When an alignment of supermassive black holes occurs

The rotation axes of several quasars, those active nuclei of galaxies hosting a supermassive black hole at their core, are aligned with each other but also with the axis of the cosmic web filament to which they belong. This group may constitute a complex structure formed at about one third of the current age of the Universe, that extends over a distance of three billion light years. This discovery was made thanks to the analysis of the polarisation of the light emitted by those quasars. The observations were carried out in Chile by astrophysicists from the University of Liege. Though the alignment phenomenon was expected by the researchers, the detection of such a large structure remains very surprising since it is not predicted by the standard model of the Universe.

In March 2014, researchers from the Department of Astrophysics, Geophysics and Oceanography of the University of Liege traveled to the ESO observatory in Chile to observe the cosmos through one of the VLTs (Very Large Telescopes), which are among the largest telescopes in the world. They knew precisely where to look. They concentrated their attention towards a part of the sky containing a high number of quasars. These are the active nuclei of galaxies which contain supermassive black holes around which the significant accretion of matter generates so much energy that they are brighter than all the stars of their host galaxy put together. They are so bright that, despite being several billion light years away, they are visible from the Earth. “Of course”, explains Damien Hutsemékers, senior research associate with the FNRS in the Department of Astrophysics, Geophysics and Oceanography (AGO) of the University of Liege, and the first author of the publication(1), “what we see are mere point-like sources like when observing the stars Their name was derived from that property ; quasar stands for quasi-stellar radio source. In the sixties, spectroscopy allowed to establish that these sources of light were not located within our galaxy but much further away”.

Since then, the quest to catalog and better understand these objects at the edge of the Universe has continued to grow. In the last few years, large sky surveys have discovered lots of these objects. From just a few thousand objects ten years ago, the list of known quasars today exceeds one million, a number which makes these objects a robust and interesting statistical sample for different reasons. “Their distance and brightness are good assets”, explains Lorraine Braibant, a doctoral student at the AGO and co-author of the publication. “We use them as beacons for sketching the distant and therefore very young Universe. For example, the group we have just studied has enabled us to go back ten billion years in time, at a time when the Universe was about one third of its current age. We can also study quasars for themselves”. When studied for themselves they are also objects of great curiosity. Not only does their core consist of a supermassive black hole, a very mysterious cosmic object, but they are also among the most massive objects of the Universe and seem able to organize themselves into structures as illustrated by this study. Studying quasars sheds some light on a poorly understood aspect of the standard cosmological model. We understand quite well the behaviour of what surrounds us but, on a bigger scale, that is to say on scales of billions of light years in terms of time and space, our knowledge of the Universe remains theoretical and full of extrapolations that need to be validated through observation.

Damien Hutsemékers and his team have focussed their interest on nearly one hundred quasars. They could have delegated the responsibility for data gathering to the astronomers at of the observatory in Chile. “But we preferred to go there ourselves”, explains Lorraine Braibant. “There was a risk that the weather conditions might not be good when we arrived. But in our case we did not require optimal atmospheric conditions. Finally, we were very lucky with the weather. We wanted to point the telescope at 70 quasars which we succeeded in doing quite quickly. We had some remaining observation time to use.  We were therefore able to observe another sub-group of quasars and increase the number of quasars observed to 93”.   

The prediction of structures in our cosmological model

In theory, numerical models predict the existence of cosmic webs, structures existing between different objects of the cosmos. These structures result from the fact that the Universe, on a smaller scale, is not homogenous. Objects are not uniformly distributed throughout space. Over the course of billions of years, they groupe and create denser volumes in the distribution of matter in the Universe. On a very large scale, the structure of the Universe appears like a web composed of interconnected filaments. These filaments mark the boundaries of “voids”, where the density of galaxies or clusters of galaxies is smaller than in the filaments. Up to a certain scale, observations confirm and support this theory. “This is the limit of our models”, explains Vincent Pelgrims, a doctoral student at AGO and coauthor of the publication. “They predict that galaxies can organize themselves into filaments that look like neuronal connections but are much bigger of course. “These cosmic webs depend on gravitational force”, continues Dominique Sluse, a researcher at the Argelander Institute of Astronomy in Bonn and the University of Liege. “The models show that, under the influence of gravity, the angular momentum, therefore the rotational axes of galaxies, are linked to the orientation of the structure to which they belong”.

