Detecting dark matter: seven hints point to primordial black holes



For several decades, astrophysicists have suspected the existence of so-called dark matter, constituting the bulk of all matter in the universe, but have not detected it directly. Rather than continue to search for new, hypothetical particles for dark matter, Sébastien Clesse suggests that it could be made of primordial black holes, citing seven supporting observations that hint at their existence.1

Beginning in the 1930s, the Swiss astronomer Fritz Zwicky drew his colleagues’ attention to the problem of 'missing' mass in the cluster of galaxies of Coma Berenices, but wasn’t until 1970 that the American astronomer Vera Rubin hypothesised that the ‘missing’ mass was ‘dark matter’. In the almost 50 years since, matter has been a dual concept: on the one hand ordinary matter (also called baryonic because it’s composed essentially of baryonic particles, including protons and neutrons) and on the other dark matter. The first, of which we and stars are made, represents about 15% of the universe’s matter. The second represents the remaining 85%.

Indirect detection

Scientists didn’t call the 'substance' dark matter without reason: it doesn’t emit traditional electromagnetic radiation – it’s dark, invisible, but interacts gravitationally with its environment just like ordinary matter.

A half century of inquiry has led researchers to find several indirect proofs of its existence, such as galaxy rotation: stars at the edge move almost at the same speed as those closer to the centre, whereas speed should decrease with distance. It doesn’t because of the presence of a huge halo of invisible material that surrounds galaxies.

Another indirect proof: light emitted by faraway galaxies is deviated by the gravity of those masses it encounters in its path, although much more so than the gravity of visible masses should do. There is thus another component whose total mass exceeds that of the visible matter and which bends the light rays. Finally, according to cosmologists, dark matter’s existence is required to explain the tiny fluctuations in the cosmic microwave background at the origin of all structures in the universe (see also ‘The Planck legacy’).

‘So what we have,’ explains Sébastien Clesse, an FNRS postdoctoral researcher in UCLouvain’s Cosmology, Universe and Relativity Research Group (CURL), ‘are observations at totally different time and distance scales, which are consistent and indicate that 85% of dark matter is required. In almost half a century, many models have been developed to try to define what this strange matter is. And of course we’ve tried to discover the particles that make it up.’

WIMPs versus MACHOs

One of these models, which was popular until very recently, was based on weakly interacting massive particles ('WIMPs'), which, as the name indicates, would be massive, electrically neutral particles that interact only very rarely with ordinary matter and which would have appeared in the primordial universe. But all attempts to discover WIMPs have been unsuccessful. A previous hypothesis, in the 1980s, was based on massive compact halo objects (MACHOs), and brought together black holes, planets, and brown dwarfs (stars not massive enough to trigger the nuclear fusion of hydrogen) toward a rather obvious goal: finding sufficient non-luminous mass to explain the anomalies described above. It was unsuccessful and has been somewhat forgotten. ‘However,’ Dr Clesse argues, ‘the idea was interesting because black holes are massive, don’t emit light, and don’t move at speeds close to that of light; in short, they behave like dark matter.’ But there’s a problem: black holes as conceived by astrophysicists are formed by the death of stars, which were not yet present in the primordial universe. Since the presence of dark matter is necessary very early in the history of the universe to explain the minute fluctuations of the cosmic microwave background, classic black holes, as well as planets and brown dwarfs, cannot be dark matter. Adiós MACHOs.  

Primordial black holes

‘So the temptation was great,’ Dr Clesse says, ‘to imagine that other types of black holes exist, which would have appeared directly in the primordial plasma, from very dense regions of the universe that would have collapsed very early, less than a millisecond after the Bing Bang. Hence the name primordial black holes and the 1970s hypothesis of their existence by Stephen Hawking, Bernard Carr and Georges Chapline, which was revived in 2016 following the discovery of gravitational waves.’
In 2015, before the detection of gravitational waves, Dr Clesse and his colleague Juan Garcia-Bellido of the University of Madrid proposed a theoretical model according to which dark matter could be composed of primordial black holes ranging in mass from 0.01 to 10,000 solar masses (one solar mass is the mass of our sun), in clusters containing up to several million black holes. ‘The detected gravitational waves account for the merger of black holes on the order of 30 solar masses, whereas most astrophysical models predicted much smaller masses, typically a few solar masses. Massive black hole mergers are therefore not uncommon, which fits well with our model of a strong presence of primordial black holes. Since then, other massive black hole mergers have been identified that generate gravitational waves.’ This was the first hint of the possible existence of primordial black holes.

The second hint: black hole spin (rotation speed). The detections suggest a very weak spin, that is, little to no rotation. ‘That's impossible if these black holes are of stellar origin’, Dr Clesse says, ‘because a star rotates and the conservation of angular momentum makes the resulting black hole rotate as well. So far, no model of a black hole of stellar origin can explain these weak spins, whereas it’s exactly what we expect for primordial black holes.’

