A black hole in a globular cluster
22 Jan, 2007 07:13 pm
This report discusses the recent discovery of strong evidence for an accreting black hole in a globular cluster in the galaxy NGC 4472, the brightest elliptical galaxy in the Virgo Cluster. It also places the discovery in context of why the topic of black holes in globular clusters has attracted so much interest in the astronomical community.
The discovery of X-ray emission from globular clusters triggered a debate about whether globular clusters might host massive black holes (i.e. black holes heavier than the ten solar mass objects we routinely observe in binary systems, but lighter than the million to billion solar mass black holes in the cores of galaxies)[4]. Since that time astronomers have found definitive proof that all the Milky Way's X-ray emitting globular clusters contain accreting neutron stars - all show strong bursts of radiation characteristic of thermonuclear fusion of material which has piled up on the solid surface of the neutron star[5]. Furthermore, a variety of theoretical work shown that "stellar mass black holes" (i.e. black holes formed in supernova explosions of massive stars) should sink to the centers of globular clusters (as they will be the heaviest objects in the clusters), should then pair up into binaries, and then should eject one another by using the gravitational potential energy of the binary systems to provide the kick needed to escape the gravity of the cluster[6,7,8].
It has never been clear, though, whether these results apply to all clusters, or just to those with certain masses and densities[6]. Furthermore, the debate over whether black holes exist in globular clusters has been re-kindled by theoretical suggestions that stellar mass black holes might merge into massive black holes in globular clusters[9], and by findings that the trajectories of stars in a few globular clusters show strong evidence of a large mass in a small volume in the inner core, without any strong excess in the amount of starlight from these regions[10,11]. Astronomers have vigorously debated whether these mass concentrations imply the presence of central black holes about 1000 times as massive as the Sun[8,9], or simply verify the belief that the heaviest objects in a globular cluster should sink to its center[12,13]. Since in an old population of stars like a globular cluster, the heaviest objects will be neutron stars and white dwarfs, which are very faint in optical light, this, too, would lead to a dark, but massive concentration in the cluster center[12,13].
In the new era of X-ray satellites, astronomers finally have the capability to make detailed and sensitive X-ray images of other galaxies, allowing them to look at hundreds of times as many globular clusters as our Milky Way contains. The downside of this capability is that it is normally impossible to determine whether the emission from a distant globular cluster comes from a single X-ray source, or many X-ray sources. The simplest means of proving an object cannot be a neutron star is to show that it is brighter than neutron stars are believed to get, a technique which obviously does not work when one could have many accreting objects in the same cluster, which cannot be resolved apart by the telescope.
In a recent Letter to Nature[14], we had the good fortune to report the discovery of strong variability from an extremely bright X-ray source in a globular cluster in the galaxy NGC 4472, the brightest galaxy in the Virgo Cluster of galaxies, about 55 million light years away. The strong variability provides proof that the X-ray emission is dominated by a single source, since it would be highly unlikely for a collection of X-ray sources all to turn off at the same time. We furthermore had the benefit of spectroscopic verification that the object we believe is a globular cluster really is. This solves the other key problem with providing proof for the existence of black holes in globular clusters - namely that background quasars (supermassive black holes accreting at high rates) can look quite similar to globular clusters in images, and can be sources of strong, highly variable X-ray emission.
The implications of this result are still being sorted out. We believe that the present data favor the interpretation that this black hole is about ten times the mass of the Sun. The variability seems to be due to an outflow from the accretion disk around the black hole, which would most likely be powered because more matter is being fed into the black hole than it can accrete, and a much heavier black hole would be able to accrete more than enough matter to produce the observed luminosity (about a million times the Sun's luminosity). However, this line of reasoning is far from concrete, and it certainly remains possible that this black hole is hundreds or thousands of times the mass of the Sun; more observations will be necessary to answer this question. Regardless, the first clear evidence that globular clusters can contain black holes is an important step on the road to determining whether they can be factories for making the intermediate mass black holes that may provide the missing link between the stellar mass black holes in binary systems, and the supermassive black holes in the nuclei of galaxies.
