Using a deceptively simple antenna roughly the size and shape of a dinner table, radio astronomers have made an unprecedented discovery: telltale fingerprints from the earliest stars in the cosmos, pressed into the afterglow of the universe's birth.
That signal, imprinted more deeply into the Big Bang's afterglow than scientists expected, could reveal much about the universe's youth and hint at the nature of dark matter, that mysterious substance that far outweighs all the normal matter in existence.
"To my mind … it's Nobel Prize-worthy twice, if it's real," said Avi Loeb, a theoretical astrophysicist at Harvard University who was not involved in the research. "Not only did they detect the signal, but it actually is bigger than one can accommodate in the standard cosmological model. And you need new physics in order to explain a signal as big as they detected."
According to the standard model, the Big Bang gave rise to the universe some 13.8 billion years ago, and the first stars were born on the order of 100 million years later. Those stars were not like the stars of today.
Because they coalesced out of the soup of neutral hydrogen (and a little helium) that filled the early cosmos, these stars grew large, burned bright and blue and then died quickly, probably surviving around 100 million years, give or take. (Our own sun, by comparison, is already 4.6 billion years old and has billions more years to go.)
When these short-lived stars went supernova, their explosive deaths forged heavier elements that seeded generations of stars to come. So understanding the stellar vanguard that brought light to the universe is key to understanding all the stars in galaxies today.
"They really lay the seeds for everything that comes after them," said Judd Bowman, an experimental astrophysicist at Arizona State University and lead author of one of the Nature papers.
But it's exceedingly difficult to glimpse actual evidence of those first stars, and thus to get a firm grip on the timeline of events in this epoch of cosmic history. That's partly because there aren't a lot of stars to see in this early era.
But it's also because the universe is expanding, and that expansion is stretching that ancient starlight into longer, "redder" wavelengths. That means even NASA's Hubble Space Telescope, which has been able to see galaxies from 400 million years or so after the Big Bang, can't spot them.
In a project dubbed EDGES (short for Experiment to Detect Global EoR Signature), Bowman and his colleagues decided to take a different approach.
In recent years, astronomers have studied the radiation afterglow of the Big Bang, known as the cosmic microwave background, or CMB. This radiation is subtle but extends over the entire sky, and astronomers have studied its tiny fluctuations in order to understand the underlying structure of the early universe.
The scientists realized that the cosmic microwave background, mixed with that soup of neutral hydrogen, might actually hold a subtle fingerprint from those primordial stars. That's because ultraviolet starlight would have shifted the hydrogen atoms' energy state, allowing them to absorb a particular wavelength out of the cosmic microwave background. Somewhere in the wavelengths that make up the CMB, they'd detect this telltale slice of missing light.
Finding this fingerprint in the Big Bang's afterglow was easier said than done. The local universe hurls an overwhelming amount of radio waves at Earth, drowning out this muted signal.
On top of that, the scientists were using a fairly simple instrument — a single radio detector roughly 6.4 feet long that resembles a dining table. With this single antenna, looking for a signal in one particular part of the sky would have been impossible.
Instead, they looked at the average radio spectrum across the entire sky and searched for discrepancies. They also placed their detector in a remote region of Australia, in the hopes of being as far away from human-generated radio waves as possible.
Sure enough, the scientists discovered a drop in the radio waves at 78 megahertz — a wavelength of light that had been dramatically stretched, thanks to the universe's expansion, from its original frequency of 1,420 megahertz. (The higher a wave's frequency, the shorter its wavelength.) This wavelength must be missing, the scientists argued, because it was absorbed by the hydrogen gas that was primed by the light from those early stars.
"I think it's a little bit like winning the lottery, in a sense," said study co-author Alan Rogers, a radio astronomer at MIT. At the same time, he added, luck often favors the prepared. "We spent a lot of time improving the calibration of the instrument."
The results show that these first stars were already shining just 180 million years after the Big Bang. As those early stars died, they likely left behind black holes, neutron stars and supernovas, producing X-rays that further heated the hydrogen gas. Thanks to all this heating, the telltale absorption signal disappears around 90 million years later.
"This is a huge potential result that's really a breakthrough in the more-than-a-decade-long effort to detect signals from the very early universe," said Gregg Hallinan, a Caltech radio astronomer who was not involved in the work. "This measurement is our first step to begin to understand that era where the first stars and galaxies actually formed."
Although the signal's location matched theoretical predictions, its shape did not. The dip in the light curve was flat-bottomed, like a U, and also twice as deep as scientists had predicted. That depth appears to imply that the hydrogen was much cooler than it should have been at that point in time.
If so, this would be groundbreaking because dark matter — which can't be seen or touched — has only been known to interact with normal matter through its gravitational influence. (That gravitational influence is pretty clear at large scales because there is more than five times as much dark matter as normal matter in the universe.)
But the first step, scientists said, would be for independent experiments to confirm that this signal really is out there.
MORE IN SCIENCE