The evolution of binary stars into contact states

Good-quality empirical results on 62 short-period binary stars recently summarised by Hilditch & Bell (1987) and Hilditch, King & McFarlane (1988) are discussed in terms of evolutionary paths from detached to semi-detached and contact states. These data suggest two evolutionary paths to the contact binaries — from detached systems directly into contact to form initially shallow-contact systems, and via case A mass transfer to semi-detached states, thence to contact systems. These empirical results support previous arguments based on evolutionary models and less detailed observational data. Concern is expressed about the paucity of high-quality spectroscopic data, particularly for low-mass systems displaying EB-type light curves and the resultant limitations on analyses of those light curves. Such systems provide tests of evolution into contact for the first time, or of broken-contact phases for WUMa-type binaries. The crucial importance of long-term monitoring (decades) of times of minima as indicators of mass transfer rates amongst these interacting binaries is also noted.     

The relation of Algols and W Serpentis stars

Evidence on the issues of whether the W Serpentis stars are a coherent class, and how they may interface with the Algol systems, is reviewed, with emphasis on the idea that they are semi-detached systems in the latter part of the rapid phase of mass transfer, with optically and geometrically thick disks of transferred gas around the (now) more massive star. We are interested in what will be seen when the gas clears away, and mainly examine the idea that it will be an Algol-type system. More particularly, consideration is given to centrifugally limited accretion as a mechanism to build up a substantial disk, and the presumed evolutionary sequence is from a W Ser to a rapidly rotating Algol to a normal Algol system. Systems such as V367 Cyg and RW Tau fit into this scheme only with difficulty. Because it is extremely difficult to measure the rotation of some W Ser (mass) primaries, it is natural to look at the rotation statistics of Algols to test this idea. The badly behaved light curves and spectroscopy of some Algols (eg. U Cep, RZ Sct) may be attributable to the double contact condition, and the ramifications of this possibility are discussed. If so, the rotation statistics of Algols should show two spikes, corresponding to the two special conditions into which a system should be driven by tidal braking and centrifugally limited spinup. Present rotation statistics do show these spikes. Algols should flip between these states fairly quickly, depending on the mass transfer rate. Thus, to the extent that the meager statistics can be accepted as meaningful, the new (fourth) morphological type of close binary (double contact) has attained demonstrable reality. The rotation statistics are presented in terms of a particular rotation parameter, R, which is zero for synchronism and unity for the centrifugal limit. Future work should develop rotation statistics to see if the rotational lobe-filling (R = 1) spike persists. It should also look into whether W Ser primaries are on the hydrogen burning main sequence, or in general what they are. We also need more light curves of W Ser type systems, high resolution line profiles for the (mass) primaries (with particular attention to the W Ser-Algol transition cases), and spectroscopy of low inclination W Serpentis systems, such as KX And.     

The relation of Algols and W Serpentis stars

Evidence on the issues of whether the W Serpentis stars are a coherent class, and how they may interface with the Algol systems, is reviewed, with emphasis on the idea that they are semi-detached systems in the latter part of the rapid phase of mass transfer, with optically and geometrically thick disks of transferred gas around the (now) more massive star. We are interested in what will be seen when the gas clears away, and mainly examine the idea that it will be an Algol-type system. More particularly, consideration is given to centrifugally limited accretion as a mechanism to build up a substantial disk, and the presumed evolutionary sequence is from a W Ser to a rapidly rotating Algol to a normal Algol system. Systems such as V367 Cyg and RW Tau fit into this scheme only with difficulty. Because it is extremely difficult to measure the rotation of some W Ser (mass) primaries, it is natural to look at the rotation statistics of Algols to test this idea. The badly behaved light curves and spectroscopy of some Algols (eg. U Cep, RZ Sct) may be attributable to the double contact condition, and the ramifications of this possibility are discussed. If so, the rotation statistics of Algols should show two spikes, corresponding to the two special conditions into which a system should be driven by tidal braking and centrifugally limited spinup. Present rotation statistics do show these spikes. Algols should flip between these states fairly quickly, depending on the mass transfer rate. Thus, to the extent that the meager statistics can be accepted as meaningful, the new (fourth) morphological type of close binary (double contact) has attained demonstrable reality. The rotation statistics are presented in terms of a particular rotation parameter, R, which is zero for synchronism and unity for the centrifugal limit. Future work should develop rotation statistics to see if the rotational lobe-filling (R = 1) spike persists. It should also look into whether W Ser primaries are on the hydrogen burning main sequence, or in general what they are. We also need more light curves of W Ser type systems, high resolution line profiles for the (mass) primaries (with particular attention to the W Ser-Algol transition cases), and spectroscopy of low inclination W Serpentis systems, such as KX And.     

