Exoplanet mass measurement : adaptive optics follow-up of microlensing events

Microlensing is a powerful tool that can be used to search for cold exoplanets in the disk and bulge of the Milky Way. Over the past 16 years ∼100 planets have been discovered using this method, but moving from modelled parameters to physical properties is difficult. I refine and apply a method of using high-angular resolution near-infrared observations from large ground based observatories to find more precise constraints on the mass and distances of these exoplanets and their stellar hosts. Microlensing events rely on the chance alignment of source and lens stars along our line of sight towards dense star fields, usually the galactic bulge at the centre of our galaxy. The relative proper motion of these two stars is such that adaptive optics (AO) observations on 8-10m class telescopes can be used measure the lens flux, or even entirely resolve them when observed at later epochs. In this thesis I present three studies of microlensing events augmented with AO photometry obtained from the Keck observatory. The first is a reanalysis of the microlensing event MACHO-97-BLG-28. Suspected to be a case where the source and lens stars could be resolved with photometry from Keck, I showed it was not. I revise estimates of the lens mass and distance by re-modelling the event light curve using additional data from the Mt. Stromlo 74” telescope, model estimates of stellar limb darkening, and fitting of the blend separately for each telescope and passband. This slightly favours the conclusion that the event is a stellar M-dwarf binary. I then use Keck photometry to constrain the mass-distance of OGLE-2017-BLG-1434, which is a cold super-Earth planet and the eighth microlensing planet discovered with a planet host star mass ratio q < 1 x 10\(^{-4}\). This planet confirms the turnover in the cold-planet mass ratio function. I obtained 60 mas images of the event and by comparing measured near-IR magnitudes with the predicted source magnitude from the light curve model we reduce the error in the lens mass and distances by more than 50%. This confirms the event as a close super-Earth at a distance of 860 ± 50 pc. Finally, I present the first detection of a white dwarf hosting a gas giant planet. This detection was made in event MOA-2010-BLG-447 using Keck H-band images taken at two epochs after the time of maximum magnification when the proper motion was such that the lens and source should be resolved. A non-detection of a main-sequence lens well above the detection limit of the Keck photometry indicates the event to be a 0.58 solar mass white dwarf orbited by a 1.3 Jupiter mass planet. These three studies contribute to the measurement of the cold planet mass function. The super-Earth OGLE-2017-BLG-1434 lies near the break of this function, while MOA- 2010-BLG-477 is the first solid detection of a single white dwarf with a planetary companion. These studies and the methods herein are a pathfinder for the Nancy Grace Roman Space Telescope and the mass measurement method which will be used for 80%+ of the microlensing targets it will observe