I present the results of a comprehensive assessment of companions to 454 solar-type stars within 25 pc. New observational aspects of this work include surveys for (1) very close companions with long-baseline interferometry at the Center for High Angular Resolution Astronomy (CHARA) Array, (2) close companions with speckle interferometry, and (3) wide proper motion companions identified by blinking multi-epoch archival images. I have also obtained and included unpublished results from extensive radial velocity monitoring programs. The many sources utilized enable a thorough evaluation of stellar and brown dwarf companions. The results presented here include eight new companion discoveries, four of which are wide common proper motion pairs discovered by blinking archival images, and four more are from the spectroscopic data. The overall observed fractions of single, double, triple, and higher order systems are 57%±3%, 33%±2%, 8%±1%, and 3%±1%, respectively, counting all stellar and brown dwarf companions. The incompleteness analysis indicates that only a few undiscovered companions remain in this well-studied sample, showing that a majority of the solar-type stars are single. Bluer, more massive stars are more likely to have companions than redder, less massive ones. I confirm earlier expectations that more active stars are more likely to have companions. A preliminary, but important indication is that brown dwarfs, like planets, prefer stars with higher metallicity, tentatively suggesting that brown dwarfs may form like planets when they are companions to stars. The period distribution is unimodal and roughly Gaussian with peak and median values of about 300 years. The period-eccentricity relation shows a roughly flat distribution beyond the circularization limit of about 12 days. The mass-ratio distribution shows a clear discontinuity near a value of one, indicating a preference for twins, which are not confined to short orbital periods, suggesting that stars form by multiple formation mechanisms. The ratio of planet hosts among single, binary, and multiple systems are statistically indistinguishable, suggesting that planets are as likely to form around single stars as they are around components of binary or multiple systems at sufficiently wide separations.

Les binaires X, composées d'un objet compact (naine blanche, étoile à neutrons ou trou noir stellaire) et d'une étoile donneuse, sont des objets fascinants qui ont permis faire des découvertes majeures dans le domaine de l'astrophysique des hautes énergies. Toutefois, l'environnement immédiat de ces dernières, à l'échelle de $\sim$ 100--10,000 unités astronomiques, n'est pour sa part que très peu compris et exploré. Dans ce mémoire, on s'intéresse à la présence de compagnons, autant de masse planétaire que stellaire, dans lesdits environnements des binaires X. Pour ce faire, on a fait appel à des techniques observationnelles qui sont normalement utilisées pour la recherche d'exoplanètes jeunes, chaudes et en orbite autour d'étoiles situées à quelques parsecs de la Terre : il s'agit de l'imagerie directe ou plus précisément de l'imagerie à haut contraste. Ainsi, pour la toute première fois, on a appliqué ces méthodes sur un échantillon de binaires X avec l'instrument proche infrarouge NIRC2 de l'Observatoire W. M. Keck, avec l'aide d'un système d'optique adaptative et d'un coronographe. Ici, on a présenté les premières images à haut contraste de 14 binaires X. Le but était de détecter des sources lumineuses dans les images et de calculer leur flux et leur position avec des méthodes d'optimisation. Ensuite, il s'agissait de définir via diverses analyses si celles-ci sont cohérentes avec des compagnons liés par la gravité à la binaire X, ou plutôt avec des étoiles de fond. Ces travaux se veulent donc une introduction à un nouveau sous-domaine de l'astrophysique qui tente de relier l'exoplanétologie et l'astrophysique des hautes énergies. / X-ray binaries, consisting of a compact object (white dwarf, neutron star or stellar-mass black hole) and a donor star, are fascinating objects that have allowed major breakthroughs in the field of high-energy astrophysics. However, their immediate environments, on the scale of $\sim$ 100--10,000 astronomical units, are still poorly understood. In this Master's thesis, we investigated the presence of companions, ranging from planetary to stellar masses, in the environments of X-ray binaries. In order to do so, we used observational techniques that are normally used for the search of young, hot and distant exoplanets orbiting stars located at a few parsecs from the Earth: direct imaging or more precisely high-contrast imaging. Thus, for the very first time, we applied these techniques on a sample of X-ray binaries with the near-infrared instrument NIRC2 of the W. M. Keck Observatory, with the help of an adaptive optics system and a coronagraph. Here, we present the first high-contrast images for 14 X-ray binaries. The goal was to detect point sources in the images and to calculate their flux and position with optimization methods. Afterward, we determined via a variety of analyses if these were consistent with companions gravitationally bounded to the X-ray binary, or rather with background stars. This work acts as an introduction, albeit exploratory, to a new subfield of astrophysics that attempts to link exoplanetology and high-energy astrophysics.