Astronomers from Eotvos Lorand University (ELTE) have found a statistical relationship in quasar data that is consistent with a prediction made a decade ago. Quasars are very bright active galactic nuclei in which a supermassive black hole swallows gas and emits powerful light. The results were published in the journal Monthly Notices of the Royal Astronomical Society.
At the centers of large galaxies are supermassive black holes whose mass can exceed the Sun's by millions and billions of times. When two galaxies merge, their black holes draw closer and ultimately merge themselves. In the final moments, very strong gravitational waves are born - ripples in space-time. If such waves travel unequally in different directions, the remaining black hole receives a kick similar to the recoil of a rifle when fired.
If the kick is not enough to eject the hole from the galaxy forever, it returns, flies through the center, slows on the other side and moves back again. In this way it oscillates around the galactic nucleus for a long time, like a pendulum. At the center, where there is the most dust, the hole moves fastest.
In 2016, Peter Raffai, Zoltan Haiman and Zsolt Frei used computer models to predict that if such oscillating holes exist, quasars with faster black holes should on average be more heavily dusted. Dust makes light redder, like the atmosphere at sunset.
A new study led by Bence Becsy, an ELTE graduate now working at the University of Birmingham, tested this on real data. The scientists selected quasars from the Sloan Digital Sky Survey catalog with sufficiently high-quality measurements. For each object they determined the black hole's speed and the degree of reddening of the light, and then checked whether these indicators are related.
The relationship did appear: quasars with faster black holes turned out on average to be redder, that is, more strongly hidden by dust. Among the most dust-obscured objects, the share of heavily shadowed quasars was three times higher among holes with the highest speed than among the slowest ones. This is a correlation - a statistical relationship that by itself does not prove causality.
The authors emphasize that proving the ejection of a black hole in an individual quasar is difficult, since similar signs can be produced by other phenomena. That is why they looked for the statistical trace of an entire population. Checks using different methods of selecting the data did not change the result, but when analyzing only the light of gas clouds that do not follow the hole, the relationship disappeared.
The most natural explanation is that some quasars contain supermassive black holes ejected after a merger. If so, the strength of the signal suggests that in roughly half of the studied quasars the merger occurred relatively recently. However, this estimate remains uncertain, and other physical processes cannot be fully ruled out - further observations and more detailed models are needed.
The work also helps in preparing the LISA space telescope, which the European Space Agency plans to launch in 2035 to search directly for gravitational waves from mergers of supermassive black holes.