Recurrent novae as a consequence of the accretion of solar material onto a 1. 38 M/sub sun/ white dwarf

We have computed three evolutionary sequences which treat the accretion of hydrogen-rich material onto 1.38 M/sub sun/ white dwarfs. In each of these sequences the accreting matter had only a solar composition of the CNO nuclei (Z = 0.015). In the first sequence we utilized an accretion rate of 1.7 x 10/sup hyphen8/ M/sub sun/ yr/sup hyphen1/ onto a white dwarf with an initial luminosity of 0.1 L/sub sun/ . It took this sequence '33 yr to reach the peak of the thermonuclear runaway which resulted in an outburst that ejected 3x10/sup hyphen8/ M/sub sun/ . of material moving at speeds up to 2900 kn s/sup hyphen1/. The light curve, the time to outburst, and the amount of mass ejected during the evolution are in excellent agreement with the observed outburst of Nova U Sco 1979. However, only 6% of the accreted envelope was ejected during the outburst. The remaining material quickly burned to helium ('2 yr) and settled back onto the white dwarf. The second study involved an accretion rate of 1.7x10/sup hyphen9/ M/sub sun/ yr/sup hyphen1/ onto a white dwarf with an initial luminosity of 10/sup hyphen2/ L/sub sun/ . It took nearly 1600 yr to reach themore » burst phase of the evolution, and by this time the dwarf had accreted '3x10/sup hyphen6/ M/sub sun/ . Peak temperature in the shell source reached 3.5x10/sup 8/ K, about 1.3x10/sup 8/ K higher than was found for model 1. This sequence ejected 3x10/sup hyphen7/ M/sub sun/, only 13% of the accreted envelope, moving at low velocities. For both of these evolutionary sequences, we find that as a result of the accretion of matter onto a massive white dwarf, the mass of the white dwarf grows toward the Chandrasekhar limit. If our study is a realistic representation of the evolution of U Sco, then this star is well on its way to becoming an SN I.« less