Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 322, 785-800 (1997)

Previous Section Next Section Title Page Table of Contents

2. The input optical sample

Up to now, the number of Pop II field binaries with known orbits has been rather limited compared to the wealth of orbital solutions available for Pop I binaries. Orbital parameters have been published so far only for about 90 systems. In Table 1, we have compiled a list of 86 Pop II field binaries of spectral types F to K, which forms our optically selected input sample. The main source for our compilation is the radial-velocity study of 80 spectroscopic binaries by Latham et al. (1988, 1992), which come from the survey of proper-motion stars by Carney and Latham (1987). From the sample of Latham et al. (1988, 1992), we selected 72 metal-poor systems with [FORMULA]. We added to this primary list another 14 systems which have been the subject of detailed investigations by Peterson et al. (1980), Mayor and Turon (1982), Lindgren et al. (1987), Ardeberg and Lindgren (1991), and Lindgren and Ardeberg (1995). Our sample stars are located mainly in the northern sky.


Table 1. Optical Properties of the Pop II Field Binaries.


Table 1. (continued)

In Table 1, the optical properties of the Pop II binaries are listed. The photometry V, [FORMULA] and the orbital parameters [FORMULA], e are taken mainly from Latham et al. (1988, 1992), as indicated by the first entry in the reference column. The orbital periods range from 1.8 to 2000 d. For HD 106516, two different orbital solutions are reported: Latham et al. (1992) found a very long period of [FORMULA], while Abt and Willmarth (1987) suggested a shorter period of [FORMULA]. Our finding, that HD 106516 possesses a quite active X-ray corona (cf. Sect. 4), seems to argue in favour for the shorter period. The metallicities [m/H] are taken, if available, from Spite et al. (1994), otherwise from Laird et al. (1988), as indicated by the second entry in the reference column. The metallicities range from -0.4 to -2.9, showing that the sample contains both intermediate and extreme Pop II stars. The surface gravities g are taken from the high-quality spectra of Spite et al. (1994). The chromospheric indices [FORMULA] are adopted from Pasquini and Lindgren (1994) or were obtained by us at the McMath telescope (Kitt Peak, AZ), as indicated by the fourth entry in the reference column. The [FORMULA] values measure the amount of flux between the [FORMULA] points of the CaII K line, normalised to the continuum flux at [FORMULA]. Thus, [FORMULA] is a measure of the chromospheric activity of the active star within the binary system, normalised to the continuum flux of the primary.

As a coronal activity measure, we will preferentially use the integral X-ray luminosity of the binary system (cf. Sect. 6). To obtain accurate values for the X-ray luminosities, a careful determination of the stellar distances d is essential. The distances quoted in Table 1 are determined as follows:

(i) Since most of our stars are more distant than 50 pc, the most reliable method to derive stellar distances is by means of spectroscopic gravities, which are available for 18 stars in our sample. For the 9 evolved stars with surface gravities [FORMULA], the distances are estimated from theoretical evolutionary tracks calculated by Van den Berg and Bell (1985) for low metallicity stars. We consider the tracks with a He fraction [FORMULA], which is close to the primordial He abundance of presumably [FORMULA] (Boesgaard and Steigman, 1985), and a metal fraction Z according to the star's metallicity. The values of [FORMULA] and [FORMULA] are used to determine the position of the star along the appropriate track. For larger distances and/or low galactic latitudes the effect of reddening is taken into account. For the 9 dwarf stars with [FORMULA], the relation between absolute magnitude [FORMULA] and [FORMULA] derived by Laird et al. (1988) for metal-poor dwarfs is applicable, so we use this relation to estimate distances. Two of the dwarf stars, HD 111980 and BD -00 4234, are not contained in the sample of Laird et al. (1988); their distances are also estimated from evolutionary tracks. The thus derived spectroscopic distances are expected to be rather reliable ([FORMULA] assumed).

(ii) Trigonometric parallaxes, as measured from the ground, have recently been found to be reliable only for distances less than about 50 pc (e.g., Grenon 1995). Therefore, trigonometric distances [FORMULA] are given only if there are no gravity measurements. Thus, trigonometric distances are quoted for 16 stars. About half of these parallaxes have acceptable statistical errors [FORMULA]. The remaining trigonometric distances have statistical errors greater than [FORMULA], and correspondingly a large uncertainty range.

(iii) For the 37 stars in our sample for which both the surface gravity and the trigonometric parallax are unknown, we use again the photometric parallaxes determined by Laird et al. (1988). If the stars are dwarfs, then the photometric distances are reliable ([FORMULA] assumed). However, the fraction of subgiants and giants within the entire proper-motion sample is estimated to be about [FORMULA] (Laird et al., 1988), and is even higher within our binary sub-sample, as is indicated by the [FORMULA] - measurements. As a consequence, at least [FORMULA] of the photometric distances are strongly underestimated. This systematic error cannot be taken into account.

In total, distances are quoted for 65 out of the 86 sample stars. The reference for the distances is indicated by the third entry in the reference column. For HD 149414 and HD 89499, the spectroscopic distances are far above the trigonometric distance ranges (cf. column ' [FORMULA] ' of Table 1), indicating that the trigonometric distances of these stars are probably wrong. In the case of HD 89499, the spectroscopic distance is one order of magnitude larger than its trigonometric value. The large value of the hydrogen column density of [FORMULA], derived from ROSAT and ASCA X-ray spectra (Fleming and Tagliaferri 1996), would seem to suggest the larger distance.

Because our sample is optically selected, it is not a priori biased towards X-ray luminous Pop II binaries. However, our sample is not volume-limited and may therefore contain a slightly higher fraction of subgiants and giants than a complete (i.e., volume-limited) sample.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998