1. Clark, G.W., 1975, ApJL, 195, 43
2. Katz, J.I., 1975, Nature, 253, 698
3. Kundu, A., Maccarone, T.J., & Zepf, S.E., 2002, ApJL, 574, 5
4. Bahcall J.N. & Ostriker, J.P., 1975, Nature, 256, 23
5. Liu, Q.Z., van Paradijs, J., & van den Heuvel, E.P.J, 2001, A&A, 368, 1021 (see also references within)
6. Kulkarni, S., McMillan S.F.W., Hut, P., 1993, Nature, 364, 421
7. Sigurdsson, S. & Hernquist, L., 1993, Nature, 364, 423
8. Portegies Zwart S.F., & McMillan, S.F.W., 2000, ApJL, 528, 17
9. Miller, M.C. & Hamilton, D.P., 2002, MNRAS, 330, 232
10. Gerssen et al., 2002, AJ, 124, 3270
11. Gebhardt, K., Rich, R.M., & Ho, L.C., 2002, ApJL, 578, 41
12. Baumgardt et al., 2003, ApJL, 583, 21
13. Baumgardt et al., 2003, ApJL, 589, 25
14. Maccarone, T.J., Kundu, A., Zepf, S.E., & Rhode, K.L., 2007, Nature, 445, 183
-
24/01/10
The Known Universe
-
11/09/08
Large Hadron Rap
-
02/04/08
Meteorites Rich in Amino Acids
-
01/04/08
New Organic Molecule in Space
-
06/12/07
LHC: The Six Billion Dollar Questions
Overall, I found the article to be interesting and well-written, but I would suggest that the author REVISE the article to include a discussion of the theoretical implications of his observations. The author discusses mass segregation; the idea that heavier objects like stellar mass black holes will settle to the center of globular clusters of certain masses and densities. How does the discovery of this flaring 10 solar mass black hole constrain the regimes in which mass segregation occurs? How does if affect the expected abundances of 10 solar mass black holes at the core of globular clusters, and the rates at which these would catalyze the formation of intermediate (1000 solar mass) black holes?
I would also revise the final sentence of the article. While the existence of solar mass black holes may be neccessary to form intermediate mass black holes, does this observation or future observations of this kind have the potential to tell us more about the population of solar mass black holes? Also, in what sense are intermediate mass black holes "missing links" between solar mass black holes in binary systems and supermassive black holes in galaxies?
These are certainly interesting questions, and I would be very interested to learn more about the ability of observations of this kind to address them.
This discovery doesn't really constrain how quickly mass segregation occurs, primarily because at the present time, our data isn't good enough to give us any structural parameters for the cluster. But the main open question is really more about whether mass segregation leads to "self-ejection" of black holes.
And with only one object, we don't gain any quantitative information about the abundances of 10 solar mass black holes, or the possible implications for formation of intermediate mass black holes in clusters. We just learn that neither can be called impossible on the grounds that black holes are all ejected by stellar dynamical processes. To say more will require a lot more detections, and a better understanding of what fraction of black holes are actually detectable. It has been suggested, for example, by Kalogera, King & Rasio (2004, ApJL, 601, 171), that most black holes in globular clusters would be in wide binary systems where the transfer rate of mass from the companion star is very low, and hence the systems would not be X-ray emitters for a large fraction of their lifetimes.
The idea that intermediate mass black holes are missing links comes from some observations indicating that it might be necessary to make black holes in the early universe grow faster than they seem to be able to grow by accreting gas - there is a three billion solar mass black hole seen at high redshift - z=6.41 (Willot, McLure & Jarvis, 2003, ApJL, 587, 15). If the black hole mass measurement there is accurate, either the black hole would have had to have started out significantly heavier than 10 solar masses, or it would have required an unusual means of growing. There certainly are alternatives for forming heavier black holes besides just mergers in the cores of a dense star cluster, but that is one of the leading possibilities.