Discussion

The indication of a non-uniform, bimodal tendency in the phase distribution of detectable emissions from certain binary classes raises interest in the physics of their configurations. RS CVn stars are a well-known source of microwave radiation, and there are many corroborative factors which link this radiation with enhanced magnetodynamic stellar `activity' (cf. e.g. Linsky, 1988, 1994). The 3 detection distribution and the background radiation trend for V711 Tau suggest some preferred longitudes for the visibility of magnetodynamic phenomena. This in keeping with `active longitude' sector concepts which have been frequently discussed. With surface fields of order several thousand gauss concentrated over regions of order 10 m in radius, magnetic reconnection events can be expected to energize electons at least to tens of MeV. Such electrons would subsequently lose energy by synchrotron radiation in the confining fields of the active region, and may be the source of the detected outbursts (Dulk, 1985). It would thus be a useful exercise to carry out a more detailed inter-comparison of radio emission and photometric properties of both RS CVn and Algol systems (Budding, 1996).

From another point of view, the long-term, quasi-continuous and relatively substantial flow of matter from a `loser' to a `gainer' characterizes Algols, or other stars undergoing Roche Lobe overflow (RLOF). The phase range 0.4-0.5, where our Algols' detections maximize, corresponds to where the mass transferring plasma stream would be seen coming towards the observer over the limb of the primary star. The possibility of a signature of the RLOF stream in the microwave range has only been briefly considered previously (e.g. in Slee et al., 1987), as far as we know. It is suggested, however, by the discussion of the location of origin for the microwave emission from polars like V834 Cen, which also indicate a phase-linked pattern to their emission (cf. Chanmugam & Dulk, 1982; Wright et al., 1988, Ferrario et al., 1989). So, insofar as certain gross features of the mass-transferring stream's microwave emission in polars may resemble that of Algols, we might expect some comparability of observed effects.

It could also be interesting to examine this idea further. In considering any potentially relevant properties of the mass transferring stream specific to Algols, reference can be made to Lubow & Shu's (1975) classical Algol model. This allows for kinetic energy transport in the transferred material typically in the order of 10 of the gainer luminosity (cf. e.g. Plavec, 1989), or up to, say, watts, for the classical Algols observed. If we estimate field strengths near the gainer-stream interaction region to be of order 10 gauss, where the hypersonic stream velocities attain , the Faraday induction across the stream could easily be of order 0.1 volts per m, developing e.m.f.s of around volts across an interaction (`hot spot') region of lateral extent m. The detections of the Algols at the mJy level require that up to watts be emitted at source over the whole microwave spectrum during the outbursts. The sporadically occurring emissions are most frequently observed when the induced electic field would be predominantly perpendicular to the line of sight, if the relevant region of the gainer's magnetic field is essentially poloidal and nearly perpendicular to the by-passing stream. Since outbursts are observed only of the time we require of the available steady kinetic power. This kinetic power is of the same order as the Poynting energy developed in the hypothesized interaction region with the foregoing parameter values.

At present, it has to be admitted that the Parkes evidence for any special effects in Algol systems is suggestive only. A significantly higher median brightness temperature of Algols over RS CVn stars was noted before (Slee et al., 1987), but we have not been able effectively to distinguish any difference in phase-dependent characterstics by comparing our Algol and RS CVn binary microwave detection statistics. Although it would be interesting to know if there was any signature of the characteristic stream of Algol binaries at radio-wavelengths, a more sensitive survey than the Parkes one would be required to find out. In the meantime it appears more pragmatic to regard the processes giving rise to Algol and RS CVn binary radio emissions as essentially similar.

Acknowledgements

E. Budding acknowledges support from the New Zealand side of the Bilateral Research Agreement Programme (BRAP) of the Ministry of Research, Science & Technology (NZ), 1996, to enable visits to the Australia Telescope National Facility, where the ideas for this paper were mooted. We also acknowledge the initial unpublished analysis of Dr K. Thompson of Monash University on the subject of orbital dependency in the Parkes stellar observations.

© Copyright Astronomical Society of Australia 1997