The detection of two candidate black holes in a dense system of stars in the Milky Way suggests that a larger population of such objects might be lurking in this system. See Letter p.71
The Milky Way globular cluster M22, a massive, dense star cluster containing close to a million stars, is starting to reveal its secrets. On page 71 of this issue, Strader et al.1 present long-exposure radio images of M22, obtained by the Karl G. Jansky Very Large Array of radio antennas, showing two previously unknown sources of radio waves. The sources are candidates for 'stellar-mass' black holes, with masses 10–20 times that of the Sun. These are the first possible black holes to be detected in massive, old Milky Way star clusters such as M22. Futhermore, by virtue of the fact that two black holes were found, and were found in the cluster centre, the discovery provides insight into the dynamic evolution of the black-hole population.
Although the existence of black holes was predicted by theory almost a century ago2, it was only in the past three decades that observational evidence became available. Because black holes are dark, the only way to detect them is through their gravitational influence on surrounding matter. In most cases, this requires that the black hole is in a binary system with another object that is either transferring mass to the black hole (accreting), at the same time emitting a large amount of energy as X-rays, or that it is bright enough to allow measurements of its radial velocity to be made.
In the present study, however, the main evidence for the interpretation of the two new sources as black holes comes from the relationship between the observed radio emission and the X-ray emission. The radio emission from a black hole is usually attributed to radiation from jets of gas emanating from either side of a disk of gas, which is accreted from a companion star (Fig. 1). The X-ray emission results from strong shear in the inner part of the disk and from gas turbulence, which together heat the gas to such high temperatures that it releases X-rays.
Although many of the details of the connection between the gas jet and disk remain unclear, observations3 of accreting, stellar-mass black holes led to the determination of a relationship between the X-ray and radio emissions, for low accretion rates. Thus, as X-ray luminosity decreases, the radio emission becomes increasingly dominant. Perhaps most importantly, observations4 as well as theoretical studies5 show that the radio-to-X-ray emission ratio increases with black-hole mass. This makes radio observations ideally suited to detect stellar-mass black holes6.
Strader et al. argue that the fact that the two new radio sources were not detected by the Chandra X-ray satellite places an upper limit on the sources' X-ray luminosity. When combined with their radio emission, this limit yields a large minimum radio-to-X-ray ratio, corresponding to stellar-mass black holes each with 10–20 solar masses. The interpretation of the sources as two black holes is compelling, not least because further support comes from, among other independent evidence, their location close to the cluster centre. In a 'self-gravitating' stellar system (in which all the individual components are held under the combined gravity of the object as a whole) which is in thermal equilibrium, the average distance of an object from the centre is a function of its mass, with more-massive objects lying farther in. Given that the core of M22 would have reached approximate thermal equilibrium in a relatively short time (0.3 billion years) compared with the cluster's age (12 billion years), the positions of the two sources can be used to determine the masses of the black holes. Using this approach, the authors deduced that the black holes are about 15 times more massive than the Sun — which is in accord with their previous calculation of 10–20 solar masses.
The detection of two accreting black holes in a globular cluster has implications for our understanding of the structure and dynamic evolution of dense stellar systems. First, there may be more than two black holes in M22, either as single entities or in binary systems in which mass accretion does not occur. Theoretical calculations7 of the formation of binary systems composed of a black hole and a white-dwarf star in globular clusters indicate that, over a period of 10 billion years, only 2–40% of the total retained black-hole population forms black-hole–white-dwarf binaries that have observable gas accretion. Therefore, if the sources detected in M22 were in binaries with white dwarfs, M22 could contain as many as 100 black holes. Second, simulations of clusters that have a large population of stellar-mass black holes showed8 that the black-hole population leads to a substantial expansion of the cluster core. This is mainly caused by frequent ejections of the black holes from the core into the outer cluster regions by close gravitational encounters with other black holes. Strader et al. suggest that this expansion could explain why M22 has the fifth-largest core radius among bright Milky Way globular clusters.
Finally, and perhaps most significantly, Strader et al. point out that the discovery of two stellar-mass black holes in M22 challenges a hypothesis that has been held for decades — that the population of black holes disappears rapidly through gravitational interactions, such that only one, or a binary system of two, black holes remain at the typical age of globular clusters (1010 years)9,10,11. This would be all the more important if the black holes detected in M22 were part of a large population; Strader and colleagues' results indicate that more black holes may be coupled to the cluster.
If many black holes are retained in globular clusters, we would expect an increased detection of gravitational waves from the merging of black-hole binaries. Although the interaction rate between black holes, and consequently the formation of black-hole–black-hole binaries, would be larger if the black holes were decoupled from the cluster, the rate of destruction of potential black-hole binaries would likewise increase, leading overall to fewer merger events12. Future gravitational-wave searches will allow this expectation to be tested, and the full impact of Strader and colleagues' findings to be revealed.
Strader, J., Chomiuk, L., Maccarone, T. J., Miller-Jones, J. C. A. & Seth, A. C. Nature 490, 71–73 (2012).
Schwarzchild, K. Sber. K. Preuss. Akad. Wiss. 7, 189–196 (1916).
Gallo, E., Fender, R. P. & Pooley, G. G. Mon. Not. R. Astron. Soc. 344, 60–72 (2003).
Merloni, A., Heinz, S. & Di Matteo, T. Mon. Not. R. Astron. Soc. 345, 1057–1076 (2003).
Heinz, S. & Sunyaev, R. A. Mon. Not. R. Astron. Soc. 343, L59–L64 (2003).
Maccarone, T. J. Mon. Not. R. Astron. Soc. 360, L30–L34 (2005).
Ivanova, N. et al. Astrophys. J. 717, 948–957 (2010).
Mackey, A. D., Wilkinson, M. I., Davies, M. B. & Gilmore, G. F. Mon. Not. R. Astron. Soc. 386, 65–95 (2008).
Kulkarni, S. R., Hut, P. & McMillan, S. Nature 364, 421–423 (1993).
Sigurdsson, S. & Hernquist, L. Nature 364, 423–425 (1993).
Kalogera, V., King, A. R. & Rasio, F. A. Astrophys. J. 601, L171–L174 (2004).
Downing, J. M. B., Benacquista, M. J., Giersz, M. & Spurzem, R. Mon. Not. R. Astron. Soc. 416, 133–147 (2011).
About this article
The Astrophysical Journal (2018)
The Astrophysical Journal (2017)
Proceedings of the International Astronomical Union (2015)