Many questions remain about what happened in the first billion years of the universe. At around 100 million years old, the universe was a dark place consisting of mostly neutral hydrogen without many objects emitting detectable radiation. This situation changed as stars and galaxies formed, leading to a phase transition known as reionisation where the neutral hydrogen was ionised. Exactly when reionisation started and how long it took is still not fully clear, but a recent discovery of the oldest massive black hole ever found can help answer this important question.

Up to about 300,000 years after the Big Bang, the universe was hot and dense, and electrons and protons were fully separated. As the universe started to expand, it cooled down and underwent a first phase transition where electrons and protons formed neutral gases such as hydrogen. The following period is known as the cosmic dark ages. During this period, protons and electrons were mostly combined into neutral hydrogen, but the universe had to cool much further before matter could condense to the level where light-producing objects such as stars could form. These new objects started to emit both the radiation we can now detect to study the early universe and also the radiation responsible for the last phase transition – the reionisation of the universe. Some of the brightest and therefore easiest-to-detect objects are quasars: massive black holes surrounded by discs of hot accreting matter that emit radiation over a wide but distinctive spectrum.

Using data from a range of large-area surveys by different telescopes, a group led by Eduardo Bañados from the Carnegie Institution for Science has discovered a distant quasar called J1342+0928, with the black hole at its centre found to be eight million solar masses. After the radiation was emitted by J1342+0928, it travelled through the expanding universe, increasing its wavelength or “red shifting” in proportion to its travel time. Using known spectral features of quasars, the redshift (and therefore the moment at which the radiation was emitted) can be calculated.

The spectrum of J1342+0928, shown in the figure, demonstrates that the universe was only 690 million years old – just 5% of its current age – at the time we see J1342+0928. The spectrum also shows a second interesting feature: the absorption of a part of the spectrum by neutral hydrogen, which implies that at the time we are observing the black hole, the universe was not fully ionised yet. By modelling the emission and absorption, Bañados and co-workers found that the spectrum from J1342+0928 is compatible with emission in a universe where half the hydrogen was ionised, putting the time of emission right in the middle of the epoch of reionisation.

The next mystery is to explain how a black hole weighing eight million solar masses could form so early in the universe. Black holes grow as they accrete mass surrounding them, but the accreting mass radiates and this radiation pushes other accreting mass away from the black hole. As a result, there is a theoretical limit on the amount of matter a black hole can accrete. Forming a black hole the size of J1342+0928 with such accretion limits would require black holes in the very early universe with sizes that challenge current theoretical models. One possible explanation, however, is that this particular black hole is a peculiar case and was formed by a merger of several smaller black holes.

Thanks to continuous data taking from a range of existing telescopes and upcoming new instrumentation, we can expect more objects like J1342+0928 or even older to be discovered, offering a probe of the universe at even earlier stages. The discovery of further objects would allow a more exact date for the period of reionisation, which can be compared with indirect measurements coming from the cosmic microwave background. At the same time, more measurements will show if black holes of this size in the early universe are just an anomaly or if there are more. In either case, such observations would provide important input for research on early black hole formation.

Further reading

E Bañados et al. 2018 Nature 553 473.