Algols as limits on binary evolution scenarios

Evolutionary scenarios must account for Algol binaries surviving their first phase of mass transfer. The outcome of this phase is dependent upon the rapidity of the initial mass transfer, which can be estimated by calculating the radial reponse of potential progenitors to mass loss. Limits on the donor's evolutionary state, and its companion mass, can be placed on systems which would transfer mass on a thermal or dynamical timescale. Slower mass transfer rates are necessary for the successful transition to an Algol. Considering 1.5 and 5.0 M_⊙ models, the former succeed in case A and Br systems, while the latter can do so only in case A systems. To evolve into an Algol binary, all systems seem to require initial mass ratios near one.     

Algols as limits on binary evolution scenarios

Evolutionary scenarios must account for Algol binaries surviving their first phase of mass transfer. The outcome of this phase is dependent upon the rapidity of the initial mass transfer, which can be estimated by calculating the radial reponse of potential progenitors to mass loss. Limits on the donor's evolutionary state, and its companion mass, can be placed on systems which would transfer mass on a thermal or dynamical timescale. Slower mass transfer rates are necessary for the successful transition to an Algol. Considering 1.5 and 5.0 M models, the former succeed in case A and Br systems, while the latter can do so only in case A systems. To evolve into an Algol binary, all systems seem to require initial mass ratios near one.     

The evolution of moderately close and moderately wide binaries

We discuss evolutionary processes in binaries where the primary becomes a red giant with a deep convective envelope before it fills its Roche lobe. Such binaries (late Case B or late Case C, if they evolve conservatively) ought to suffer drastic mass transfer, on a hydrodynamic timescale. In some circumstances this may lead to a common envelope, spiral-in, and finally either a very short-period binary or coalescence. But there appear to be other circumstances in which the outcome is an ordinary Algol, or a wide binary with a white dwarf companion as in Barium stars and some symbiotics. We try to demonstrate that stellar-wind mass loss, enhanced one or two orders of magnitude by tidal interaction with a companion, can vitally affect the approach to RLOF, and indeed may prevent RLOF in binaries with periods over 1000 d. Such mass loss is probably accompanied by angular momentum loss, by magnetic braking combined with tidal friction. The result is that it will not be easy to predict definitively the outcome of evolution in a given zero-age binary.     

The evolution of moderately close and moderately wide binaries

We discuss evolutionary processes in binaries where the primary becomes a red giant with a deep convective envelope before it fills its Roche lobe. Such binaries (late Case B or late Case C, if they evolve conservatively) ought to suffer drastic mass transfer, on a hydrodynamic timescale. In some circumstances this may lead to a common envelope, spiral-in, and finally either a very short-period binary or coalescence. But there appear to be other circumstances in which the outcome is an ordinary Algol, or a wide binary with a white dwarf companion as in Barium stars and some symbiotics. We try to demonstrate that stellar-wind mass loss, enhanced one or two orders of magnitude by tidal interaction with a companion, can vitally affect the approach to RLOF, and indeed may prevent RLOF in binaries with periods over 1000 d. Such mass loss is probably accompanied by angular momentum loss, by magnetic braking combined with tidal friction. The result is that it will not be easy to predict definitively the outcome of evolution in a given zero-age binary.     

NLTE analysis of hot binaries

We present quantitative spectroscopic NLTE analyses of the components of well detached early type binaries (DH Cep, Y Cyg, V453 Cyg, and CW Cep). The position of the stars in the logL-logT _eff diagram is discussed. We find significantly higher temperatures for the components of Y Cygni from spectral analysis by means of unblanketed NLTE model photospheres than those given by the orbit analysis. Therefore the comparison with evolutionary tracks yields larger masses. The spectroscopic temperatures of V453 Cygni and CW Cephei agree with the orbit data, but the evolutionary tracks point to larger masses also. However, if we account for some 2000K lower effective temperatures due to line blanketing, the luminosities, temperatures and masses of all stellar components are in good agreement, except for the case of DH Cep.     

