Context. The evolution of massive stars is dominated by interactions within binary and multiple systems. In order to accurately model this evolution, it is necessary to investigate all possible forms of an interaction in binary systems that may affect the evolution of the components. One of the “laboratories” plausible for this kind of investigation is the massive eccentric binary system MACHO 80.7443.1718 (ExtEV), which exhibits an exceptionally large amplitude of light variability close to the periastron passage of its 32.8-day orbit. Aims. We examine whether the light variability of ExtEV can be explained by a wind-wind collision (WWC) binary system model. We also critically review other models proposed to explain the light curve of ExtEV. Methods. We conducted an analysis of (i) the broadband multicolor photometry of ExtEV spanning a wide range of wavelengths from the ultraviolet to near-infrared, (ii) the time-series space photometry from the Transiting Exoplanet Survey Satellite (TESS), (iii) ground-based Johnson UBV photometry, and (iv) time-series high-resolution spectroscopy. To derive the parameters of the primary component of the system, we fit the spectral energy distribution (SED) and calculated evolutionary models of massive stars that included mass loss. Using radial-velocity data, we determined the spectroscopic parameters of the system. We also fit an analytical model of light variations to the TESS light curve of ExtEV. Results. The ExtEV system exhibits an infrared excess, indicating an increased mass-loss rate. The system does not match the characteristics of B[e] stars, however. We rule out the possibility of the presence of a Keplerian disk around the primary component. We also argue that the scenario with periodic Roche-lobe overflow at periastron may not be consistent with the observations of ExtEV. Analysis of the SED suggests that the primary component has a radius of about 30 R⊙ and a luminosity of ∼6.6 × 105 L⊙. With the analysis of the radial-velocity data, we refine the orbital parameters of ExtEV and find evidence for the presence of a tertiary component in the system. Using evolutionary models we demonstrate that the primary component’s mass is between 25 and 45 M⊙. We successfully reproduced the light curve of ExtEV with our analytical model, showing that the dominant processes shaping its light curve can be attributed to the atmospheric eclipse and light scattered in the WWC cone. We also estimate the primary’s mass loss rate due to stellar wind for 4.5 × 10−5 M⊙ yr−1. Conclusions. ExtEV is most likely not an extreme eccentric ellipsoidal variable, but rather an exceptional WWC binary system. The mass loss rate we derived exceeds theoretical predictions by up to two orders of magnitude. This implies that the wind in the system is likely enhanced by tidal interactions, rotation, and possibly also tidally excited oscillations. Therefore, ExtEV represents a rare evolutionary phase of a binary system that may help to understand the role of companion-driven enhanced mass loss in the evolution of massive binary systems.
Read full abstract