Astronomy and Astrophysics
Aims: Multi-epoch optical spectra of NGC 2004#115 are used to determine if this binary can be explained by either of these two scenarios, and hence shed additional light on these interesting systems.
Methods: VLT-FLAMES and SALT-HRS data covering a baseline of ∼20 years were analyzed to determine radial velocities and orbital parameters, while non-LTE model atmospheres were used to determine stellar parameters and the relative brightness of the system components. Archive MACHO, Gaia, and XMM-Newton data provide additional constraints on the system.
Results: NGC 2004#115 is found to be a triple system consisting of an inner binary with a period P = 2.92 d, eccentricity e ∼ 0.0, and mass function f = 0.07 M⊙. The only firmly detected star in this inner binary is a B2 star, the primary, with a projected rotational velocity (ve sin i) of 10 km s−1 and a luminosity of log L/L⊙ = 3.87. It contributes ∼60% of the total V-band light, with the tertiary contributing the other ∼40% of the light, while the secondary is not detected in the optical spectrum. The possibility that the primary is a low mass inflated stripped star is excluded since its Roche radius would be smaller than its stellar radius in such a compact system. A main sequence star of mass 8.6 M⊙ is therefore inferred; however, the assumption of synchronous rotation leads to a secondary mass in excess of 25 M⊙, which would therefore be a black hole. The tertiary is detected as a fainter blended component to the hydrogen and helium lines, which is consistent with a slightly less massive B-type star, though with ve sin i ∼ 300 km s−1. The data do not permit the characterization of the outer period, though it likely exceeds 120 days and is therefore in a stable configuration. The disk-like emission is variable, but may be associated with the inner binary rather than the rapidly rotating tertiary. XMM-Newton provides an upper limit of 5 × 1033 ergs s−1 in the 0.2-12 keV band which is consistent with, though not constraining, the system hosting a quiescent B+BH binary. A number of caveats to this scenario are discussed.