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


Astron. Astrophys. 317, 701-706 (1997)

Previous Section Next Section Title Page Table of Contents

3. Results and discussion

We present the measured rotational velocities Vsini in Table 1. These rotation rates are plotted in Fig. 1 as a function of the color index (B-V). In spite of the limited sample studied here, this figure shows clearly the sudden decline in rotation near (B-V) [FORMULA] 0.55, which corresponds to the spectral type F8IV. Such rotational discontinuity is particularly very well defined for the single subgiants and for those binary subgiants with orbital period greater than about 20 days, typically binary systems with orbital eccentricity greater than about 0.10. In fact, except for HD 71071, all the binary systems with orbital period lower than 20 days and circular orbit located to the right of the rotational discontinuity present enhanced rotational velocities which are very likely to result from the synchronization between the axial rotation and orbital revolution. The orbital parameters for the double-lined system HD 31738 are not yet determined, but from 19 radial velocity measurements obtained with the CORAVEL spectrometer we found a velocity range of about 50 km.s-1 indicating a short orbital period for such star. To the left of the rotational discontinuity, one sees a wide range of Vsini values, which reflects the broad distribution of rotation rates for the progenitors of the subgiant stars. Unfortunately, the number of stars to the right of the discontinuity presented in Fig. 1 is very limited, but as shown by De Medeiros (1990) in this side of the discontinuity the rotation for the single subgiants decreases smoothly from the G to the K spectral region, with a mean rotational velocity of about 6.0 km.s-1 near the spectral type G0IV and a mean rotation velocity of about 1.0 km.s-1 near the spectral type K0IV. Fig. 2 presents Lithium abundance as a function of the color index (B-V) for all the stars listed in Table 1. This figure shows the familiar decrease of Lithium abundance with increasing (B-V), but rather than a gradual decrease it seems to exist a drop in the distribution of the Lithium abundance as a function of colors near (B-V) = 0.55, corresponding to the spectral type F8IV or to the effective temperature of 6000K. A careful comparison between Figs. 1 and 2 indicates that the location of the sudden decline in rotation almost coincides nearly with the drop in Lithium abundance. Based on this result, one might ask whether these two properties, the rotation discontinuity and the drop in Lithium abundance, have the same root cause. A strong argument may be in favor of this hypothesis if a correlation could be found between rotational velocity and Lithium abundance.

[FIGURE] Fig. 1. Rotational velocity as a function of (B-V) color index for all of the programme stars. Single stars are identified by squares and binaries by circles, which are filled for the systems with circularized orbit
[FIGURE] Fig. 2. Lithium abundance as a function of (B-V) color index for the programme stars. Squares and triangles represent respectively measured and upper limit Li abundances for single stars; circles and diamonds represent respectively measured and upper limit Li abundances for binaries, which are filled for systems with circularized orbit

In Fig. 3, we plot Lithium abundance versus rotational velocity for the stars listed in Table 1, which are separated in three intervals of color (B-V) [FORMULA] 0.55, 0.55 [FORMULA] (B-V) [FORMULA] 0.75 and (B-V) [FORMULA] 0.75. We have carried out least-squares regression analysis for the stars presented in Fig. 3. A log-linear least-squares fit for Logn(Li) versus LogVsini was derived first for the sample of single stars and then for the binary stars. In our calculations for the single stars we have included only those presenting at least two CORAVEL radial velocity measurements, that is, we have ignored HD 184663 which has just one radial velocity measurement, not enough to define its single or binary status. From the sample of single stars given in Table 1, except for HD 184663, we have found a linear correlation coefficient of about 0.70 and a standard deviation of 0.52. Thus, the Lithium abundance is almost linearly proportional to the rotation rate. For the entire sample of binary stars our least-squares solution yields a linear correlation coefficient of 0.01 and standard deviation of 0.74. Our calculations give also very poor linear correlation coefficient and standard deviation if we take into account just the binary stars with orbital period lower than about 20 days and eccentricity lower than about 0.10. We regard the slopes of the relations resulting from these calculations, except for single stars, as indistinguishable from zero and conclude that, at least for the present sample, Lithium abundance is independent of rotation rate in binary subgiant stars. Similar least-squares fits were obtained for each color interval given in Fig. 3, and an equally poor correlation was found for the color intervals (B-V) [FORMULA] 0.55, 0.55 [FORMULA] (B-V) [FORMULA] 0.75 and (B-V) [FORMULA] 0.75. Additional evidence of no correlation between Lithium abundance and projected rotational velocity Vsini in binary subgiants was provided by Randich et al. (1994), particularly for active subgiant stars.

[FIGURE] Fig. 3. Lithium abundance versus rotational velocity for all of the programme stars. Triangles denote stars with (B-V) [FORMULA] 0.55, circles denote stars with 0.55 [FORMULA] (B-V) [FORMULA] 0.75 and squares those with (B-V) [FORMULA] 0.75. Single stars are identified by open symbols and binaries by filled symbols

Concerning the good coincidence between the location of the rotational discontinuity and the drop in Lithium abundance near the spectral type F8IV, it is important to emphasize here that a magnetic braking seems to play a relevant role as the root cause of the rotational discontinuity in subgiant stars (Gray & Nagar 1985; Rutten & Pylyser 1988; De Medeiros & Mayor 1989, 1990). Could we conclude that Lithium depletion in single subgiant stars is also affected by magnetic braking?