Semidetached systems: Evolutionary viewpoints and observational constraints

The present knowledge on the evolution of semidetached systems is reviewed. Characteristics of observed systems are discussed and general properties tested by the behaviour of theoretical models. New models of mass accreting companion stars are computed. The accretion phase is divided into a fast and slow phase with an accretion rate depending on the initial mass of the mass losing star and on the initial mass ratio, assuming the systems are undergoing a case B of mass transfer. The results are compared with observed systems with masses of the gainers located within the theoretical range. Up to now no computations exist for the evolution of medium mass close binaries including overshooting of the convective core. However some of the influences of extended convective mixing on the interaction of close binaries are investigated. A larger probability for the occurrence of case A of mass exchange and a larger remnant mass at the end of the process are the most important results. Finally the investigation into the origin of individual systems (in mass, mass ratio and period) is discussed, showing that progress both in observations and in theoretical models result in a more detailed and more restricted determination of the initial parameters of the individual systems.Research associate, NFWO, Belgium.     

The evolution of massive stars

The evolution of stars with masses between 15 M ₀ and 100M ₀ is considered. Stars in this mass range lose a considerable fraction of their matter during their evolution.The treatment of convection, semi-convection and the influence of mass loss by stellar winds at different evolutionary phases are analysed as well as the adopted opacities.Evolutionary sequences computed by various groups are examined and compared with observations, and the advanced evolution of a 15M ₀ and a 25M ₀ star from zero-age main sequence (ZAMS) through iron collapse is discussed.The effect of centrifugal forces on stellar wind mass loss and the influence of rotation on evolutionary models is examined. As a consequence of the outflow of matter deeper layers show up and when the mass loss rates are large enough layers with changed composition, due to interior nuclear reactions, appear on the surface.The evolution of massive close binaries as well during the phase of mass loss by stellar wind as during the mass exchange and mass loss phase due to Roche lobe overflow is treated in detail, and the value of the parameters governing mass and angular momentum losses are discussed.The problem of the Wolf-Rayet stars, their origin and the possibilities of their production either as single stars or as massive binaries is examined.Finally, the origin of X-ray binaries is discussed and the scenario for the formation of these objects (starting from massive ZAMS close binaries, through Wolf-Rayet binaries leading to OB-stars with a compact companion after a supernova explosion) is reviewed and completed, including stellar wind mass loss.     

Algol, Beta Lyrae, and W Serpentis: Some new results for three well studied eclipsing binaries

The properties of the eclipsing binaries Algol, Beta Lyrae, and W Serpentis are discussed and new results are presented. The physical properties of the components of Algol are now well determined. High resolution spectroscopy of the H-alpha feature by Richards et al. and by Gillet et al. and spectroscopy of the ultraviolet resonance lines with the International Ultraviolet Explorer satellite reveal hot gas around the BBV primary. Gas flows also have been detected apparently originating from the low mass, cooler secondary component and flowing toward the hotter star through the Lagrangian L1 point. Analysis of 6 years of multi-bandpass photoelectric photometry of Beta Lyrae indicates that systematic changes in light curves occur with a characteristic period of -275 ± 25 days. These changes may arise from pulsations of the B8II star or from changes in the geometry of the disk component. Hitherto unpublished u, v, b, y, and H-alpha index light curves of W Ser are presented and discussed. W Ser is a very complex binary system that undergoes complicated, large changes in its light curves. The physical properties of W Ser are only poorly known, but it probably contains one component at its Roche surface, rapidly transfering matter to a component which is embedded in a thick, opaque disk. In several respects, W Ser resembles an upscale version of a cataclysmic variable binary system.     

Photometric effects of accretion disks in long-period eclipsing binaries

Summary A multi-year photometric program on long-period eclipsing binaries has begun to uncover some properties of accretion disks in these systems. Emission and transmission properties can sometimes be found from light curve features produced by partial eclipses of the disk by the cool star, and by partial occultations of the cool star by the disk. These disks do not have the classical alpha structure. They are optically thin normal to the orbital plane, but may be geometrically thicker than purely gravitationally-stratified disks. Disk gas may be contaminated by dust particles acquired from the outer layers of the cool loser. In some systems, high states, produced by elevated mass accretion by the hot star, occur, suggesting that the mass distribution in the disk is clumpy. However mass-transfer rates are found, they lie between 10^-7 and 10^-6 solar masses per year.While this binary sample is small at the moment, some of its properties are shared with other systems. The author has five-color observations of about a dozen additional systems, which may fill out this picture more fully.     

