SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 354, 1014-1020 (2000)

Previous Section Next Section Title Page Table of Contents

4. Limits for a white dwarf wind

For winds from the hot component in symbiotic binaries, qualitatively different results have been found for different systems. In AG Peg a strong P Cygni profile in N V  [FORMULA] reveals the existence of a white dwarf wind with a terminal velocity of [FORMULA] (Nussbaumer et al. 1995). The spectrum of AG Peg has been analyzed by Schmutz (1996), who derived a mass-loss rate of [FORMULA]. For EG And Vogel (1993) found [FORMULA] and [FORMULA]. For RR Tel no trace of a wind from the hot component can be found in HST GHRS spectra (Nussbaumer & Dumm 1997). For the eclipsing symbiotic nova PU Vul, a white dwarf wind with [FORMULA] has been detected in the N V  [FORMULA] emission line (Nussbaumer & Vogel 1996). In Fig. 3 we compare the N V  [FORMULA] line profile of RW Hya taken at [FORMULA], with that of AG Peg. In AG Peg the P Cygni profile corresponds to [FORMULA]. In the spectrum of RW Hya we see a P Cygni absorption feature corresponding to [FORMULA]km/s, but there is no indication for a P Cygni absorption from a faster wind.

[FIGURE] Fig. 3. The P Cygni profile in the N V [FORMULA] doublet of AG Peg (dotted) indicates a fast wind with a `zero' intensity expansion velocity (Howarth & Prinja 1989) [FORMULA], and an absorption feature corresponding to [FORMULA]. The RW Hya spectrum (solid) shows only the absorption feature at [FORMULA]. Both spectra have been corrected for the system velocity, the flux is in units of erg/(cm2 s Å).

With the wind momentum-luminosity relation (WLR) for radiation driven winds (Kudritzki 1998) we can calculate a mass-loss rate [FORMULA] for the white dwarf. The luminosity of the white dwarf in RW Hya has been determined by Kenyon & Mikolajewska (1995) and Schild et al. (1996). Inserting the upper limit given by Schild et al. (1996), [FORMULA], into the WLR leads to a mass-loss rate [FORMULA]. The lower value of Kenyon & Mikolajewska (1995), [FORMULA], corresponding to a distance of [FORMULA] derived by Schild et al. (1996) yields [FORMULA]. Both groups find a similar white dwarf radius, [FORMULA]. Schild et al. (1996) derived a white dwarf mass of [FORMULA]. This leads to an escape velocity of [FORMULA]. Typically, the terminal velocities of the winds of central stars of planetary nebulae (CSPN) are about 2.5 times the escape velocity, indicating for RW Hya [FORMULA].

One has to keep in mind, that the observed properties of winds from the hot components in AG Peg, EG And, and PU Vul strongly differ from those predicted by the WLR. This might be due to the large extrapolation, as the least luminous objects used to derive the WLR have luminosisties [FORMULA] and temperatures [FORMULA].

In order to test whether a possible white dwarf wind is significant compared to the red giant wind, we have to compare the momentum fluxes of the two winds. Along the binary axis they are balanced at a distance r from the white dwarf

[EQUATION]

where s is the binary separation.

A hot wind velocity of [FORMULA], together with [FORMULA], [FORMULA] and [FORMULA], leads to [FORMULA] corresponding to [FORMULA]. We will discuss the implications of a wind from the white dwarf in Sect. 6.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: February 25, 2000
helpdesk.link@springer.de