First Direct Identification of a Multi-Star Microlens System Hosting a Planet

Abstract

Nearly 25 years have passed since gravitational microlensing delivered its first exoplanet discoveries. The relative proper motion between lens and source systems allows them to be resolved as they separate on the sky over many years. This enables scientists to study light coming directly from the lenses and place tight constraints on the physical properties of these planetary systems. In this paper we present an analysis of high angular resolution images of the microlensing target OGLE-2006-BLG-284 using Keck adaptive optics and HST data taken more than 15 years after the event. We clearly detect the lens system and measure its motion across the Keck and HST epochs. We measure a lens-source relative proper motion of $5.80 \pm 0.15$ mas yr$^{-1}$. A combination of the light curve parameters and the host star(s) brightnesses in $V,\, I$, and $K$ passbands gives primary and secondary host star masses of $M_{\rm 1} = 0.43 \pm 0.05\,M_{\odot}$ and $M_{\rm 2} = 0.12 \pm 0.02\,M_{\odot}$, and a planetary companion mass of $m_p = 165^{+40}_{-33}\,M_{\oplus}$. Although we are unable to determine if the planet orbits one or both stars in the system, our analysis motivates next-generation high-precision RV measurements that can distinguish these two possibilities. This is the first direct detection of flux coming from a two-star microlensing system which hosts a planet.