A low-amplitude X-ray and optical outburst from the periodic transient A0538-66: Accretion onto a magnetosphere?

We present ASCA observations made in 1995 February around the time of a predicted outburst from the 16.65 day period recurrent transient A0538-66, together with simultaneous optical photometry. Optical spectroscopy was also obtained near the time of a predicted outburst approximately 2 months earlier. We find that A0538-66 is active, both in X-rays and optically, although at a much lower level than during the largest super-Eddington outbursts that have been seen in the past. The optical spectra show variable Balmer emission, but not the strong He II 4686 Angstrom emission that accompanies large outbursts. The average X-ray luminosity is similar to 5.5 x 10(36) ergs s(-1), and we find evidence for the presence of an iron line in the X-ray spectrum. This can be interpreted either as a single line at 6.54 +/- 0.04 keV or, alternatively, as a blend of emission at both 6.4 and 6.7 keV. A simple power law, or power law plus high-energy cutoff, does not fit the continuum spectrum well. Instead, a more complicated model is required, and we find that a power law plus blackbody gives a good fit to the continuum, although other models cannot be excluded completely. No evidence for the presence of previously detected 69 ms pulsations is found in our data, although our sensitivity is limited by the low count rate during our observations and the unknown orbital parameters of this system. We consider X-ray production processes that may account for the observed behavior. In particular, accretion onto the magnetosphere rather than the surface of a neutron star is discussed. Our data do not, however, exclude conclusively the possibility that accretion onto the neutron star is still occurring either because the magnetic field of the neutron star is low or because material is leaking past the magnetic barrier. If the X-ray flux is powered by energy released at the magnetosphere, then the presence of an iron line in the spectrum at greater than 6.4 keV, which is rare for Be star X-ray binaries, may be an indicator that such a process is occurring.