Which black hole formed first? Mass-ratio reversal in massive binary stars from gravitational-wave data

Population inference of gravitational-wave catalogues is a useful tool to translate observations of black-hole mergers into constraints on compact-binary formation. Different formation channels predict identifiable signatures in the astrophysical distributions of source parameters, such as masses and spins. One example within the scenario of isolated binary evolution is mass-ratio reversal: even assuming efficient core-envelope coupling in massive stars and tidal spin-up of the stellar companion by the first-born black hole, a compact binary with a lighter, non-spinning first-born black hole and a heavier, spinning second-born black hole can still form through mass transfer from the initially more to less massive progenitor. Using current LIGO/Virgo observations, we measure the fraction of sources in the underlying population with this mass-spin combination and interpret it as a constraint on the occurrence of mass-ratio reversal in massive binary stars. We modify commonly used population models by including negligible-spin subpopulations and, most crucially, non-identical component spin distributions. We do not find evidence for subpopulations of black holes with negligible spins and measure the fraction of massive binary stars undergoing mass-ratio reversal to be consistent with zero and $<32\%$ ($99\%$ confidence). The dimensionless spin peaks around $0.2\unicode{x2013}0.3$ appear robust, however, and are yet to be explained by progenitor formation scenarios.