Who is who in Algol-land ? - Part II -

In part I (De Greve and Packet) we have investigated the occurrence of reversed phases of mass-transfer during Case A evolution in close binaries. If the initial period of a system is shorter than 1–2 days (Early Case A) the reversed phase starts before core hydrogen exhaustion of the gainer (part I). This type of evolution is characterized by at least two phases of slow mass-transfer.We have computed the evolution of four Early Case A systems with initial masses of the loser equal to 3 Mo and 5 Mo. These four systems start mass-exchange when Xc of the primary has decreased to 0.525 (75% of its initial value). They all experience two phases of slow mass-transfer.We find that both phases have about the same duration for all systems. The mass ratios are clearly distinct, being closer to unity during the first phase. In the Hertzsprung-Russell, mass-radius and mass-luminosity diagrams both components remain close to the main-sequence band during slow mass-transfer. Evolution as an Algol is ended when both components overflow their outer critical surface after a second reversal of the mass-transfer.Observed Algol systems evolving in Early Case A are scarce. A search thruogh the catalogue by Giuricin et al. gives us the following candidates: X Tri, SX Aur and V Pup. Based on their mass ratios, SX Aur can tentatively be assigned to the first phase of slow mass transfer and X Tri to the second phase. For V Pup (which is more massive) this choice can not be made with certainty.     

Fundamental parameters of the W Serpentis stars

Application of digital cross-correlation spectroscopy to the spectra of the W Serpentis binaries SX Cas and RX Cas has allowed an accurate determination of the orbits and rotations of the (mass-losing) K-subgiant secondary components. The distortion of the primary radial-velocity curves due to the influence of the prominent accretion disks in these systems has been modelled to first order. This enables us to estimate k ₁, and thereby the mass ratio q ≈ 0.30, to within ～ ± 20%. The absolute radii of the secondaries are derived independently from the observed rotations and periods, assuming synchronous rotation. They show that the stars fill their Roche lobes, or at least very nearly so. Rough fits to the available photometry shows both primaries to be unevolved mid-B stars; that in RX Cas appears completely obscured by the disk. Preliminary spectroscopic data for W Ser and W Cru show some promise for similar analyses of these systems.     

Long-term cycles in cosmic X-ray sources

Most of what we know about galactic X-ray binaries comes from their time variation, particularly periodic variations corresponding to neutron star rotation, and binary motion. Longer cycles or quasi-cycles are much harder to observe because of the shortage of instrumentation suitable for long-term monitoring. Nonetheless, cycle with periods up to a few years have been seen in several galactic binaries.Cycles of 30–300 days have been confirmed for four high-mass systems, LMC X-4, Her X-1, SS433, and Cyg X-1, and are suspected in several others. These cycles are observed in both the X-ray and optical bands, and represent cyclic variations in both the inner and outer parts of the accretion disk. Some component of these systems is precessing, but we are not certain which. It could be a misaligned companion star; the outer rim of the accretion disk, driven by radiative feedback; or the neutron star.Several low-mass X-ray binaries have quasi-periodic cycles, with periods ranging from 1/2 to 2 years. The amplitude of modulation ranges between 50 and 100%, i.e., both persistent and transient objects fall into this class. This activity is reminiscent of the superoutburst cycles of the SU UMa cataclysmic variables, and may be caused by similar mass-transfer instabilities.Periodic outbursts in the Be/neutron star systems seem to result from variable mass transfer in a wide, eccentric orbit. The relationship between the orbital cycle and the flux outbursts, however, is not well understood, and even the equivalence of the outburst and binary cycles remains hypothetical for most objects. Most likely, the periodic outbursts result from enhanced mass transfer at periastron.Compared to other aspects of X-ray astronomy, long-term activity has been much less intensively studied by both observers and theoreticians. A simple all-sky monitor in permanent operation could provide for the X-ray sky the same kind of data base provided to optical observers by the Harvard plates.     

