(dloga dlogm) / m0a0, we measure an upper limit (at the 95% confidence level) of 8.8% and 12.6% for 1-13 MJup companions between 10-100 AU for hot start and cold start evolutionary models, respectively. For massive gas giant planets in the 5-13 MJup range like those orbiting HR 8799, GJ 504, and ß Pictoris, we find that fewer than 5.3% (7.8%) of M dwarfs harbor these planets between 10-100 AU for a hot start (cold start) formation scenario. Our best constraints are for brown dwarf companions; the frequency of 13-75 MJup companions between (de-projected) physical separations of 10-100 AU is 2.1+2:1-1:2%. Altogether, our results show that gas giant planets, especially massive ones, are rare in the outskirts of M dwarf planetary systems. If disk instability is a viable way to form planets, our constraints for the most common type of star imply that overall it is an inefficient mechanism.