Astronomical Science DOI: 10.18727/0722-6691/5039 ALMA Observations of z ~ 7 Quasar Hosts: Massive Galaxies in Formation Bram P. Venemans1 An effective way to learn more about To further constrain the build up of mas- the constituents of galaxies at the highest sive galaxies, it is important to locate redshifts is to study the brightest (and and study bright quasars at even higher 1 Max Planck Institute for Astronomy, most massive) members of this popula- redshifts. Over the last ten years, with the Heidelberg, Germany tion. Such luminous galaxies are very rare advent of wide-field near-infrared cam- and not found in the deep, pencil-beam eras, large near-infrared surveys started studies that are typically used for high- to image the sky to sufficient depth to Luminous high-redshift quasars are redshift galaxy searches with, for exam- uncover distant quasars. Using various thought to be hosted by the most mas- ple, the Hubble Space Telescope (HST). public surveys, such as the UKIRT Infra- sive and luminous galaxies in the early red Deep Sky Survey (UKIDSS), the Visi- Universe. Over the past few years, we Quasars are among the most luminous ble and Infrared Survey Telescope for have discovered several quasars at non-transient sources in the Universe. Astronomy (VISTA) Kilo-degree INfrared 9 z ~ 7 powered by > 10 M⊙ black holes, As a result they can be seen out to very Galaxy survey (VIKING), the VISTA Hemi- which allow us to study the formation high redshift, z > 7. Quasars are powered sphere Survey (VHS), and the Panoramic and evolution of massive galaxies at the by (supermassive) black holes that Survey Telescope And Rapid Response highest redshifts. ALMA and PdBI/ accrete matter from their environment. System survey (Pan-STARRS1), we have NOEMA observations have revealed The host galaxies of accreting super- discovered more than a dozen luminous that these z ~ 7 quasars are hosted by massive black holes at z > 2 are among quasars with redshifts above z = 6.5 (for far-infrared-bright galaxies with far- the brightest and most massive galaxies example, Mortlock et al., 2011; Venemans 12 infrared luminosities > 10 L⊙, indi- found at these redshifts (for example, et al., 2013, 2015). cating star formation rates between Seymour et al., 2007). Therefore, an –1 100 and 1600 M⊙ year . High-resolution effective method to pinpoint the most When compared to the quasars found ALMA imaging of a quasar host at massive and luminous galaxies in the around z ~ 6, the absolute magnitudes, 9 z = 7.1 shows that a high fraction of both early Universe is to locate bright quasars black hole masses (of ≥ 10 M⊙) and the dust continuum and [C II] 158 μm at the highest redshifts. (metal) emission line strengths of these emission comes from a compact region new z ~ 7 quasars are remarkably similar < 2 square kiloparsecs across. Obser- A crucial initial step is to find such bright (for example, De Rosa et al., 2014), indi- vations of emission from CO and neu- quasars at large cosmological distances. cating little evolution in the quasar prop- tral carbon in our z ~ 7 hosts provide This is extremely difficult as they are erties over the 200 Myr between z = 6 the first constraints on the properties of very rare (only one luminous quasar is and z = 7.1 (~ 25 % of the Universe’s age the interstellar medium and suggest expected at z ~ 6 per > 100 square at that time). Furthermore, not only are that the gas heating is dominated by degrees on the sky) and requires large these z > 6.5 quasars among the most star formation. optical/near-infrared surveys that cover distant sources found today, but they a large area. Indeed, the Sloan Digital also have luminosities that are about two Sky Survey discovered about 20 bright orders of magnitude larger than those of Distant quasars: beacons in the early quasars around z ~ 6 (less than 1 Gyr galaxies found at similar redshifts. These Universe after the Big Bang), which are shown to z > 6.5 quasars therefore provide a good 9 host supermassive black holes (> 10 M⊙; opportunity to gain crucial insight into the Among the outstanding important ques- for example, Jiang et al., 2007; De Rosa formation and evolution of massive galax- tions in astronomy are when the first et al., 2011). Direct imaging of the stellar ies at the highest redshifts, and to under- galaxies formed, and what their physical light of the galaxies hosting these high stand the origin of the apparent correla- properties were. Galaxy candidates have redshift quasars via rest-frame ultraviolet/ tion between bulge mass and black hole now been identified through deep optical/ optical studies has proven very difficult, mass