Optical and infrared telescopes study radiation streaking through the universe with wavelengths in the visible-light and infrared parts of the electromagnetic spectrum. And using observations with these telescopes, astronomers today know that star-forming activity in the universe peaked some 8-10 billion years ago – and then declined by about 10-times to its current rate.
However, we don’t know why this slowdown happened – and not for lack of trying. The principal issue is that we didn’t, and still don’t, have enough information about a key ingredient of stars: neutral atomic hydrogen.
A neutral hydrogen atom consists of one proton, in the nucleus, and one electron orbiting it. All subatomic particles have a property called spin, which denotes the particle’s angular momentum. If the proton and the electron in the hydrogen atom have both their spins oriented in the same direction, the atom as a whole has slightly more energy than if the proton’s and the electron’s were pointing in opposite directions (a.k.a. anti-parallel). The higher energy state is called the excited state.
Sometimes, in an extremely rare event, an excited hydrogen atom may transition to the lower energy state by emitting energy in the form of electromagnetic radiation – i.e. a photon – of wavelength 21.106 cm (rounded down to 21 cm for convenient communication). So observing radiation of this wavelength is a sign that neutral hydrogen atoms are present.
This event is so rare that it is nearly impossible for physicists to observe it happening spontaneously in a laboratory on Earth. And even though astronomers know the precise wavelength of the radiation to look for, the emission itself is quite weak, and is quite tough to spot when it’s coming from distant galaxies.
Axiomatically, observing and measuring 21-cm radiation from distant galaxies is bound to be a notable feat. These are practically photons carrying a shade more than a trillionth of a trillionth of a joule of energy each that have travelled for billions of lightyears (one lightyear is approximately 9.4 trillion km) – and some of which have, by happenstance, encountered Earth.
A team of researchers from the National Centre for Radio Astrophysics (NCRA), Tata Institute of Fundamental Research, Pune, and the Raman Research Institute (RRI), Bengaluru, recently reported measuring just this 21-cm radiation from distant galaxies, corresponding to a time when star-forming activity had just begun to dip. Their findings, made with the Giant Metre-wave Radio Telescope (GMRT) in Pune, were published on October 14, 2020.
“This the earliest epoch in the universe for which there is a measurement of the atomic gas content of galaxies, and represents a significant leap in our understanding of gas in galaxies and its connection to star-formation,” Aditya Chowdhury, a PhD student at the NCRA and one of the members of the study team, told The Wire Science.
Both space and time in the universe are united in some ways by the speed of light. A prominent effect of this is that objects that are one lightyear away in space are effectively one year away in time. If we also include the fact that the universe is expanding, light emitted by a galaxy eight billion years ago will have to travel many more billions of lightyears to reach Earth – and when it does, it will carry information of a reality that existed eight billion years ago. On the flip side, the information will have become much weaker because of the distance it has travelled.
In the face of this challenge, the NCRA-RRI team captured individual 21-cm emission signals coming from 7,653 galaxies by lining them up in three dimensions – two corresponding to their position and one to the redshift. This way, they were able to adjust for different differences and obtain an average signal.
The researchers could figure out how far away each of these galaxies was from Earth in a straightforward way. Say a galaxy has emitted some light that is travelling towards Earth. The universe is constantly expanding, as a result of which space itself is expanding. So the wavelength of this light is stretched out, making it seem redder. Astronomers know this distortion as the redshift.
The higher the redshift of a galaxy, the farther away it is.
The galaxies whose atomic hydrogen content the researchers had detected were located at redshifts of 0.74 to 1.45. On this scale, 0 refers to current time, 0.1 is about a billion years ago, 0.75 is about 5.78 billion years ago and 1.45 is 9.2 billion years ago.
The older the galaxies are, the farther away they are in space. But because the universe is also expanding at an accelerated pace, distance and time are not linearly related. So galaxies that are about 8 billion years away in time are actually more than 20 billion lightyears away in space.
According to Kanan Datta, an assistant professor of physics at the Presidency University, Kolkata, it is difficult to overstate the challenge of directly measuring the amount of neutral hydrogen from the early universe – i.e. in distant parts of it – because the 21-cm signal from individual galaxies that are so distant is very faint.
