The outcome of massive star formation

Massive stars play a crucial role in the Universe. They shape their surroundings by injecting large amounts of energy and momentum and they produce new, heavy elements that are the building blocks of new stars, planets, and life. They are usually observed in close binaries. Due to the lack of observations covering the earliest stages of their lives, the formation process of massive (binary) stars is poorly understood. In this thesis we present observational studies of the outcome of massive star formation. We present the first spectroscopically confirmed population of massive pre-main sequence stars in the giant HII region M17, where their temperature, luminosity, radius, and projected radial velocity are measured. We also discuss their multiplicity properties and find that the young stars in M17 show a very low radial velocity dispersion in comparison to somewhat older stellar clusters of similar mass. We propose that massive stars are formed in binaries with wide orbits that shrink in the first few million years of evolution. By studying the young populations in three other Galactic star forming regions we find a possible relation between the age of the clusters and the radial velocity dispersion of their massive stars. The latter supports the hypothesis that massive stars form in wide orbits. We present a study of diffuse interstellar bands (DIBs) towards M17 and find a relation of the DIB strength with the grain size distribution of the interstellar medium. Finally, we present two ongoing projects to test the wide binary hypothesis.