A 17-billion-solar-mass black hole in a group galaxy with a diffuse core (original) (raw)
Quasars are associated with and powered by the accretion of material onto massive black holes; the detection of highly luminous quasars with redshifts greater than z = 6 suggests that black holes of up to ten billion solar masses already existed 13 billion years ago 1. Two possible present-day 'dormant' descendants of this population of 'active' black holes have been found 2 in the galaxies NGC 3842 and NGC 4889 at the centres of the Leo and Coma galaxy clusters, which together form the central region of the Great Wall 3-the largest local structure of galaxies. The most luminous quasars, however, are not confined to such high-density regions of the early Universe 4,5 ; yet dormant black holes of this high mass have not yet been found outside of modern-day rich clusters. Here we report observations of the stellar velocity distribution in the galaxy NGC 1600-a relatively isolated elliptical galaxy near the centre of a galaxy group at a distance of 64 megaparsecs from Earth. We use orbit superposition models to determine that the black hole at the centre of NGC 1600 has a mass of 17 billion solar masses. The spatial distribution of stars near the centre of NGC 1600 is rather diffuse. We find that the region of depleted stellar density in the cores of massive elliptical galaxies extends over the same radius as the gravitational sphere of influence of the central black holes, and interpret this as the dynamical imprint of the black holes. We observed NGC 1600 (Fig. 1) as part of the MASSIVE Survey 6 , the aim of which is to study the structure, dynamics, and formation history of the 100 most massive early-type galaxies within 108 megaparsecs (Mpc) of Earth. This volume-limited survey probes galaxies with stellar masses above 5 × 10 11 M ((where M (is the mass of the Sun) in diverse, large-scale environments that have not been systematically studied before. The stellar mass (8.3 × 10 11 M (), halo mass (∼ 1.5 × 10 14 M (), and distance (64 Mpc) of NGC 1600 are fairly typical of the galaxies in the survey. We obtained stellar spectra covering the central 5-arcsec-by-7-arcsec region of NGC 1600 with roughly 0.6-arcsec spatial resolution, using the Gemini multi-object spectrograph (GMOS) at the Gemini North Telescope. We further obtained large-area (107-arcsec-by-107arcsec) stellar spectra of NGC 1600 using the Mitchell integral field spectrograph (IFS) at the McDonald Observatory. The stellar luminosity distribution of the galaxy is provided by surface photometry from the Hubble Space Telescope (HST) and the Kitt Peak National Observatory 7. We measured the distribution of the line-of-sight stellar velocities at 86 locations in NGC 1600 by modelling the deep calcium triplet absorption lines in our GMOS IFS spectra and several optical absorption features in our Mitchell IFS spectra. The galaxy shows little rotation (less than 30 km s −1), and the line-of-sight velocity dispersion rises from 235-275 km s −1 at large radii to a maximum value of 359 km s −1 near the centre, consistent with previous long-slit measurements 8. We used orbit superposition models 9 to determine the mass of the central black hole, M BH , of NGC 1600: we find a value for M BH of (1.7 ± 0.15) × 10 10 M ((68% confidence interval). We rule out statistically the possibility that the M BH is less than 10 10 M (with more than 99.9% confidence (see Methods). Defining the sphere of influence of the black hole, r SOI , as the radius at which the enclosed stellar mass equals M BH , we find r SOI = 3.8 arcsec (or 1.2 kiloparsec, kpc) for NGC 1600, using our measured M BH and our calculated value for M /L of (4.0 ± 0.15)M (/L ((in the R-band; L (is the solar luminosity). Our velocity data resolve the central spatial region of NGC 1600 down to about 200 pc, where M BH exceeds the enclosed stellar mass by a factor of 100. Even if the unresolved stellar mass near the centre were ten times larger because of, for example, an extreme population of undetected dwarf stars, this would not have a measurable effect on M BH (see Methods). The stellar bulge mass of NGC 1600 is 8.3 × 10 11 M (, according to our dynamical modelling, and is consistent with the value inferred from the absolute K-band magnitude of − 25.99 for NGC 1600 (ref. 6). The M BH of NGC 1600 is 2.1 per cent of its bulge mass (M bulge)-three to nine times more than the percentage predicted from the known scaling relations of black-hole and galaxy bulge mass 10-12. Other galaxies with high M BH-to-M bulge ratios are all notable for their compact size, suggesting that they are tidally stripped objects or relics from the young Universe with stunted late-time growth. In contrast, NGC 1600 is much less compact, as judged by its half-light radius, which is comparable to that of other giant elliptical galaxies of similar luminosity. The M BH of NGC 1600 is also about ten times larger than the mass expected from its average stellar velocity dispersion (Extended Data Fig. 1). NGC 1600 has a remarkably faint and flat core: it has the largest core radius detected in a study of 219 early-type galaxies using HST