Since their discovery1 in 1963, astronomical objects called quasars have been among our most powerful probes of the early Universe. Initially seen as mysterious sources of extreme luminosity, quasars are now known to be supermassive black holes that are voraciously consuming gas from their immediate surroundings, emitting large amounts of radiation in the process. In a paper in Nature, Bañados et al.2 report observations of the most distant quasar found so far. The light detected from this object was emitted when the Universe was a mere 690 million years old — just 5% of its current age.
Almost 90 years ago, the astronomer Edwin Hubble discovered that the Universe is expanding3. The expansion stretches light waves travelling through space, such that light that was emitted from a distant source as blue might be detected as red. This phenomenon is called redshift, and is associated with both distance and time: the larger the redshift, the farther away the source was when it emitted its light, meaning that the light was emitted at an earlier time.
If we rewind the expansion, we find that the Universe started out in a hot, dense state, filled mostly with ionized hydrogen. As it expanded, it also cooled, and after about 380,000 years, the temperature was low enough for neutral hydrogen to form. For the first few hundred million years, the Universe was devoid of any sources of light — no stars, galaxies or quasars existed. The first stars were then born, but the Universe remained dark because neutral hydrogen is highly effective at absorbing ultraviolet radiation (the main type of emission from these stars).
However, the present-day Universe is filled with sources of light, and the hydrogen that exists in the space between galaxies (the intergalactic medium) is completely ionized and therefore transparent to the ultraviolet emission from early galaxies and quasars. The process of this phase change from a neutral to an ionized Universe, known as reionization, is poorly understood.
The neutral fraction of hydrogen in the Universe can be estimated by analysing the absorption of light by hydrogen in quasars. Studies of quasars observed as they were when the Universe was 0.85 billion to 1.2 billion years old (corresponding to redshifts of 6.5 to about 5, respectively) have shown that the neutral fraction decreased sharply from 0.1% to 0.01% during this time4. However, most of the reionization process occurred before this epoch.
Bañados and colleagues’ quasar, known as ULAS J1342+0928, has a redshift of 7.54. This means that its strong ultraviolet emission has been shifted into the near-infrared, beyond the sensitivity of typical imaging surveys of the sky. Finding such a high-redshift quasar was not possible until about a decade ago, when sufficiently sensitive near-infrared detectors began scanning large areas of the sky5,6. By studying the absorption spectrum of ULAS J1342+0928 (the fraction of incident radiation absorbed by the intergalactic medium over a range of frequencies), the authors determined that the neutral proportion of hydrogen was at least 10% when the Universe was 690 million years old, which sets a strong constraint on how the intergalactic medium was reionized.
The quasar’s black hole is extremely massive — about 800 million times the mass of the Sun. Black holes grow by consuming (accreting) gas from a surrounding structure called an accretion disk (Fig. 1). The gas emits radiation as it falls in. However, such systems have a maximum luminosity, which occurs when the pressure of the emitted light pushes away the infalling gas, halting further growth. This luminosity depends on the mass of the accreting black hole, and therefore defines a maximum growth rate, known as the Eddington limit, for the system.
Bañados et al. suggest that the large mass of the black hole in ULAS J1342+0928 can be explained if the object began its life as an initial (seed) black hole of at least 1,000 solar masses. This result could rule out models in which black-hole seeds were created from the deaths of the first massive stars7, and instead favour models in which these seeds formed from the direct collapse of primordial gas8. In addition, the black hole would need to have grown continuously (and, therefore, exponentially) at the Eddington limit, starting from when the Universe was roughly 65 million years old. Although this scenario is physically possible, it requires extreme, sustained accretion for about 600 million years, which is substantially longer than the typical lifetime of a quasar9.
So far, only two quasars with redshifts greater than 7 have been discovered. The previous record holder was reported10 in 2011, and early models of quasar evolution predicted that more should have been found by now11. The methods for finding quasars, even at these high redshifts, are sound and have been proved effective. Therefore, the dearth of high-redshift quasars might indicate that these objects were uncommon in the early Universe, and could imply a sharp decline in quasar activity towards early times12. If so, this suggests that we might be observing extremely rare systems as they were beginning to emerge in the Universe.The authors’ work offers a glimpse into the conditions of the intergalactic medium at the earliest epoch of structure formation in the Universe, and could place key constraints on cosmological models of this era. However, a single quasar is insufficient for providing a complete picture of the Universe in the reionization era or of the evolution and growth of supermassive black holes from initial seeds. The task ahead is, then, to mine the upcoming near-infrared sky surveys for additional quasars that can paint a more complete picture of the rapidly evolving early Universe.
Nature 553, 410-411 (2018)