The power of a cosmic lens to magnify and split the light from a distant, mass-accreting giant black hole into four components has allowed researchers to measure the black hole's spin. See Letter p.207
Quasars are the most powerful, continuously emitting sources of radiation in the Universe. They reside at the centre of a small fraction of galaxies, and are powered by supermassive black holes, which have masses millions to billions times greater than that of the Sun. Although giant black holes are present in most — possibly all — galaxies, not all of them are in an active state, in which they accrete gas from a surrounding disk. In fact, most of these objects are in a quiescent phase. It is the active type of supermassive black hole that drives quasars. The formation history of supermassive black holes is thought to be closely tied to that of their host galaxies, but how exactly they form and grow remains unclear. In this issue, Reis et al.1 (page 207) describe how a cosmic lens has enabled them to find that a supermassive black hole powering a distant quasar has grown through coherent, rather than random, episodes of mass accretion.
Astronomers believe that supermassive black holes formed in the early Universe from small 'seeds' with masses of up to 10,000 solar masses. These seeds would have then grown to reach millions to billions of solar masses either through multiple mergers during galaxy collisions or through gas accretion from their host galaxies; this accretion would have consisted either of many short, unrelated accretion episodes or of fewer, longer and ordered accretion phases. Different models of galaxy evolution predict a different mix of these processes, so reconstructing the formation history of giant black holes would provide a way for us to understand how galaxies evolved.
Supermassive black holes are simple systems. They are characterized by just two quantities, their mass and their angular momentum (spin). Whereas the total amount of accretion and any mergers that a supermassive black hole undergoes are encoded in its mass, how this mass was assembled is encoded in its spin2. A few ordered accretion events or mergers of large black holes produce high spins, and short, random accretion processes produce low spins. Measuring these spins is therefore a major goal of extragalactic astronomy: the spins of supermassive black holes hold a key to understanding the evolution of their host galaxies.
But how can we measure the spins? According to Einstein's general theory of relativity, a black hole's gravitational field twists space-time around it. Such twisting depends on the black hole's spin, so measuring the twisting allows the spin to be estimated. The signature of space-time distortion is imprinted on the emission of radiation from regions close to the black hole's event horizon — the surface beyond which no radiation can escape. In quasars, the bulk of the huge, observed luminosity is emitted by the accretion disk at optical and ultraviolet wavelengths. However, this primary emission is nearly featureless, so, despite its vicinity to the event horizon, it does not provide an easy means to detect space-time distortions. The best way to perform such a measurement is to observe X-rays reflected by the disk.
The main source of X-ray emission in quasars is believed to be a compact cloud of hot electrons in the inner part of the black hole's accretion disk. Some of this radiation illuminates the accretion disk and is reflected towards the observer's line of sight. This reflected emission usually accounts for less than 1% of the total energy produced by quasars, but contains narrow spectral features — most notably, an iron spectral line at the object's rest-frame energy of 6.4–7 kiloelectronvolts — the shape of which is strongly altered by the space-time warping around the black hole3,4,5. The shape of these features can be measured in high-quality X-ray spectra, providing a measurement of the spins of supermassive black holes6,7,8.
This type of analysis is at the heart of Reis and colleagues' work. Until now, astronomers have struggled to obtain unambiguous spin measurements using this method. The X-ray spectra of active galactic nuclei are quite complex, and the reflection component, which contains the signatures of space-time distortions, is relatively weak. Moreover, certain absorption features can mimic the distortions9. As a result, only long observations of a few very bright sources in the local Universe made with the most powerful X-ray observatories — NASA's Chandra, Europe's XMM-Newton, Japan's Suzaku and, more recently, NASA's NuSTAR — have provided convincing results6,7,8. In their study, Reis et al. break new ground by obtaining a spin measurement of a quasar at a distance of more than 6 billion light years from Earth, from a time when the Universe was about half its current age.
This remarkable result was possible owing to the exceptional nature of the observed source — a quadruply imaged, gravitationally lensed quasar (Fig. 1). The light from the distant quasar is both magnified and split into four different images by the gravitational field of a foreground elliptical galaxy (the lens) that, by chance, is on the line of sight of the quasar. For this reason, the authors could analyse four 'copies' of the X-ray spectrum of the quasar, each with an intensity significantly magnified by the lens. The resulting X-ray spectra have a quality that matches the best that has been obtained for nearby sources, and allowed a robust measurement of the black hole's spin. As it turns out, the spin is large (close to the highest possible value that theory predicts), suggesting that the black hole acquired its mass through coherent phases of mass accretion.
Although X-ray spectra of a quality comparable to that obtained here cannot be currently obtained for standard, non-lensed sources at similar distances, Reis and colleagues have opened a window on what astronomers could observe with the next generation of X-ray telescopes, such as Europe's proposed ATHENA mission.
Reis, R. C., Reynolds, M. T., Miller, J. M. & Walton, D. J. Nature 507, 207–209 (2014).
Volonteri, M., Madau, P., Quataert, E. & Rees, M. J. Astrophys. J. 620, 69–77 (2005).
Fabian, A. C., Rees, M. J., Stella, L. & White, N. E. Mon. Not. R. Astron. Soc. 238, 729–736 (1989).
Dauser, T., Wilms, J., Reynolds, C. S. & Brenneman, L. W. Mon. Not. R. Astron. Soc. 409, 1534–1540 (2010).
Dovčiak, M., Karas, V. & Yaqoob, T. Astrophys. J. Suppl. 153, 205–221 (2004).
Brenneman, L. W. & Reynolds, C. S. Astrophys. J. 652, 1028–1043 (2006).
Walton, D. J. et al. Mon. Not. R. Astron. Soc. 428, 2901–2920 (2013).
Risaliti, G. et al. Nature 452, 449–451 (2013).
Miller, L., Turner, T. J. & Reeves, J. N. Astron. Astrophys. 483, 437–452 (2008).
Claeskens, J.-F., Sluse, D., Riaud, P. & Surdej, J. Astron. Astrophys. 451, 865–879 (2006).