2020ApJ...905L..15B

The Laser Interferometer Gravitational-Wave Observatory (LIGO)/Virgo Collaboration reported the detection of the most massive black hole-black hole (BH-BH) merger to date with component masses of 85M_☉ and 66M_☉ (GW190521). Motivated by recent observations of massive stars in the 30 Doradus cluster in the Large Magellanic Cloud (M_* >= 200M_☉; e.g., R136a) and employing newly estimated uncertainties on pulsational pair-instability mass loss (that allow for the possibility of forming BHs with mass up to M_BH ∼ 90M_☉), we show that it is trivial to form such massive BH-BH mergers through the classical isolated binary evolution (with no assistance from either dynamical interactions or exotica). A binary consisting of two massive (180M_☉ + 150M_☉) Population II stars (metallicity: Z ≃ 0.0001) evolves through a stable Roche lobe overflow and common envelope episode. Both exposed stellar cores undergo direct core collapse and form massive BHs while avoiding pair-instability pulsation mass loss or total disruption. LIGO/Virgo observations show that the merger rate density of light BH-BH mergers (both components: M_BH < 50M_☉) is of the order of 10-100Gpc^–3yr^–1, while GW190521 indicates that the rate of heavier mergers is 0.02-0.43Gpc^–3yr^–1. Our model (with standard assumptions about input physics), but extended to include 200M_☉ stars and allowing for the possibility of stellar cores collapsing to 90M_☉ BHs, produces the following rates: 63Gpc^–3yr^–1 for light BH-BH mergers and 0.04Gpc^–3yr^–1 for heavy BH-BH mergers. We do not claim that GW190521 was formed by an isolated binary, but it appears that such a possibility cannot be excluded.