An international team of researchers published a study in the journal Nature in 2026 identifying a mass gap in the population of merging black holes at approximately 45 solar masses, providing observational evidence for pair-instability supernovae. The analysis, based on data collected by four gravitational wave detectors over multiple years, found a sharp cutoff in the mass distribution of the smaller black hole component in observed merger events.

The researchers analyzed possible types of black hole collision scenarios, categorizing mergers by whether the black holes involved were first-generation — meaning they formed directly from stellar collapse and had not undergone prior mergers — or second-generation, meaning they were themselves products of earlier mergers. A first-generation merger involves two black holes that should each fall below the pair-instability mass cutoff, while a merger between a first-generation and a second-generation black hole may involve the larger component exceeding that limit. Mergers between two second-generation black holes, in which both components may exceed the cutoff, were estimated to account for about 1 percent of all observed mergers.

Most mergers involving a second-generation black hole were found to pair a first-generation black hole with a second-generation one. In such events, the smaller black hole is first-generation and should adhere to the pair-instability mass limit. The study identified this limit at approximately 45 solar masses, with an uncertainty of about five solar masses. An independent analysis using black hole spin data yielded a mass cutoff near the same value, and a previous analysis on a subset of the data had found a similar result. The spin magnitudes of the more massive black holes in the analyzed mergers were high, consistent with them being products of previous mergers.

Theoretical models predict a sharp cutoff in black hole mass distribution due to pair-instability processes. In very massive stars, photon energy in the core can convert into electron-positron pairs, reducing radiation pressure and inducing rapid core contraction. In sufficiently massive stars, this contraction triggers sudden oxygen fusion that releases enough energy to completely unbind the star, leaving no remnant — a pair-instability supernova. In some massive stars, pulsational oxygen fusion can instead expel the outer layers, leaving a smaller star that later forms a lower-mass black hole.

Black holes form when a star's core collapses during a supernova. In a typical supernova, a star's outer layers explode outward while the inner layers collapse inward, with part of the collapsing material forming a black hole or, if the mass is lower, a neutron star. While theoretical models predict pair-instability supernovae, these events have been difficult to confirm through direct observation. Several candidate events have been proposed, but clear observational criteria to distinguish them from other types of supernovae have not been established. Additional years of gravitational wave observations are expected to reduce uncertainties in mass cutoff estimates.