The story of our quest for Sagittarius A*, the supermassive black hole at the heart of the Milky Way.

Once upon a time

Holmdel, New Jersey, early 1930s. Engineer Karl Jansky is conducting an experiment with an antenna he built, using a set of four Ford Model-T tires to rotate it. His goal is to find possible sources of interference with radio transmissions. Most come from thunderstorms, but he can also detect a faint steady hiss of unclear origin. Further investigation showed that it came from the direction of the Sagittarius constellation (the Archer), where the centre of the Milky Way is.

The mystery became even more… mysterious in the 1950s, when astronomers discovered more and more very bright radio sources disseminated across the Universe, located at the centre of other galaxies. These galaxies however looked like very faint stars when observed in visible light and were thus named “quasars”, short for “quasi stars”.

The puzzle started coming together in 1963, as astronomer Maarten Schmidt calculated the distance to one of these quasars, 3C 273, found to lay 2.5 billion light years away. By combining its distance and apparent brightness as seen from Earth, it turned out that 3C 273 had to shine extraordinarily brightly in radio waves, emitting more than 1000 times the energy radiated by the whole Milky Way with its 100 billion stars.

How could these objects be producing so much energy? One explanation was that the energy came from enormous amounts of matter getting hot while being swallowed by a gigantic, supermassive black hole — the last light before oblivion. But possible explanations are not enough: science needs evidence, and astronomers wondered how to prove that supermassive black holes are indeed quasars’ invisible engines.

Well, black holes are famous for their gravitational pull, so why not trace the orbits of stars or gas clouds close to the quasars? Great idea, with one flaw though: quasars were too far away to directly see stars or gas clouds orbiting them. Hence, astronomers had to focus on the radio sources which, though not as powerful as quasars, were in closer galaxies. And which galaxy is closer than the one you are in?

All eyes on me

So astronomers turned their eyes to the centre of the Milky Way and the stars lying there. There was one problem: the thick, gigantic clouds of interstellar dust between us and the galactic centre. This dust absorbs the light from the core of the Milky Way, forcing astronomers to give up on using visible light and go for longer wavelengths, such as infrared and radio waves, which pass through the clouds almost undamped.

A wide-field view towards the central regions of the Milky Way. The dark bands are dust clouds that block our view in visible light, requiring astronomers to use infrared or radio observations to look through the dust.

Credit: ESO/S. Guisard (www.eso.org/~sguisard)

This is what Charles Townes and his team did at Berkeley in the 1970s, as they tracked the motion of gas around Sgr A* in infrared light. From this motion they inferred that an amount of mass between 2 and 4 million times the mass of the Sun should be enclosed within a few light years from Sgr A*, compatible with a supermassive black hole.

An astonishing result, but not the ultimate answer, as two problems were still in place. First, there was no guarantee that gravity was the only force influencing the orbit of the observed gas. In fact, magnetic fields and stellar winds could have also distorted the orbit in a way that resembled the gravitational pull of a black hole. Second, the available technology did not allow the team to get closer to the radio source more than a few light years, which is very small on galactic scales but still millions of times larger than the size of a supermassive black hole. Hence, it was not possible to exclude that other objects were responsible for the observed orbits of stars and gas. Better tracers and closer distances were needed, but this required new technology.

Two are better than one

The UT4 telescope at ESO’s VLT is seen here using its laser to perform adaptive optics observations of the galactic centre.

Credit: Juan Carlos Muñoz-Mateos/ESO

In the first half of the 90s two teams of astronomers looked independently at the orbits of stars around the galactic centre. One was led by Reinhard Genzel, former member of Townes’ team in Berkeley and now co-director of the Max Planck Institute for Extraterrestrial Physics in Germany; the other was led by Andrea M. Ghez, currently a professor at the University of California, Los Angeles, in the US. Genzel’s team used two ESO facilities in Chile: the 3.6m New Technology Telescope (NTT) at La Silla Observatory, and the Very Large Telescope (VLT) a few years later. Ghez’s team on the other hand used the 10 m Keck telescope in Hawaii.

