The Tarantula Massive Binary Monitoring VII. On the nature of the eccentric O+BH binary candidate VFTS 812

TL;DR

This study reevaluates VFTS 812, an eccentric $P=17.28$ day O4 V SB1 in the Tarantula Nebula, as a potential OB+BH binary. Utilizing 26 high-resolution FLAMES-IFU spectra and spectral disentangling, the authors derive a primary evolutionary mass of $M_1 = 53.4^{+6.0}_{-5.2}\,M_\odot$ and a mass-function–based secondary minimum of $M_{2, ext{min}} > 5.1\,M_\\odot$, while finding no detectable luminous companion down to roughly $6\,M_\odot$. Two main outcomes emerge: either VFTS 812 hosts a dark BH companion, or a lower-mass luminous companion remains undetectable with current data. The work highlights the importance of UV and higher-S/N spectroscopy to confirm the companion's nature and to search for subtle accretion signatures, contributing to our understanding of OB+BH binaries and massive binary evolution.

Abstract

Massive O-type stars ($M\gtrsim15\,M_\odot$) with an X-ray quiet black hole (BH) companion represent a crucial stage in massive binary evolution leading to binary BH mergers. The population of such binaries remains elusive, with $\lesssim5$ candidate or confirmed systems. The Tarantula nebula harbors thousands of massive stars, 2-3 % of which are expected to have BH companions. It is therefore an ideal place to hunt for such systems. Here we analyse 30 epochs of VLT/FLAMES IFU high-resolution observations of the H$δ$ region, as well as archival FLAMES spectroscopy, of VFTS 812, a 17-day single-lined spectroscopic binary with an O4V primary and a minimum secondary mass of $5.1\,M_\odot$. Following careful removal of the nebular contamination, spectral disentangling on the new data did not reveal any signature of the hidden companion. We derive $T_\mathrm{eff}=49^{+3}_{-4}$ kK, $\log L/L_\odot=5.7\pm0.1$ and $v_\mathrm{rot,max}{\rm \,sin\,}i=110^{+25}_{-35}$ km/s for the O4V component, yielding a (single star) evolutionary mass of $53^{+6}_{-5}$ $M_\odot$ and an age in the range of 0-1.6 Myr. Using injection tests of various luminous artificial companions in our data, we exhaustively rule out the presence of any luminous signature from a main sequence star more massive than $6\,M_\odot$. We discuss the possible nature of the companion, suggesting that the rejuvenated O star + BH companion is the most suitable scenario to consistently explain the location, (rejuvenated) young age, eccentricity and lack of companion signature. While this establishes VFTS 812 as a strong candidate O+BH system, follow-up observations are deemed necessary for robust confirmation and to search for accretion signatures on the O4V star.

Figures (6)

• Figure 1: Reduced chi-squared map from grid disentangling as a function of $K_1$ (y-axis) and $K_2$ (x-axis).
• Figure 2: Disentangled secondary spectra (light red) from the no-injection case (first panel) and different mock datasets (remaining panels) along with labels showing masses of the injected models. Also shown are the injected model spectra (dark red) and the best fit flat line (black). The disentangled spectra are binned by a factor of 3 for better visualization.
• Figure B.1: Four examples of IFU data shown with their Modified Julian Dates (MJDs) at start of observation. In every panel, the 20 spaxels barring the corners are scaled with their median flux, with light colors representing low and dark representing high flux. The position and point spread function of the star can be seen varying across the four epochs.
• Figure B.2: For the same four examples as Figure \ref{['fig:ifu']}, we show spectra for the star spaxels (dashed), sky spaxels (dotted) and the resulting sky-subtracted spectrum (solid). Also highlighted is the "line region" (yellow) from 4103.9--4106.7 that was most affected by nebular contamination. The sky subtracted spectra in the line region clearly show variable residual features across the four epochs.
• Figure C.1: Spectral fit of VFTS 812. Data is shown in black; the dark green line denotes the best fit and the light green region the 1-$\sigma$ uncertainty region of the fit without turbulent pressure. The dashed orange line and the hatched orange area show the best fit and 1-$\sigma$ uncertainty of the fit with turbulence.