Another interesting trend emerges from Fig. 2. Despite the limited number of stars with (B-V) [FORMULA] 0.55, it seems that synchronized binary subgiants located to the right of the rotational discontinuity have a tendency to retain more of their original Lithium than both their single counterparts, and the non synchronized binary subgiants. Zahn (1994) has shown that late-type binaries of short enough orbital period, typically orbital periods below 8 days for solar-type stars of population I, and 6 days for halo stars, retain more of their original Lithium than their single counterparts. Spite et al. (1994) have found the same trend in a sample of old disk and halo stars. The results of the present work indicate that the same behavior could be present in K-type binary subgiant stars.

3.1. The rotational discontinuity and the drop in Lithium abundance

Because the magnetic braking, associated to the origin of the rotational discontinuity, is expected to operate when the convection zone becomes deep, while Lithium dilution also depends on the growth of the convection zone, a relationship between the two would, in principle, be expected. Nevertheless, one should first inquire if the drop in the abundance of Lithium shown in Fig. 2 is real. Firstly, we must be cautious with the size of the sample and the nature of the Lithium abundances analyzed here. In this sense, Lebre et al. (1995) have claimed a drop in the distribution of the equivalent width of the Lithium line 6707.81 Å near the spectral type F8IV, on the basis of a larger sample of subgiant stars than the one used in the present work. Single subgiant stars located to the right of this spectral type show no important Lithium line feature. Concerning the sources of the Lithium abundances presented in Table 1, as we have pointed out in Sect. 2, the detailed error analysis effectuated by Balachandran (1990), Pallavicini et al. (1987) and Randich et al. (1993, 1994) shows that their data have the same high quality, particularly with a same interval of values for the signal-to-noise ratio. The Lithium abundances from these authors are the most relevant in the definition of the drop in Lithium shown in Fig. 2. These two points appear to indicate that the drop in Lithium is not a result of selection effects.

An important aspect of the sample studied here concerns its evolutionary stage. Balachandran (1990) has shown that F stars, slightly evolved off the main sequence, present a large scatter in Lithium abundances, indicating that the Lithium depletion is not related to age, Vsini or spectral types alone. This author has identified two groups of F slightly evolved stars, a first one with Lithium abundances between 2.0 and 3.5, and a second one presenting upper limits with abundances lower than 2.0 presumably presenting Lithium depletion. Following the analysis of Balachandran (1990), the first group is composed by stars with masses between 1.1 [FORMULA] and 1.85 [FORMULA], whereas the second group appears to be concentrated between 1.1 [FORMULA] and 1.5 [FORMULA]. From the 35 stars of Balachandran (1990) listed in Table 1 of the present paper, 22 stars belong to the first group defined above and 13 stars, those with upper limits, belong to the second group. So, one can conclude that in the spectral region of the drop in Lithium abundance stars have probably masses between 1.1 [FORMULA] and 1.85 [FORMULA]. If we consider that the subgiant branch represents an evolutionary sequence, then we expect the stars on the right side of the drop in Lithium to have evolved from the same population as the stars on the left side of the drop. Unfortunately, the number of stars on the right side of the drop in Lithium is very limited and, consequently, a study to determine the kinematic-age relation between stars on both side, of the drop is not possible here. Nevertheless, a kinematic-age analysis by De Medeiros (1990) for subgiants F, G and K of population I, based on the dispersion of the radial velocity, shows no significant increase of the velocity dispersion from the left side through the right side of the rotational discontinuity. This fact indicates that cool G and K subgiants arose from the same population in mass as the hotter F subgiants.

It is interesting to point out that a study of the Lithium behavior in halo subgiants by Pilachowski et al. (1993) shows no drop in the distribution of the Lithium abundances with the effective temperature or spectral type. These authors have shown that the spread in Lithium abundance for such stars near Teff = 6500 K, which corresponds to the spectral type F8IV, is small and that the mean value is near logn(Li) = 2.1. Note that unlike the population I subgiants studied in the present paper, which seems to have masses between 1.1 [FORMULA] and 1.85 [FORMULA], the halo subgiants evolve from a restricted range of mass on the main sequence. The masses of the halo subgiants are substantially lower than those of the population I subgiant stars. Further, it is important to underline that differently of the population I subgiants there is no sign of rotational discontinuity for halo subgiant stars. This difference may result from the fact that the precursors of the population I subgiants present a spread in rotational velocity from a few km.s-1 to about 150 km.s-1, whereas the halo subgiants seem to have had slowly rotating precursors. According to Pinsonneaut et al. (1989), the strength of the Lithium depletion in late F-type stars will depend upon the magnitude of the change in the rotational velocity. On the basis of this scenario, the sudden decrease in the rotation of the subgiants presented in Fig. 1 should correspond to a similar decrease in the abundance of Lithium for this class of stars. This fact indicates that the rotational discontinuity and the drop in the abundance of Lithium for population I subgiants, both observed near the spectral type F8IV, seem to be controlled by the same root cause.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: July 8, 1998
helpdesk.link@springer.de