1995A&A...299..724C

We have studied a sample of selected eclipsing binaries with eccentric orbits and accurate absolute dimensions in order to analyse the validity of available theories of tidal evolution concerning synchronization and circularization time scales. In the present investigation we pay special attention to stars with radiative envelopes though some stars with outer layers in convective equilibrium have been included to show different aspects of tidal evolution. In the present paper, we have adopted for the comparison with observational parameters, the tidal theory by Tasssoul (1987ApJ...322..856T , 1988ApJ...324L..71T). The formalism introduced by Zahn (1977A&A....57..383Z, 1989A&A...220..112Z) will be the subject of a separate paper. The critical values of the age, log g and R for circularization and synchronization have been computed using a set of stellar models which include recent opacity libraries, core overshooting and mass loss. These values can be computed as a function of the orbital period for different masses. In this way, it is relatively simple to compare them with observational values and, therefore, to analyse the behaviour of parameters which indicate the actual situation, namely, the eccentricity and the rotational velocities. The comparison of observed rotational velocities with expected values for each binary system configuration has shown that nearly all systems in our sample are actually pseudo-synchronized with their orbital periods (i.e. with the fastest orbital velocity in eccentric systems) in good agreement with theoretical predictions. These results, however, should be taken with some caution since the rotation of the stellar interior may be decoupled from that of the surface layers. With respect to circularization, a cutoff value around log t=log t_cri(or log g_obs=log g_cri) shows a satisfactory agreement for observational values of the eccentricity. Furthermore, we concluded that the circularization process is still active during the main sequence stage for early-type binaries.