Abstract
Filling out the dearth of detections between direct imaging and radial velocity surveys will test theories of planet formation and (sub)stellar binarity across the full range of semi-major axes, connecting formation of close to wide separation gas giants and sub-stellar companions. Direct detection of close-in companions is notoriously dicult as classical techniques fail near the diffraction limit. In this dissertation, I present a new pipeline which searches for faint companions using kernel-phase interferometry (KPI), a technique utilizing the full unobscured telescope aperture. I demonstrate the pipeline, and the power of KPI, by performing two surveys and accompanying demographic analyses.
First, I search for companions in the entire HST/NICMOS F110W and F170M image archive of nearby brown dwarfs, demonstrating significant sensitivity to companions at half the diffraction limit. I discover no new companions but recover and refine astrometry of 19 previous imaging companions and confirm two previous kernel-phase detections. I also present contrast curves to enable a population study and to demonstrate the strength of this technique at separations inaccessible to classical imaging techniques.
Second, after estimating physical properties of a subset of this sample, I use a Bayesian framework to compare these detections and detection limits to a model companion population. When correcting for Malmquist bias, I find a smaller population of companions with tighter separations than seen in previous studies. This is due to our use of KPI, which enables us to resolve the peak of the semimajor axis distribution with significant sensitivity to low-mass companions. I confirm the previously-seen trends of decreasing binary fraction with decreasing mass and a strong preference for tight and equal-mass systems in the field-age sub-stellar regime; only ~1% of systems are wider than 20 au or have a mass-ratio q < 0.6.
Third, I present the results of a second KPI based survey of sub-stellar objects in Taurus and Upper-Scorpius using archival HST/ACS. I present 6 new candidate detections at extremely tight separations, in addition to previously known companions. I find an observed companion frequency of F = 0.24+0.10-0.08 which is slightly higher than previous studies, consistent with obs 0.08 our new detections.
Combining these two surveys and demographic analyses, I find a significant excess of young wide (> 10 au) companions compared to the field. I also find that the field population of wide companions is consistent with my population of systems formed in low density regions if it were to be diluted by single systems formed in high density regions. This is consistent with dynamical evolution preferentially dissolving systems born in high-density star-forming regions over those in low-density regions. I attribute the characteristics of the BD binary population to turbulent fragmentation setting the initial conditions followed by a brief period of dynamical evolution, removing the widest and lowest-mass companions, before the birth cluster dissolves.