The third hint: gravitational microlensing. Observing stars in galaxies close to ours over a long period makes it possible to notice when a compact non-luminous object – such as a black hole – passes between us and an observed star, because its luminosity is temporarily amplified. ‘Such events have been detected, but they were deemed as not having been due to the presence of black holes in our galaxy’s halo. On the other hand, the few events seemed to exclude the possibility that dark matter is composed of black holes of less than 10 solar masses. But such observations are consistent with our scenario because our primordial black holes cover a wide range of masses and only very few of them would be detectable by this microlensing phenomenon. We can therefore think that these microlensing events are due to primordial black holes, which could be the dark matter.’

Galaxies dominated by dark matter

The fourth hint comes from observing small galaxies composed of only a few thousand stars but which are known to be dominated by dark matter, which may be 1,000 times more abundant than ordinary matter. If these ‘dwarf’ galaxies are composed mainly of primordial black holes, they would be expected to intersect regularly, exchange kinetic energy, and make galaxy size unstable, to make it grow. ‘Some observations can support an argument for reducing the abundance of primordial black holes’, Dr Clesse explains. ‘But we can reverse the argument: Would there not exist a critical radius within which we would no longer see dwarf galaxies dominated by dark matter? This is the case: we observe that dwarf galaxies all have a radius greater than 30-40 light years – never less. This could indicate that smaller ones are unstable because of the compact nature of dark matter.’ For Dr Clesse, this is a clear way to test his scenario: if we observe a single galaxy dominated by dark matter and with a radius of less than 30 light years, his scenario collapses; if, on the other hand, they are all larger, it would be a good indication of the compact and non-corpuscular nature of dark matter. 

For the fifth hint, we remain among dwarf galaxies but not necessarily those dominated by dark matter. Some models find that a particular type of dark matter could form the core of such galaxies. What if such dark matter is composed of primordial black holes interacting gravitationally with each other? ‘We come to the same conclusion, namely the presence of a dark matter core’, Dr Clesse confirms. ‘Our theory thus sticks to the observations and is not invalidated by this type of galaxy.’

Supermassive black holes

To find the sixth hint, let's change scale again and consider supermassive black holes. The origin of these monsters – millions or billions of solar masses – hidden in the centre of massive galaxies, remains an enigma. They are observed very early, less than a billion years after the Big Bang. It’s therefore impossible that they were formed from the explosions of the first stars: there wasn’t enough time for them to attain such mass. ‘In our hypothesis,’ Dr Clesse explains, ‘even if most primordial black holes don’t exceed a few dozen solar masses, there are enough that have the size of a few thousand solar masses to explain, after merging and accreting matter, the formation of these supermassive monsters.’

For the seventh and last hint, we have to search the diffuse infrared and X-ray backgrounds. Fluctuations in these forms of radiation are correlated with each other. To explain this correlation, we must find very early in the primordial universe more halos of dark matter than expected. ‘That's exactly what we expect if the dark matter is primordial black holes; their statistical distribution shows that they’re likely to generate this type of halo. To date, it’s even the best solution for explaining the correlation.

Such hints lead to considering primordial black holes candidates for dark matter: ‘Each of these observations, taken individually, can be explained otherwise, but our scenario for primordial black holes explains them in a unified and coherent way.’

It’s a scenario that must be confirmed, which the UCLouvain group will try to do, including by studying gravitational waves.2 ‘The best way to test whether primordial black holes exist is to observe black holes that could not have been created from stars, those that are either very massive (more than 100 solar masses) or of very little mass. Gravitational waves should allow us to observe black holes of less than 1.4 solar masses, the limit below which there can be no stellar black holes. If we find some, it’ll be a clear indication that we’re probably dealing with a primordial black hole.’

Henri Dupuis


(1) ‘Seven Hints for Primordial Black Hole Dark Matter’, S. Clesse & J. Garcia-Bellido, Phys. Dark Univ. Preprint available at: 
(2) A group of Belgian researchers including several from UCLouvain (G. Bruno, K. Piotrzkowski, C. Ringeval and S. Clesse) recently joined the VIRGO collaboration, a gravitational wave detector experiment that will test the scenario.

A glance at Sébastien Clesse's bio

‘As an adolescent, I hesitated between history, maths, physics and computer science.’ Any outliers? None in Sébastien Clesse’s eyes. ‘The only way to combine the four was in physics, more specifically cosmology. It covers mathematics and computer science and even history – the longest, the universe’s!’ Thus he studied physics at the University of Namur, where he obtained his bachelor’s degree in 2006. He then began a joint PhD at ULB and UCLouvain (under the supervision of Michel Tytgat and Christophe Ringeval) and defended his thesis in 2011. Next, a long postdoctoral research journey in Cambridge, Munich, Namur (thanks to a BESPO return grant), and Aachen. He returned to UCLouvain and UNamur in 2017, as an FNRS postdoctoral researcher at the new Cosmology, Universe and Relativity Research Group. ‘In cosmology, we study the ultimate; no other sciences can explain the object of our research in other ways. Cosmology, just like particle physics, is at the very end of the science chain, and at the forefront of understanding the world.’ 

Published on January 29, 2019