The standard model predicts structures of size up to 350 megaparsecs; a parsec is a unit of measure corresponding to 3.26 light years. “In the case of structures of galaxies whose scale is about 100 megaparsecs, and on scales that are smaller and nearer to us, the Universe is clearly not homogeneous”, explains Damien Hutsemékers. “But on larger scales, we consider that it has to be homogeneous . In particular, calculations show that the Universe had not enough time to form bigger structures. In summary, according to the standard model, a structure much bigger than 350 megaparsecs cannot exist”.

The suspected existence of a quasar structure

In this context, the observation of these 93 quasars has become interesting. Such a high density of these objects in the same region of the sky raised suspicions. “The recent discovery of this group was already surprising”, explains Dominique Sluse. “Observing this overdensity of quasars was a first step which allowed astronomers to speculate that it was probably a structure that was bigger than one gigaparsec, that is to say, more than 3.26 billion light years, at a time when the Universe was only 5 billion years old. This is ten times bigger than the structures that are usually observed and three times bigger than the predicted limit of the standard model. But was this the result of chance or not? Was this really a single individual structure or were we missing something? It was difficult to say”.    

In order to establish the facts once and for all, it was necessary to pay more attention to this structure. The answer was to be found in the emission of light from the accretion disc. As a reminder, quasars are too far for their internal structures to be directly observed. Only the light they emit can be studied. Another important point to remember is that the orientation of the rotation axis of an object in the Universe without the influence of another force is completely random. Statistically, the alignment of the rotation axes of different objects therefore has little chance of being random, and is rather an indication that they belong to the same structure.  By being able to determine the orientation of the rotation axes of the quasars observed, it now becomes possible to verify whether these axes are aligned or not. In the affirmative case, the theory of the presence of a structure is reinforced. 

The polarisation of the light reveals an alignment

The orientation of the axis of the accretion disc can be determined thanks to the observation and analysis of the polarisation of light. Usually, light is not polarised. It is a transversal wave which vibrates and propagates in all directions in a homogeneous way from its source and in all vibration planes. Light is polarised when the oscillation of the wave preferentially occurs in one direction. “In normal circumstances”, explains Damien Hutsemékers, “the reflection of light on a puddle of water, on a wind-screen or a mirror constitute phenomena that produce polarisation. In the case of quasars, a similar mechanism is at work: a source of light which is assumed not to be polarised, is reflected by particles such as electrons or dust. This reflection causes polarisation”. Vincent Pelgrims continues, “The angle of polarisation indicates the direction in which the electric field of the photons that reach us is the strongest. We have one a direction and therefore an axis which is related to that of the accretion disc”.  

For the 93 quasars observed, 19 candidates emitted a sufficient quantity of polarised light to enable them to be studied. “With Vincent and Dominique”, continues Damien Hutsemékers, “we have looked at how the angles of polarisation organize themselves on the structure. At first, we observed that they were aligned despite the fact that they are separated by billions of light years. We then went further and we noticed that they also tended to be aligned with the axis of the filamentary structure in which they were found. This was surprising, even though we somehow expected it. This was what we had been trying to observe”.

The predictions of the standard model therefore seemed to have been exceeded. “The information about the polarisation backed up the idea that we were witnessing something unique and whose constituent parts were experiencing a phenomenon of alignment”, summarize the researchers. “This may prove that there is an ingredient missing in our current models”. In any case, the nature of alignments on such scales is not easy to understand. “The nucleus of a quasar is made up of a supermassive black hole which can reach a mass equivalent to several billion times the mass of the sun. It has not been established that such objects behave like less massive galaxies. It is a very interesting first channel of investigation. We can legitimately suspect, by extrapolating from what we know about galaxies that the same mechanism can apply to quasars. The theory must be tested firstly by verifying that what explains alignments on smaller scales can include such vast sprawling structures, and secondly, must be confirmed by new observations”.

Towards other groups of quasars

This web of quasars is a challenge to astrophysicists today. According to theory, the expanding Universe did not have time to form such a huge structure. “It is intriguing that we have found such an ordered structure that is one gigaparsec in size. This could be due to a statistical fluctuation. But if we find others, we will have to reconsider the model and integrate new factors. We also need to check that we are not in the presence of a poorly-defined structure. We must characterize these alignments to verify whether they have a real meaning. All this work on quasars is very recent and there is a lot of work still to be done. We are observers more than theorists but we will further study the question. In the shorter term, we have requested more observing time on the VLT. We will look at other clusters of quasars to see if we notice the same behaviour and reinforce our observations”, the researchers point out.

(1) Damien Hutsemékers, Lorraine Braibant, Vincent Pelgrims, Dominique Sluse, Alignment of quasar polarizations with large-scale structures, Astronomy & Astrophysics, 19 novembre 2014.