The long-period massive binary HD 54662 revisited^⋆,⋆⋆

¹ Space sciences, Technologies and Astrophysics Research (STAR) Institute, Université de Liège, Allée du 6 Août 19c, Bât B5c, 4000 Liège, Belgium
e-mail: emossoux@ulg.ac.be
² Instituut voor Sterrenkunde, KU Leuven, Celestijnenlaan 200D, Bus 2401, 3001 Leuven, Belgium

Received: 12 October 2017
Accepted: 7 February 2018

Abstract

Context. HD 54662 is an O-type binary star belonging to the CMa OB1 association. Because of its long-period orbit, this system is an interesting target to test the adiabatic wind shock model.

Aims. The goal of this study is to improve our knowledge of the orbital and stellar parameters of HD 54662 and to analyze its X-ray emission to test the theoretical scaling of X-ray emission with orbital separation for adiabatic wind shocks.

Methods. We applied a spectral disentangling code to a set of optical spectra to determine the radial velocities and the individual spectra of the primary and secondary stars. The orbital solution of the system was established and the reconstructed individual spectra were analyzed by means of the CMFGEN model atmosphere code. We fitted two X-ray spectra using a Markov chain Monte Carlo algorithm and compared these spectra to the emission expected from adiabatic shocks.

Results. We determine an orbital period of 2103.4 days, a surprisingly low orbital eccentricity of 0.11, and a mass ratio m₂/m₁ of 0.84. Combined with the orbital inclination inferred in a previous astrometric study, we obtain surprisingly low masses of 9.7 and 8.2 M_⊙. From the disentangled primary and secondary spectra, we infer O6.5 spectral types for both stars, of which the primary is about two times brighter than the secondary. The softness of the X-ray spectra for the two observations, the very small variation of best-fitting spectral parameters, and the comparison of the X-ray-to-bolometric luminosity ratio with the canonical value for O-type stars allow us to conclude that X-ray emission from the wind interaction region is quite low and that the observed emission is rather dominated by the intrinsic emission from the stars. We cannot confirm the runaway status previously attributed to HD 54662 by computing the peculiar radial and tangential velocities. We find no X-ray emission associated with the bow shock detected in the infrared. Conclusions. The lack of hard X-ray emission from the wind-shock region suggests that the mass-loss rates are lower than expected and/or that the pre-shock wind velocities are much lower than the terminal wind velocities. The bow shock associated with HD 54662 possibly corresponds to a wind-blown arc created by the interaction of the stellar winds with the ionized gas of the CMa OB1 association rather than by a large differential velocity between the binary and the surrounding interstellar medium.