Tight pairs of supermassive black holes are expected to emit gravitational waves that could give astronomers a new way to explore the cosmos. One relatively tight pair has been discovered within a rare triple system. See Letter p.57
Observations have demonstrated that every galaxy with a mass comparable to that of the Milky Way or larger hosts at its centre a supermassive black hole — millions to billions of times more massive than the Sun. These large galaxies are thought to form by mergers of two (or more) smaller galaxies, so the natural expectation is that astronomers should frequently observe pairs of supermassive black holes. Such binary black hole systems are of great interest because they could have a profound influence on the centre of the galaxy, stirring up its gas and ejecting its stars. However, despite many searches1,2, so far we know of only a few binary black hole systems. On page 57 of this issue, Deane et al.3 present the discovery of a rare triple system, and report an improved method for finding tight pairs.
As they coalesce, tight binary supermassive black hole systems emit powerful gravitational waves — ripples in the fabric of space-time that propagate at the speed of light. The detection of these waves would provide additional confirmation of Einstein's general theory of relativity, and would give astrophysicists a new way to explore the cosmos. These waves may be detectable with future space-based instruments4, which would work by bouncing lasers off widely separated spacecraft and looking for subtle changes in the measured distance between them due to a passing gravitational wave. Another detection method under active development is to use ground-based pulsar timing arrays5, such as NANOGrav6. This method takes advantage of the wonderful regularity of pulsating stars by carefully timing the arrival of pulses from many such stars and then looking for tiny changes in the arrival times induced by passing gravitational waves.
Many searches have been carried out for tight binary supermassive black holes. These searches have been made at wavebands from radio to X-rays, because supermassive black holes strongly emit across the electromagnetic spectrum. Radio observations are particularly interesting because the technique of very long baseline interferometry (VLBI) — which creates a giant radio telescope, thousands of kilometres across, from a collection of smaller telescopes — provides the highest spatial resolution achievable. Such high resolution allows the examination of black hole separations down to parsec scales (1 parsec is 3 × 1016 metres), even in the most distant galaxies. The Caltech–Jodrell Bank VLBI survey, initiated in the 1990s, led to the discovery7 of the tightest binary black hole system found so far, which is known to astronomers as 0402 + 379 and has a separation of just 7 pc (Fig. 1). Additional VLBI surveys examined ten times more radio sources than the Caltech–Jodrell Bank VLBI survey, but no further tight binary black hole systems were found.
The crucial issue for finding tight binaries is how quickly the black hole merger takes place. The initial decay (shrinkage) of the orbits of the binary supermassive black holes probably happens quickly thanks to frequent gravitational interactions between the two black holes and stars8. At a black hole separation of under 10 pc, the density of stars drops off, and the binary system could 'stall' before gravitational radiation can begin to carry away angular momentum and allow the orbital decay to resume. If this stalling phase exists, then tight binary black holes should be commonplace. However, if gas or stellar interactions can bridge the gap to sub-parsec scales9 at which gravitational radiation becomes efficient at shrinking the orbits, then tight binary black hole systems could be truly rare.
The triple black hole system discovered by Deane and colleagues, known as J1502 + 1115, was initially identified10 as a quasar (a mass-accreting supermassive black hole system) with double-peaked optical emission lines. Such double-peaked lines have been suggested as a sign of binary black hole systems, in which each of the black holes has its own peak and the two peaks are shifted, with respect to the systematic velocity of the host galaxy, by their orbital velocity. Unfortunately, many single supermassive black hole systems have complicated mass-accretion disks around them that can also produce two (or more) emission-line peaks11.
To confirm a candidate system, other evidence such as multiple compact radio jets is needed. And in J1502 + 1115 there is a bonanza of three such compact jets, with two of them separated by just 140 pc, as Deane et al. now report using VLBI observations. Furthermore, the authors find that, in this system, the large-scale radio emission associated with the tight pair has an 'S'-symmetry suggestive of precession of the jets such as might be expected in a binary system. Such a symmetry could provide a way to search for other tight pairs. The authors convincingly argue against other possible explanations for the observed radio structures (for example, the possibility of a double jet from a single black hole). What is more, they suggest that tight binary black holes are actually more commonplace than previously thought, on the basis of their success in finding one new system out of six candidate double-peaked systems that have moderate radio emission and host-galaxy brightness.
If tight binary black holes actually are long-lived and therefore commonplace, we should not have long to wait for many more of them to be found. Many candidate systems already fit Deane and colleagues' observational selection criteria, and new optical surveys using the Panoramic Survey Telescope and Rapid Response System (Pan-STARRS) and the Large Synoptic Survey Telescope (LSST) will greatly increase the numbers of known active galaxies containing at least one supermassive black hole.
Our best evidence for tight systems relies heavily on high-resolution VLBI observations. However, some care is warranted because jets can exhibit a wide variety of morphologies and spectra, and are thus subject to misinterpretation. The ultimate test for a tight binary black hole system would be to observe the orbital motions of the two components directly. Such a feat would require decades of monitoring with VLBI even for the 0402 + 379 system, which has a predicted orbital period of 150,000 years, but it could be accomplished in just a few years if an even tighter system can be found.
About this article
The Astrophysical Journal (2017)