The Enigma of AADor

AADor (LB 3459) is an eclipsing, close binary (P = 0.26 d) consisting of a sdOB primary star and an unseen secondary with an extraordinary small mass. The secondary is possibly a former planet which may have survived a common-envelope phase and has even gained mass. In order to investigate on an existing discrepancy between the components’ mass derived from NLTE spectral analysis and subsequent comparison to evolutionary tracks and masses derived from radial-velocity and the eclipse curves, we performed phase-resolved high-resolution and high-SN spectroscopy with the UVES attached to the ESO VLT. From the obtained spectra, we have determined AADor’s orbital parameters (P = 22 600.702±0.005sec, A = 39.19±0.05km/sec, T0 = 2451917.152690) and the rotational velocity (vrot = 47 ± 5km/sec) of the primary. Hilditch et al. (1996) have analyzed that the primary component of AADor has a mass of M1 = 0.5M⊙ and the cool secondary has a very small mass of M2 = 0.086M⊙. The latter is in excellent agreement with lowest mass ZAMS models of Dorman et al. (1989). In a recent spectral analysis (Rauch 2000) of the primary, based on high-resolution CASPEC and IUE spectra, M1 = 0.330M⊙ has been derived from comparison of Teff and log g to evolutionary tracks of post-RGB stars (Driebe et al. 1998) and M2 = 0.066M⊙ has subsequently been calculated from the system’s mass function. Thus, the secondary is possibly a planet which may have survived a common envelope phase and has even gained mass (“late case B mass transfer”, Iben & Livio 1993). However, a reason for the discrepancy between these masses and those derived from the radial-velocity and the eclipse curves (Kilkenny et al. 1979) is not known. Two possible reasons are the inaccuracy of the theoretical models by Driebe et al. (1998) for AADor and the error range for the photospheric parameter determination in Rauch (2000). This analysis was hampered by the long exposure times (some hours) and hence, a large orbital velocity coverage. In order to make progress and to minimize the effects of orbital motion, 105 UVES (attached to the ESO VLT) spectra had been taken on Jan 8, 2001 with an exposure time of 180 sec each. We achieved a resolving power of 48 000 at an average S/N ≈ 20. The spectra cover a complete orbital period of AADor.