Context : Massive binaries hosting a black hole (OB+BH) represent a critical phase in the production of BH mergers in the context of binary evolution. In spite of this, such systems have so far largely avoided detection. Single-lined spectroscopic (SB1) O-type binaries are ideal objects to search for elusive BH companions. Moreover, SB1 binaries hosting two main sequence stars probe a regime of more extreme mass ratios and longer periods compared to double-lined binaries (SB2), and they are thus valuable for establishing the natal mass ratio distribution of massive stars.
Aims : We characterise the hidden companions in 51 SB1 O-type and evolved B-type binaries identified in the Large Magellanic Cloud (LMC) in the framework of the VLT-FLAMES Tarantula Survey (VFTS) and its follow-up, the Tarantula Massive Binary Monitoring (TMBM). The binaries cover periods between a few days to years (0 < log P < 3 [d]). Our goals are to hunt for BHs and sample the low-mass end of the mass-ratio distribution.
Methods : To uncover the hidden companions, we implemented the shift-and-add grid disentangling algorithm using 32 epochs of spectroscopy acquired in the framework of TMBM with the FLAMES spectrograph, allowing us to detect companions contributing as little as ≈1-2% to the visual flux. We further analysed OGLE photometric data for the presence of eclipses or ellipsoidal variations.
Results : Out of the 51 SB1 systems, 43 (84%) are found to have non-degenerate stellar companions, of which 28 are confident detections and 15 are less certain (SB1: or SB2:). Of these 43 targets, one is found to be a triple (VFTS 64), and two are found to be quadruples (VFTS 120, 702). Our sample includes a total of eight eclipsing binaries. The remaining eight targets (16%) retain an SB1 classification. We modelled the mass-ratio distribution as f (q) ∝ q^κ, and derived K through a Bayesian approach. We used massratio constraints from previously known SB2 binaries, newly uncovered SB2 binaries, and SB1 binaries, while accounting for binary detection bias. We found K = 0.2 ± 0.2 for the entire sample and κ = −0.2 ± 0.2 when excluding binaries with periods shorter than 10 d. In contrast, K = 1.2 ± 0.5 was retrieved for tight binaries (P < 10 d), and it is proposed here to be a consequence of binary interactions. Aside from the unambiguous O+BH binary VFTS 243, which was analysed in detail in a separate paper, we identified two additional OB+BH candidates: VFTS 514 and 779.
Conclusions : Our study firmly establishes a virtually flat natal mass-ratio distribution (κ = 0) for O-type stars at LMC metallicity, covering the entire mass-ratio range (0.05 < q < 1) and periods in the range 0 < log P < 3 [d]. The nature of the OB+BH candidates should be verified through future monitoring, but the frequency of OB+BH candidates is generally in line with recent predictions at LMC metallicity.

Disentangled spectra are only available at the CDS via anonymous ftp to ftp://cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsarc.u-strasbg.fr/viz-bin/cat/J/A+A/665/A148