AbstractAn evolutionary scenario for classical novae is proposed, which is intended to solve the discrepancies that exist between theory and observations: the space densities of classical novae deduced from surveys in the solar neighbourhood are lower by about two orders of magnitude than those predicted theoretically, and the mass transfer rates in nova binaries, as estimated from observed luminosities in quiescence, are higher than those allowed by the thermonuclear runaway model for nova outbursts. These discrepancies disappear if mass transfer (at a high rate) takes place for only a few hundred years before and a few hundred years after an eruption, but declines afterwards and remains off for most of the time between outbursts. We show that such a behavior is to be expected if one takes into account the variation of binary separation, due to mass ejection on the one hand and angular momentum losses on the other hand.One of the aspects of this scenario, on which we report in more detail, is the possibility of enhanced Roche-lobe overflow of the secondary, due to its expansion that results from irradiation by the high nova luminosity. We followed the evolution of a 0.5M⊙ main sequence star illuminated by a changing flux, typical of a classical nova. The numerical results indicate that, in spite of the slight binary separation that may occur after eruption, mass loss from the irradiated and thus bloated secondary should continue for a few hundred years. Other aspects of the variable Ṁ scenario are briefly summarized.
The Thermonuclear Runaway and the Classical Nova Outburst
