Because of their proximity, the Magellanic Clouds have been the main target of these studies in the past and many research papers or reviews have been devoted to them. But the situation will improve dramatically with the study of double-lined eclipsing binaries within them. In fact, it is well known that binary stars for which accurate absolute dimensions can be derived are the best source of fundamental information for studies on stellar structure. Double-lined eclipsing systems are the best way to obtain stellar masses and radii without any need for astrophysical calibrations.
It is also well-known that currently available theoretical models reproduce quite satisfactorily the stellar component of our galaxy providing additional information on chemical compositions and ages. But there are still some uncertainties in the models concerning the treatment of convection, rotation, and distribution of the chemical composition. The possibility to test the models in different conditions is expected to be very fruitful in this concern if accurate enough data are used.
In the past, the monitoring of eclipsing binaries in the Magellanic Clouds was not an easy task. Gaposhkin (1970, 1977) provided lists of candidates with periods and low quality light curves derived from photographic plate surveys aimed to the detection of new pulsating stars and the determination of the distance scale. Though this sample was used by Dworak (1974) to estimate also the distance to the Clouds using the photometric elements and an average mass-luminosity relation, the results were, as could be expected, not accurate enough (de Vaucouleurs 1978). The obtention of better light curves was attempted by several authors (Herczeg 1982) using photoelectric photometry, but it was not until the work by Jensen et al. (1988) when it was clearly shown that CCD light curves could be obtained with an accuracy (better than 0.01 mag) comparable to that of the binaries in our galaxy.
This work triggered many other studies, both in the photometric and the spectroscopic sides. Observations in La Silla, using the 1.52 m Danish telescope, and following the work by Jensen et al. (1988), has resulted in a wealth of accurate uvby light curves within both Clouds, their publication only waiting for accurate spectroscopic radial velocity curves. Although some data are available from observations with the 3.5 m NTT at La Silla, good radial velocities are still the main obstacle for the advance of research in this domain. Binaries are too faint for medium size telescopes and large apertures are needed. Moreover, since the source of accurate dimensions is only provided by well-detached binaries, orbital periods are necessarily long to keep large massive stars separated enough.
Spectroscopic observations have nevertheless been published by Niemela & Bassino (1994) while the first results on absolute dimensions were published by Bell et al. (1991) using HV2226 in the SMC. Unfortunately this system was found to be semi-detached, thus leaving little room for model testing. A similar situation was found in the case of HV5936 (Bell et al. 1993). Spectroscopic observations were obtained using the 3.9 m Anglo-Australian Telescope in Australia.
Photometric studies were also started from Mt. John Observatory in New Zealand using a CCD camera with a modest telescope (Tobin 1994). They discovered the case of the well detached binary HV2274 in the LMC (Watson et al. 1992). This is a very important target due to its eccentric orbit and evidences of apsidal motion which allowed Claret (1996) to establish limits on the masses of the component stars using internal structure constants derived from theoretical models. Radial velocities are now available and accurate absolute dimensions and distance determinations will be published soon (Guinan et al. 1997). Other binaries for which light curves have been obtained include HV12634 (West et al. 1992), HV12484 (Tobin et al. 1993), HV 1761 (Duncan et al. 1993) and the eccentric system HV982 (Pritchard et al. 1994).
Observational difficulties in establishing accurate values of stellar temperatures and interstellar absorption for O-B stars from standard multicolour photometry are moreover being solved thanks to the use of IUE and HST observations in the ultraviolet domain (Guinan et al. 1996). HST observations are also being used by Guinan et al. (1997) to obtain radial velocities in the ultraviolet where numerous spectral features are present.
Although some problems of theoretical and observational nature still
remain, tidal evolution of close binary stars can be considered as an
additional test to the stellar models. Besides the classical tests of
isolated binaries, we can study critical times for binaries which are
found in clusters. From systematic observations of radial velocity curves
of binaries in clusters, some authors (e.g. Mathieu et al. 1992) found a
cut-off period (
) below which the binaries present circular
orbits. These critical periods seem to depend on the age of the cluster.
Integrating the differential equations for tidal evolution we are able to
compare the observed versus age of the clusters with
theoretical predictions (see Claret & Cunha 1997, Fig. 5). The time
required for eccentricity (or the level of assynchronysm) to decay to 0.05%
of its initial value is the theoretical critical time. The ages of the
clusters can be considered only as upper limits for the critical time. The
investigation of binaries in clusters in the Magellanic Clouds may bring
more light on the subject and the necessary tools for such studies in terms
of model computations are given in this paper.
On the other hand, the search for gravitational microlensing events is currently providing in a serendipitous way many additional candidate eclipsing binaries in the LMC. The results of both the EROS and the MACHO programs will no doubt increase the sample of good well-detached systems for the study of stellar structure outside of our galaxy. In addition to the mentioned studies, well detached binaries in other galaxies can of course be used as accurate extragalactic distance indicators (Giménez et al. 1994; Guinan et al. 1996).
In this paper we compute the necessary models to compare the accurate
data being obtained from detached binaries in the SMC with theoretical
predictions. The models thus include values not only for the usual
fundamental data (radius, 
, 
, 
) as a
function of mass, age and initial chemical composition, but also the
internal structure constants, 
, and other parameters needed for
the calculation of predicted levels of orbital circularization and
synchronization. In Sect. 2 we present such a grid of models, their main
characteristics and the effects of core overshooting and of the initial
hydrogen content.
Finally, the need for a grid of models with Z=0.004 is not only given by the study of stars in the SMC but also to allow the interpolation of metallicities in the lower end of the already published series of grids between Z=0.01 (Claret & Giménez 1995) and Z = 0.03 (Claret 1997) which was found to be necessary for the LMC and low metallicity binaries in our galaxy.
Copyright The European Southern Observatory (ESO)