A study of the F-giant star $\theta$ Scorpii A: a post-merger rapid rotator?

Abstract

We report high-precision observations of the linear polarization of the F1$\,$III star $\theta$ Scorpii. The polarization has a wavelength dependence of the form expected for a rapid rotator, but with an amplitude several times larger than seen in otherwise similar main-sequence stars. This confirms the expectation that lower-gravity stars should have stronger rotational-polarization signatures as a consequence of the density dependence of the ratio of scattering to absorption opacities. By modelling the polarization, together with additional observational constraints (incorporating a revised analysis of Hipparcos astrometry, which clarifies the system's binary status), we determine a set of precise stellar parameters, including a rotation rate $\omega\, (= \Omega/\Omega_{\rm c})\ge 0.94$, polar gravity $\log{g_p} = 2.091 ^{+0.042}_{-0.039}$ (dex cgs), mass $3.10 ^{+0.37}_{-0.32}$ solar masses, and luminosity $\log(L/Lsun) =3.149^{+0.041}_{-0.028}$. These values are incompatible with evolutionary models of single rotating stars, with the star rotating too rapidly for its evolutionary stage, and being undermassive for its luminosity. We conclude that $\theta$ Sco A is most probably the product of a binary merger.