Ultra Compact Binary Systems

MSSL has been at the forefront of the discovery of binary systems called ultra-compact binary systems: binaries with orbital periods of ~10 mins or less. The first of these, RX J1914+24, was discovered using the ROSAT X-ray satellite. Cropper et al (1998) showed that only one period (9.5mins) was detected and that its folded X-ray light curve was off for half its cycle. This lead them to suggest that it was a double degenerate polar - a polar with a white dwarf, rather than a main-sequence star, as a secondary. Further observations, Ramsay et al (2000, 2002), still showed no other evidence for a second period and also reported the optical counterpart. This was important since the optical light curve is anti-phased with the X-ray data. Further, no strong emission lines were detected.

Our discovery of the optical counterpart to the second ultra-compact system, RX J0806+15, caused a lot of interest in the wider media, being reported in Science, New Scientist and elsewhere. Indeed, a second paper reporting a separate discovery of this source was placed on astroph within a day of us putting our publication on astroph! RX J0806+15 has an orbital period of 5 mins making it the binary system with the shortest known orbital period. Recently, we have determined that the orbital period of RX J0806+15 is spining up (Hakala et al (2003), Hakala, Ramsay & Byckling (2004)

The current known sample of these ultra-compact binaries is very small, but with three within 400 pc, binary evolution calculations (eg Tutukov and Yungelson 1996) suggest they should be relatively common. In order to establish the birthrates for SN1a -- in some models the systems are driven into coalescence in ~10^5 years -- work is required to evaluate their space density and period distribution.

A number of different scenario's have been proposed to explain the nature of RX J1914+24 (A review of ultra-compact binaries is given in astro-ph/0302240). The most viable at this stage appears to be the Unipolar Inductor (or electric star) model of Wu et al (2002), while the intermediate polar interpretation of Norton et al (2004) remains, in our view, a less likely option. However, there remains considerable uncertainty regarding the true nature of RX J1914+24. The first is its highly unusual X-ray spectra (paper submitted to MNRAS). Grating spectra with a higher signal to noise than these present data is required to investigate this in more detail. In addition, more theoretical work is needed to determine if the Unipolar Inductor model can produce such an X-ray spectrum. Secondly, more theoretical work is needed to determine system parameters, such as the asynchronisity parameter alpha, which can produce the observed change in the spin history of RX J1914+24 and also produce the observed X-ray luminosity. Thirdly, unless the spin and magnetic axes are very closely aligned, the asynchronisity should give rise to a beat period. More work is needed observationally to search for such a period, and theoretically to predict the amplitude of such a beat period. Fourthly, more work is needed to determine the nature of the object giving rise to the K star features in the optical spectrum: phase resolved spectroscopy is urgently required. If RX J1914+24 is a triple system, then this will obviously have implication for the systems evolutionary history.

Work is being undertaken to discover more of these systems and their closely related systems, the AM CVn systems (double degenerate systems with discs). This is being done through observational work and also through optical ground based work. For the latter see the RApid Time Survey (or RATS) home page.