Until now, ultra-luminous X-ray sources (ULXs) were thought to be black holes, because their high luminosity implied a mass exceeding by far the maximal mass of a neutron star. The most luminous of them were thought, furthermore, to be of a rare class of intermediate-mass black holes. The surprising discovery of pulsations from one of them now shakes this interpretation, and suggests that at least some neutron stars can become much more luminous than previously thought.

ULXs were discovered in nearby galaxies by the Einstein Observatory in the 1980s. These sources are characterized by X-ray luminosities that are intermediate between normal X-ray binaries and active galactic nuclei (AGN). If luminosity simply scaled with the mass of the accreting compact object, ULXs should be intermediate black holes with masses typically 100 to 10,000 times that of the Sun. This is an unusual mass range for black holes, which are more commonly found either with masses of about 10 solar masses, typical of stars, or with millions of solar masses, as in the case of those powering AGNs at the centres of galaxies.

A simple mass-luminosity relation arises naturally from the equilibrium between the inward gravitational force and the outward radiation pressure acting on matter accretion. Indeed, accretion can only increase as long as the resulting luminosity does not exceed what is known as the Eddington limit, at which the radiation pressure stops accretion and generates an intense outward wind. The Eddington luminosity is linearly proportional to mass and has a value of about 1031 W for a solar-mass star, which is about 10,000 times the luminosity of the Sun.

Although the Eddington limit holds, strictly, only for isotropic accretion, it serves as an order-of-magnitude upper limit to the luminosity of a source of a given mass. A ULX with a luminosity of 1033 W should, therefore, indicate the presence of a black hole at least 100 times the mass of the Sun. This argument is now disproved strongly by the detection of pulsed X-ray emission from a ULX in the nearby galaxy Messier 82 (M82), reported by Matteo Bachetti from the University of Toulouse and colleagues.

This source, M82 X-2, is the second brightest X-ray source in this star-forming galaxy, and can reach a luminosity exceeding 1033 W. The clear detection of pulsations with a period of 1.37 s and an orbital modulation of 2.5 days identifies the source as a binary system that is composed of a neutron star accreting gas from a massive companion star. The pulsed emission was observed in the 3–30 keV X-ray range by the Nuclear Spectroscopic Telescope Array, a NASA satellite launched from below an aeroplane on 13 June 2012. Confirmation that the pulsating source is indeed the ULX M82 X-2 came from contemporaneous observations by the Chandra X-ray Observatory and the Swift satellite.

The discovery of pulsations in M82 X-2 was made possible thanks to a long observation campaign in early 2014 of the M82 galaxy triggered by the explosion of the supernova SN 2014J (CERN Courier October 2014 p17). It proves that at least some ULXs can be accreting pulsars, rather than massive black holes. Theorists are now left with the challenge of proposing a model to explain how a pulsar can radiate at about 100 times its Eddington luminosity.