Chowdhury said detecting the same signal from an individual galaxy at redshifts of 0.74-1.45 would require more than 10,000 hours of observing time, and only with two currently operational radio-telescopes – the GMRT and the MeerKAT array in South Africa.
Star formation is associated with regions that contain dust and cold, dense gas, including molecular hydrogen (H2), that can collapse to form stars. However, neutral atomic hydrogen (HI) eventually forms H2. “So on larger, galactic scales, HI is considered to be the fuel for star formation,” Khandai said.
In star-forming galaxies close to us, astronomers have studied how the mass of galaxies and the star formation rates are related to their cold gas – i.e. HI and H2 – content. But estimates of HI content in distant galaxies have been lacking.
Chowdhury and his colleagues found that how rapidly new stars formed could have been mediated by the availability of hydrogen itself.
Specifically, they found that galaxies about 8-10 billion years ago contained nearly 2.5-times more atomic hydrogen than galaxies do today. This historical availability of hydrogen could have contributed to the prolific creation of stars until that time. The researchers also estimated that the gas’s abundance would have dwindled to a point where star-formation became too resource-intensive around one or two billion years later.
“Once the atomic hydrogen content is exhausted, and if fresh atomic hydrogen does not accumulate from the galaxies’ environments, the star-formation activity will rapidly decline,” Chowdhury said. “This provides a plausible explanation for the decline in star-formation activity over the past eight billion years.”
Similarly, he expects based on the current rate – of accumulating 0.002-0.4 solar masses per year – in nearby star-forming galaxies, like the JKB18 and the Andromeda, that they’ll be able to keep making new stars for another 7-9 billion years.
Suman Majumdar, an assistant professor of astronomy and astrophysics at IIT Indore, explained that the study provides observational evidence for theoretical studies that predicted a decline in star formation rate in galaxies from the early universe to now.
The NCRA-RRI team’s work is also notable because it used the GMRT, the most sensitive radio telescope in India, located near Pune. One of the foundational goals of this instrument – installed by the noted radio-astronomer Govind Swarup in 1995 – is to study the 21-cm emission of neutral hydrogen in outer space. Swarup passed away on September 7 this year.
In 2016, the same group of researchers had tried to detect the 21-cm emission signals with GMRT using the same approach – by stacking signals from multiple galaxies. But the telescope’s specifications only allowed them to obtain an upper limit on the strength of the averaged signal.
The GMRT was upgraded the following year. A galaxy’s redshift determines the amount by which light from it is stretched. So the 21-cm signal from galaxies at different redshifts can only be observed at different frequencies. This in turn means the telescope’s bandwidth determines the redshift range that can be observed at one time. The pre-upgrade GMRT had a bandwidth of 33 MHz; after the upgrades, this became 400 MHz. The current study was the product of this revamp.
“This work demonstrates that the GMRT can play a very important role in direct estimation of the neutral hydrogen contents of the early universe. We are about to see this field flourish in the near future and this work is definitely an important milestone,” Datta said.
Indeed, radio astronomy is set to receive a boost with upcoming observatories, perhaps especially the Square Kilometre Array (SKA ). The GMRT uses 30 radiation-collecting dishes, or antennae, each 45 metres wide for a total collecting area of 20,000-30,000 sq. metres, from low to high frequencies (120 MHz to 1.4 GHz). The SKA will use thousands of antennae, built in Australia and South Africa, for a total collecting area of 1,000,000 sq. metres – or 1 sq. km – to study radiation from 50 MHz to 14 GHz. Its first phase (which includes the MeerKAT telescope) is expected to be ready around 2023.
Majumdar agreed with Datta, saying astronomers would be able to use the SKA to conduct studies like the NCRA-RRI group’s, and faster and with more precision. He also noted that while the new results marked the first step in resolving a long-standing puzzle, they were also limited to one relatively small patch of the sky. “They have only studied 7,653 galaxies for this – which may sound like a large number, but it is not large enough at the cosmological scale,” Majumdar said.
“One of the main cosmological principles states that no single direction is special.” So astronomers will have to observe many more galaxies in different parts of the sky to validate and then build on the finding.
“Astronomy is an observationally driven field and theories of galaxy formation hinge upon such important observations to make further progress,” Khandai said. “Future observations will further constrain the HI content of galaxies at higher redshifts, which would answer many questions about how galaxies form and evolve.”