found in nearby galaxies (for exam- near-infrared imaging surveys out to red- if not impossible. On the other hand, ple, Kormendy & Ho, 2013). With the shifts of z > 10, only ~ 450 million years observations of the molecular gas in the completion of the Atacama Large Millime- after the Big Bang (see, for example, host galaxies in the radio and (sub)- ter/submillimeter Array (ALMA) we now the recent review by Stark, 2016). How- millimetre (targeting the rest-frame far- have a state-of-the-art facility to study the ever, their faintness typically prohibits infrared emission) allow the determination host galaxies of distant quasars in detail. detailed studies of their nature and char- of the total gas mass and dynamical acteristics, and, often even their spectro- mass of the hosts. In the last decade, it scopic confirmation. Indeed, studying has been demonstrated that large reser- The host galaxies of z ~ 7 quasars the detailed physical properties of these voirs of metal-enriched atomic and objects and their contribution to cosmic molecular gas and dust can exist in mas- To extend the study of quasar host galax- reionisation is one of the main drivers sive quasar host galaxies up to z ~ 6.4 ies to z ~ 7, we initiated a programme tar- for the James Webb Space Telescope (see Carilli & Walter, 2013 for a review). geting all quasars discovered at z > 6.5 (JWST) and the next generation of large These observations already provide at mm wavelengths, independent of their optical telescopes (in particular the tight constraints on models of galaxy for- far-infrared brightness, with the aim of ESO Extremely Large Telescope, ELT). mation and evolution (for example, Kuo & sampling the range of properties of qua- Hirashita, 2012; Valiante et al., 2014). sar host galaxies and investigating the 48 The Messenger 169 – September 2017 Figure 1. Compilation of [C II] 158 μm observations ) of five quasars at z > 6.5. The top four spectra are Y :"((< from our ALMA Cycle 1 programme (Venemans et al., 2016, 2017a) and the bottom spectrum is from PdBI (Bañados et al., 2015). ) Y :"((< variations, all the quasar hosts have far- infrared luminosities of L ≥ 1012 L and FIR ⊙ can be classified as ultraluminous infra- red galaxies (ULIRGs), in contrast to the AD@L ) vast majority of normal galaxies known X Y:"((< at these redshifts. The high infrared lumi- L) nosities indicate that stars formed in SX these quasar hosts at rates of at least DMRH ~ 100 M y r –1 and up to ~ 1600 M yr –1 C ⊙ ⊙ (Venemans et al., 2016). Alternatively, %KTW using the [C II] luminosity to estimate ) the star formation rates in the quasar Y:"((< hosts results in very similar values. The far- infrared detections further imply that significant amounts of dust, MDust ≥ 10 8 M , have already formed at z ~ 7, only ⊙ 750 Myr after the Big Bang. Such high / dust mass requires a very efficient dust Y :"((< production and/or a high stellar mass in the host galaxy (see, for example, Gall et al., 2011). l From the detection of the [C II] emission .ARDQUDCEQDPTDMBX&'Y line, we can start to constrain the dynam- ical masses of the quasar host galaxies. If we assume that the gas is located in relationship between star formation and The first object in our project was the a rotating, thin disc, we can compute the supermassive black hole growth at z ~ 7. quasar J1120+0641, at z = 7.1 the dynamical masses of the hosts from Early studies of the host galaxies of z ~ 6 highest redshift quasar known at that the observed width and spatial extent of quasars, by their nature, concentrated time (Venemans et al., 2012; Figure 1). the [C II] emission. We estimate that the on the far-infrared-bright quasars, and The host galaxy of this quasar was found dynamical masses of the host galaxies 10 11 the results from these studies may have to display somewhat different character- at z > 6.5 are 10 −10 M⊙. If we com- introduced a biased view of the charac- istics in its far-infrared (FIR) properties; the pare the host galaxy mass to that of the teristics of the typical galaxy hosting far-infrared luminosity of LFIR = (6−18) × central black hole, we find a ratio that is 11 a luminous quasar in the early Universe. 10 L⊙ is relatively faint compared to higher by a factor 3–4 than found locally Using ALMA and the Institut de Radioas- well-studied z ~ 6 quasar hosts (which (Figure 2; Venemans et al., 2016). 12 tronomie Millimétrique (IRAM) Plateau have LFIR ≥ 5 × 10 L⊙; for example, de Bure Interferometer (PdBI), we imaged Wang et al., 2013), and suggests a star We find that the ratio of black hole mass –1 five z > 6.5 quasars, targeting the [C II] formation rate of “only” 100–350 M⊙ y r .