deeplink to this page


A low-mass X-ray binary
X-ray binaries (XRBs) are binary stars in which a donor star transfers mass to a neutron star or black hole. In the accretion process the transferred gas is accelerated and heated, and emits X-rays. These highly-energetic X-rays are visible with X-ray satellites, such as ROSAT, XMM-Newton and Chandra. X-ray binaries are divided in low-mass and high-mass X-ray binaries (LMXBs and HMXBs), where the mass refers to the mass of the companion star (donor) of the neutron star or black hole.

We studied the formation of ultra-compact X-ray binaries (UCXBs). These stars are LMXBs with orbital periods less than about an hour. A relatively large fraction of observed UCXBs is observed in the densely populated globular clusters. There are two suggested scenarios to form these systems. Firstly, they could be formed through stable mass transfer, under the influence of strong angular-momentum loss, e.g. due to a process called magnetic braking. This will allow the system to decrease its orbital period while mass transfer is going on. In the second scenario, either a common-envelope phase in a binary with a giant and a neutron star, or envelope injection due to a direct stellar collision between a giant and a neutron star, would bring the core of the giant (a white dwarf) and the neutron star in a close orbit. This orbit will then decay due to the emission of gravitational radiation, until the orbital period is a few minutes, and the white dwarf will start mass transfer onto the neutron star, and an ultra-compact X-ray binary is formed.

An ultra-compact X-ray binary
The UCXB with the shortest period observed (11 minutes) seems to have a shrinking orbit, which suggests that the first scenario explains its formation. However, we investigated whether these observed systems can indeed be formed by stable mass transfer during which the orbit of the binary shrinks due to magnetic braking. The conclusion of our calculations is that this magnetic capture scenario can work in principle, but requires initial conditions too specific to occur frequently in nature. This must mean that the observed systems are formed in the collision-scenario instead (the common-envelope scenario requires a massive companion star, which are no longer present in the old globular-cluster population). The apparent shrinkage of the orbit of the 11-minute system can then be attributed to the acceleration of the system in the cluster potential, as was suggested earlier. These results were published in Van der Sluys et al. (2005a) and Van der Sluys et al. (2005b).

Closely related to this is a project that concerns the formation of the (probably ultra-compact) X-ray binary 2S 0918-549. This binary is observed in the field (rather than in a globular cluster), and was therefore probably formed in a common envelope. For field-stars, a collision is practically impossible, because of the low stellar density, while the field population is much younger than that in globular clusters, so that stars massive enough for the common-envelope scenario to work are still present. Observations of 2S 0918-549 indicate a high abundance of neon, compared to that of oxygen. We show that the neon-to-oxygen ratio for this system is naturally explained when we assume that the donor star in the system is a helium white dwarf, and hence the common envelope happened on the red-giant branch (In `t Zand et al, 2005).

The images on this page were made by the author, using Rob Hynes' visualisation program BinSim.