The nature of ultraluminous X-ray astronomical sources has long been unclear. The latest observations of these rare systems provide some crucial clues, but still leave theorists scratching their heads. See Letters p.198 & p.202
In the late 1970s, astronomers discovered objects that emit unusually bright X-rays1. Given their extreme X-ray luminosity, these ultraluminous X-ray sources were thought to contain black holes. However, the mass of the black holes powering such objects has been a topic of much debate. Two studies in this issue, by Motch et al.2 (page 198) and Bachetti et al.3 (page 202), together with two recent reports by Pasham et al.4 and Liu et al.5, are changing our views about these systems.
Most black holes are created during the violent deaths of massive stars. Although such stellar-mass black holes weigh about 3 to 100 times the mass of our Sun, they can be difficult to see. Their extreme gravitational pull attracts anything that strays too close, even light. So, to learn more about them, we must observe them indirectly, by studying the effect they have on their environment.
If the stellar-mass black hole is orbited by a companion star, we can study its effects on the star. The black hole can pull material from the star's wind and/or surface. As material falls in (accretes), forming an accretion disk, some of the material's gravitational potential energy is lost as light — mainly X-rays. Such X-ray binary systems (Fig. 1) contain not just a disk but also an optically thin (transparent) medium, which is thought to sit either above and below the disk (a corona) or between the disk and the black hole (a hot inner flow). As the rate at which material travels through the accretion disk changes, the geometry of the system, and so its accretion state, will change accordingly.
X-ray binaries can also contain neutron stars — the smaller cousins of stellar-mass black holes. Like stellar-mass black holes, neutron stars are born in violent star deaths, but they are lighter, weighing only around 1.4 solar masses. The gravitational pull of these systems is again very strong, drawing in material. But unlike black holes, light can escape from them, and we can see their surface.
Black holes also have much heavier cousins, which reside in the centres of galaxies. They are known as supermassive black holes, and weigh millions to billions of solar masses. They show similar accretion states to stellar-mass black holes, with their luminosity apparently scaling with their mass. Although we understand the formation of stellar-mass black holes, that of supermassive black holes is not understood.
Ultraluminous X-ray sources (ULXs) are intermediate in luminosity between stellar-mass and supermassive black holes, and so it was thought that they might be X-ray binaries containing intermediate-mass black holes (weighing hundreds to thousands of solar masses). In the past few years, evidence has been found suggesting that the shape of the X-ray spectra of these sources is distinctly different from that of previously observed accretion states6,7. This suggests that we may be observing stellar-mass black holes that are behaving oddly, possibly residing in a new extreme (ultraluminous) accretion state.
Motch et al.2 present a multi-wavelength analysis of an X-ray source dubbed P13 — a ULX in the spiral galaxy NGC 7793 that has been observed to vary dramatically in luminosity (by a factor of 40). The authors' X-ray analysis of P13 shows evidence of the ultraluminous state6 at its brightest X-ray emission. Furthermore, their optical data indicate that the source has a massive, luminous companion star, known as a supergiant B star. By making many optical observations over 8 years, the authors find that the star is being continually bombarded with X-ray radiation from the black hole's accretion disk, leading to X-ray heating of the star's surface even when the X-ray luminosity is decreased. This suggests that the star is exposed to bright X-ray emission from the inner parts of the disk, even though the geometry of the system is shielding this X-ray-emitting region of the disk from Earth's line of sight. On the basis of these observations, the researchers conclude that P13 contains a black hole of around 15 solar masses, confirming the idea that ULXs can be powered by stellar-mass black holes in an extreme accretion state.
This hypothesis is supported by Liu and colleagues' study5 of a source called M101 ULX-1. Their optical data indicate that this source can be explained by a black hole of about 20–30 solar masses fed by a Wolf–Rayet star, an ageing massive star with strong winds. These results, together with those of Motch and colleagues2, suggest that the mysterious nature of ULXs may now have been solved.
However, things are also becoming complicated. Analyses3,4 of two of the ULXs in the cigar-shaped galaxy M82 — M82 X-1 and M82 X-2 — show that some of these sources may be more massive or more extreme than stellar-mass black holes in an ultraluminous state.
M82 X-1 is the brightest ULX in this galaxy. Pasham et al.4 performed an X-ray study of this ULX and found two almost-periodic variations. Such quasi-periodic oscillations have previously been observed in stellar-mass black-hole systems, and seem to scale with mass8. Pasham and colleagues used this mass-scaling relationship to infer a black-hole mass of about 400 times that of the Sun. This places the object in the range of intermediate-mass black holes.
Conversely, M82 X-2 seems to be more extreme than was previously thought possible. Using several X-ray telescopes, Bachetti et al.3 find that M82 X-2 does not contain a black hole, but instead contains a neutron star. This object was seen to be pulsing, something not possible for black holes. Such pulsing can come only from objects known as pulsars — magnetic neutron stars with fields so strong that material from the accretion disk is funnelled onto their poles, creating a hotspot. As these pulsars spin, beams of light from the hotspots sweep in and out of Earth's line of sight. Astronomers see this phenomenon as pulses of light in the same way as pulses are observed from a lighthouse as its lamp rotates. However, this system seems to be pumping out massive amounts of energy, suggesting an accretion rate that is 10 times larger than anything previously measured for these sources. It is also 100 times brighter than the original theoretical limit for such sources. This finding challenges our current theories of accretion physics, specifically of accretion onto magnetic neutron stars.
These results3,4 will surprise the ULX community. They show that ULXs are heterogeneous, and that more-detailed studies of individual sources can push the boundaries of our understanding, not only of the field of ULXs but also of other fields. The confirmation that stellar-mass black holes can accrete at extreme rates2,3 could help to unlock the mystery of the rapid growth of massive black holes in the early Universe. With the discovery of intermediate-mass black holes4, models of the formation of supermassive black holes and hence galaxy formation can be tested: research indicates that these two processes are linked. The finding that neutron stars can reach such high X-ray luminosities3 will probably leave theorists scratching their heads to find ways of pushing neutron-star and accretion physics to new extremes. This is an exciting time for the study of ULXs — a small field that can have a far-reaching impact.
Fabbiano, G. Annu. Rev. Astron. Astrophys. 27, 87–138 (1989).
Motch, C., Pakull, M. W., Soria, R., Grisé, F. & Pietrzyński, G. Nature 514, 198–201 (2014).
Bachetti, M. et al. Nature 514, 202–204 (2014).
Pasham, D. R., Strohmayer, T. E. & Mushotzky, R. F. Nature 513, 74–76 (2014).
Liu, J.-F., Bregman, J. N., Bai, Y., Justham, S. & Crowther, P. Nature 503, 500–503 (2013).
Gladstone, J. C., Roberts, T. P. & Done, C. Mon. Not. R. Astron. Soc. 397, 1836–1851 (2009).
Bachetti, M. et al. Astrophys. J. 778, 163 (2013).
McHardy, I. M., Koerding, E., Knigge, C., Uttley, P. & Fender, R. P. Nature 444, 730–732 (2006).