Direct imaging and dynamical mass of a benchmark T-type brown dwarf companion to HD 167665

Abstract

A low-mass companion potentially in the brown dwarf mass regime was discovered on a sim 12 yr orbit (sim 5.5 au) around HD\,167665 using radial velocity (RV) monitoring. Joint RV--astrometry analyses confirmed that HD\,167665B is a brown dwarf with precisions on the measured mass of sim 4--9<!PCT!>. Brown dwarf companions with measured mass and luminosity are valuable for testing formation and evolutionary models. However, its atmospheric properties and luminosity are still unconstrained, preventing detailed tests of evolutionary models We further characterize the HD\,167665 system by measuring the luminosity and refining the mass of its companion and reassessing the stellar age. We present new high-contrast imaging data of the star and of its close-in environment from SPHERE and GRAVITY, which we combined with RV data from CORALIE and HIRES and astrometry from Hipparcos and Gaia The analysis of the host star properties indicates an age of 6.20pm 1.13 Gyr. GRAVITY reveals a point source near the position predicted from a joint fit of RV data and Hipparcos Gaia proper motion anomalies. Subsequent SPHERE imaging confirms the detection and reveals a faint point source of contrast of $ H2$\,=\,10.95pm 0.33 mag at a projected angular separation of sim 180 mas. A joint fit of the high-contrast imaging, RV, and Hipparcos intermediate astrometric data together with the Gaia astrometric parameters constrains the mass of HD\,167665B to sim 1.2<!PCT!>, 60.3pm 0.7 J . The SPHERE colors and spectrum point to an early or mid-T brown dwarf of spectral type T4$^ $. Fitting the SPHERE spectrophotometry and GRAVITY spectrum with synthetic spectra suggests an effective temperature of sim 1000--1150 K, a surface gravity of sim 5.0--5.4 dex, and a bolometric luminosity log($L$/$L_ $ dex. The mass, luminosity, and age of the companion can only be reproduced within 3sigma by the hybrid cloudy evolutionary models of Saumon Marley (2008), whereas cloudless evolutionary models underpredict its luminosity.