Genzel’s NTT observations relied on a technique called speckle imaging, which uses very short exposures to freeze the atmospheric distortion. But this technique is limited to relatively bright stars, so another technique came to the rescue: adaptive optics. In adaptive optics, a thin mirror is deformed hundreds or even thousands of times per second, correcting the atmospheric blur. This requires monitoring in real time a bright reference source, either an actual star within the field of view or an artificial “star” created with a laser high up in the atmosphere.

Using adaptive optics, Genzel’s and Ghez’s groups managed to see stars orbiting within a light month from Sgr A* at the tremendous speeds of about 2000 km/s, yet another hint that a very compact mass of a few million solar masses was lurking there.

Could more compelling evidence be collected? Yes, if you were able to spot stars even closer to the galactic centre.

And such a star, known as S2, was eventually discovered, coming as close to Sgr A* as 17 light hours, equivalent to four times the orbit of Neptune around the Sun. The result confirmed that an object of a few million solar masses was concentrated on a planetary scale, ruling out many possible alternatives to a supermassive black hole. By 2002, astronomers were convinced that a supermassive black hole was the most likely object in the galactic centre.

In the following years, astronomers could track even more stars, observe flashes of light from matter disappearing beyond the event horizon and even witness “in real time” a gas cloud meeting its destiny, ripped apart by the gravitational force of the black hole.

But there was an appointment astronomers could not miss: the S2 star takes 16 years to complete its orbit around Sgr A*, and in 2018 it would pass again at its closest distance from the black hole. This was an invaluable opportunity to put General Relativity to some of its hardest tests and collect more evidence on the true nature of Sgr A*.

Be there or be square

To make the best out of this appointment, Genzel’s team had to achieve even better sharpness than what could be achieved with a single 8 m telescope. So they decided to use ESO’s Very Large Telescope Interferometer (VLTI), which combines the light collected by individual telescopes with a technique known as interferometry, offering unparalleled sharpness. Together with ESO they built GRAVITY, one of the second-generation instruments mounted at the VLTI, able to accurately pinpoint the location of stars with extreme accuracy.

And the results were spectacular. The team was able to observe effects which were in splendid agreement with the predictions of general relativity. They detected gravitational redshift, where the light from a star loses a fraction of its energy as it climbs the extreme gravitational potential of the black hole, hence stretching to longer wavelengths. They also observed Schwarzschild precession, a rotation of the star’s orbit around the black hole, which they detected by comparing the position of the star in 2018 to where it was in 2002.

So, all done? Of course not, you’re never done in science.

In the following years, the team zoomed in even deeper, and discovered fainter stars such as S29, which made its closest approach to the black hole in late May 2021 at the record speed of 8740 km/s and at a distance of 13 billion kilometres (90 times the distance between the Sun and Earth). These measurements allowed them to make the most precise estimate of the supermassive black hole’s distance to us — 26996 light-years — and mass, equal to 4.297 million times that of the Sun.

GRAVITY observations of stars around Sgr A*. Using new processing techniques the GRAVITY team was able to detect fainter stars than before.

Credit: ESO/GRAVITY collaboration

For their work on Sgr A*, in 2020 Genzel and Ghez shared the Physics Nobel Prize. As you can see, we’ve really come to the point where we are literally writing down Sgr A*’s identity card, and no ID is complete without a picture. Could the Event Horizon Telescope (EHT) Collaboration deliver one? They have used interferometry to link radio observatories scattered across the globe — including ALMA and APEX, co-owned by ESO — to create a virtual telescope the size of the Earth. The EHT is so powerful that it was able to capture the first image ever of a black hole — the supermassive one at the centre of the M87 galaxy, more than 50 million light years away. This image was released in 2019, and the next EHT results are widely anticipated.

And more will come when ESO’s Extremely Large Telescope (ELT) starts operating in Chile later this decade, as it will unveil even fainter objects close to Sgr A*, thus providing even more information on the black hole.

But that, dear reader, will be a story for another post.