X-ray binaries and stellar evolution

Observational evidence suggests that most — if not all — binary X-ray sources are neutron stars. The evolutionary status and possible formation mechanisms of the type I (massive) and type II (low-mass) X-ray binaries are discussed. The difference between the standard massive X-ray binaries and the Be/X-ray binaries is ascribed to a somewhat different evolutionary history and status, and possible reasons for the existence of short- and long — period X-ray pulsars are discussed. Type II X-ray sources in globular clusters were most probably formed by capture processes; their formation rate inferred from the observations indicates that only a small fraction ( 1 to 10 percent) of the originally formed neutron stars have remained in their clusters. Type II sources in the galactic bulge may also have formed from cataclysmic binaries in which a white dwarf was driven over the Chandrasekhar limit by accretion.     

Evolution into contact of the low-mass close binary systems

We investigated the effect of mass accretion on the secondary components in close binomy systems (M _total ≤ 2.5 M _⊙ M _2,0 ≤ 0.75 M _⊙) exchanging mass in the case A. The evolution of the low-mass close binary systems (M _total ≤ 2.5 M _⊙) exchanging the mass in the case A depends on the three main factors:

-the initial mass ratio (q ₀ = M _2,0/M _1,0), which determines the rate of mass transfer between components;

-the inital mass of the secondary component (M _2,0) and

-the effectiveness of the heating of the photosphere of the secondary component, by infalling matter.

The second factor allows to divide all systems into two essentially different groups:

1. systems in which the secondary component is a star with a radiative envelope, or with a thin convection zone in the uppermost layers;
2. and systems in which secondary component has a thick convective envelope or is fully convective.

The systems from the first group evolve into contact in a characteristic time scale 10⁵ – 10⁷ years, and reach contact after transfering of 0.03 – 0.3 M _⊙. The mass exchange proceeds only in a thermal time scale. For the systems from the group b the effectiveness of the heating of the stellar surface is the most important. In the case when the entropy of the newly accreted matter is the same as the surface entropy of the secondary, a convective star should shrink upon accretion. Then contact binaries are not formed. In the case when the entropy of the infalling matter is greater then that on the surface, the reaction of the secondary is different. The radius of the secondary component grows rapidly in response to accretion, and the systems reaches contact after the 10³ – 3 10⁶ years, and after transfer of 0.002 – 0.2. M _⊙. The reaction of the secondary is determined by the formation of the temperature inversion layer below the stellar surface. Full references in: Sarna, M.J. and Fedorova, A.V. (1988) “Evolutionary status of W UMa-type Binaries — Evolution into contact”, Astron. Astrophys., in press.     

Fundamental parameters of the W Serpentis stars

Application of digital cross-correlation spectroscopy to the spectra of the W Serpentis binaries SX Cas and RX Cas has allowed an accurate determination of the orbits and rotations of the (mass-losing) K-subgiant secondary components. The distortion of the primary radial-velocity curves due to the influence of the prominent accretion disks in these systems has been modelled to first order. This enables us to estimate k ₁, and thereby the mass ratio q 0.30, to within ± 20%. The absolute radii of the secondaries are derived independently from the observed rotations and periods, assuming synchronous rotation. They show that the stars fill their Roche lobes, or at least very nearly so. Rough fits to the available photometry shows both primaries to be unevolved mid-B stars; that in RX Cas appears completely obscured by the disk. Preliminary spectroscopic data for W Ser and W Cru show some promise for similar analyses of these systems.     

Algol,Beta Lyrae,and W Serpentis: Some new results for three well studied eclipsing binaries

The properties of the eclipsing binaries Algol, Beta Lyrae, and W Serpentis are discussed and new results are presented. The physical properties of the components of Algol are now well determined. High resolution spectroscopy of the H-alpha feature by Richards et al. and by Gillet et al. and spectroscopy of the ultraviolet resonance lines with the International Ultraviolet Explorer satellite reveal hot gas around the BBV primary. Gas flows also have been detected apparently originating from the low mass, cooler secondary component and flowing toward the hotter star through the Lagrangian L1 point. Analysis of 6 years of multi-bandpass photoelectric photometry of Beta Lyrae indicates that systematic changes in light curves occur with a characteristic period of -275 ± 25 days. These changes may arise from pulsations of the B8II star or from changes in the geometry of the disk component. Hitherto unpublished u, v, b, y, and H-alpha index light curves of W Ser are presented and discussed. W Ser is a very complex binary system that undergoes complicated, large changes in its light curves. The physical properties of W Ser are only poorly known, but it probably contains one component at its Roche surface, rapidly transfering matter to a component which is embedded in a thick, opaque disk. In several respects, W Ser resembles an upscale version of a cataclysmic variable binary system.