2010MNRAS.404.1835K

Over the past few decades, numerous wide (>10³ au) binaries in the Galactic field and halo have been discovered. Their existence cannot be explained by the process of star formation or by dynamical interactions in the field, and their origin has long been a mystery. We explain the origin of these wide binaries by formation during the dissolution phase of young star clusters: an initially unbound pair of stars may form a binary when their distance in phase space is small. Using N-body simulations, we find that the resulting wide binary fraction in the semimajor axis range 10³au < a < 0.1pc for individual clusters is 1-30 per cent, depending on the initial conditions. The existence of numerous wide binaries in the field is consistent with observational evidence that most clusters start out with a large degree of substructure. The wide binary fraction decreases strongly with increasing cluster mass, and the semimajor axis of the newly formed binaries is determined by the initial cluster size. The resulting eccentricity distribution is thermal, and the mass ratio distribution is consistent with gravitationally focused random pairing. As a large fraction of the stars forms in primordial binaries, we predict that a large number of the observed `wide binaries' are in fact triple or quadruple systems. By integrating over the initial cluster mass distribution, we predict a binary fraction of a few per cent in the semimajor axis range 10³au < a < 0.1pc in the Galactic field, which is smaller than the observed wide binary fraction. However, this discrepancy may be solved when we consider a broad range of cluster morphologies.