ON THE ORIGIN OF HYDROGEN-DEFICIENT SUPERGIANTS AND THEIR RELATION TOR-CORONAE-BOREALIS STARS AND NON-DA WHITE-DWARFS

Scenarios for the formation and evolution of hydrogen-deficient superg iants are analyzed. Some of these supergiants may possibly be identifi ed as R CrB stars. The main scenarios involve (1) a final, post-hydrog en-burning, helium shell flash in the central star of a planetary nebu la, (2) the merger of hydrogen-deficient components of evolved close b inaries, and (3) the merger of a neutron star with a helium-rich star. R CrB stars produced in all scenarios are progeny of relatively low m ass stars (0.8-6.0 M.) and many are among the oldest stars in our Gala xy, a result which helps explain their relatively large average distan ces from the Galactic plane. In the single star scenario, the nucleosy nthesis and mixing required to explain observed abundances has been de monstrated explicitly by model calculations, whereas the corresponding processes in binary star scenarios must be postulated until appropria te two-dimensional or three-dimensional hydrodynamic merger models hav e been constructed. Birthrates, observed numbers, and other main param eters of R CrB stars and of other hydrogen-deficient stars which arise in close binaries after a merger of evolved components are estimated with a numerical scenario model which provides distributions over mass es, ages, and distances from the Galactic plane. A comparison of scena rio model predictions with properties of observed R CrB stars suggests that there are lower limits on the initial stellar mass and on the ma ss of the degenerate (CO or ONe) core of a helium star if it is to bec ome an R CrB star. Estimates of both limits require further calculatio ns of helium giant star models with appropriate envelope compositions and opacities adequate for these compositions. It is also clear that a stellar wind plays a decisive role in determining the number of R CrB stars produced by binary and single stars. The birthrate of R CrB sta rs produced by single stars and by components of wide noninteracting b inary stars is estimated to be between 0.01 yr(-1) and 0.1 yr(-1), and the birthrate of those produced by mergers of components of close bin ary stars is estimated to be in the range similar to 0.004-0.007 yr(-1 ). The helium fuel initially available for binary-produced R CrB stars is typically 20 times larger than is available in R CrB stars produce d by single star evolution, suggesting that most observed R CrB stars originate in close binaries. The fact that only similar to 10% of obse rved R CrB stars are associated with an extended nebula which is visib le at optical, IR, or both wavelengths supports this inference. On the other hand, since relevant model calculations which show that merger products generate appropriate compositions do not exist, the single st ar scenario cannot be catagorically excluded as the major producer of R CrB stars. Mergers may be progenitors of hydrogen-deficient stars of other types. Roughly 10% of all single stars and stars in wide binari es undergo a final, post-hydrogen-burning helium flash and evolve into non-DA white dwarfs, whether or not they pass through an R CrB phase. Approximately 15% of all single stars and stars in wide binaries unde rgo a final helium flash before hydrogen burning ceases during the pla netary nebula stage. These stars probably evolve into shell helium-bur ning giants and, if wind mass loss is efficient enough, they will ulti mately evolve into non-DA white dwarfs. Roughly 20% of all close binar ies produce single hydrogen-deficient stars which evolve into non-DA w hite dwarfs, whether or not they pass through an R CrB phase. White dw arfs of mass equal to or greater than 0.8 M. which descend from single stars and stars in wide binaries lose their hydrogen-rich envelope du ring the first 10(8) yr of the white dwarf cooling phase and also beco me non-DA dwarfs. Thus the theory of stellar evolution predicts that p erhaps up to similar to 45% of all white dwarfs are of the non-DA vari ety. Those helium supergiants which are formed by way of mergers and h ave sufficiently massive (ONe) degenerate cores may terminate their ev olution as supernovae. The frequency of such events in our Galaxy, as given by the scenario code, is similar to 6 x 10(-4) yr(-1), suggestin g that perhaps one out of every 10-20 helium supergiants made by close binaries may complete its evolution as a hydrogen-free supernova.