Generative Data Intelligence

Astronomers Say They Have Spotted the Universe’s First Stars



A group of astronomers poring over data from the James Webb Space Telescope (JWST) has glimpsed light from a rare isotope of helium in a distant galaxy, which could indicate the presence of the universe’s very first generation of stars.

These long-sought, inaptly named “Population III” stars would have been ginormous balls of hydrogen and helium sculpted from the universe’s primordial gas. Theorists started imagining these first fireballs in the 1970s, hypothesizing that, after short lifetimes, they exploded as supernovas, forging heavier elements and spewing them into the cosmos. That star stuff later gave rise to Population II stars more abundant in heavy elements, then even richer Population I stars like our sun, as well as planets, asteroids, comets and eventually life itself.

“We exist, therefore we know there must have been a first generation of stars,” said Rebecca Bowler, an astronomer at the University of Manchester in the United Kingdom.

Now Xin Wang, an astronomer at the Chinese Academy of Sciences in Beijing, and his colleagues think they’ve found them. “It’s really surreal,” Wang said. Confirmation is still needed; the team’s paper, posted on the preprint server on December 8, is awaiting peer review at Nature.

Even if the researchers are wrong, a more convincing detection of the first stars may not be far off. JWST, which is transforming vast swaths of astronomy, is thought capable of peering far enough away in space and time to see them. Already, the gigantic floating telescope has detected distant galaxies whose unusual brightness suggests they may contain Population III stars. And other research groups vying to discover the stars with JWST are analyzing their own data now. “This is absolutely one of the hottest questions going,” said Mike Norman, a physicist at the University of California, San Diego who studies the stars in computer simulations.

A definitive discovery would allow astronomers to start probing the stars’ size and appearance, when they existed, and how, in the primordial darkness, they suddenly lit up.

“It’s really one of the most fundamental changes in the history of the universe,” Bowler said.

Population III

About 400,000 years after the Big Bang, electrons, protons and neutrons settled down enough to combine into hydrogen and helium atoms. As the temperature kept dropping, dark matter gradually clumped up, pulling the atoms with it. Inside the clumps, hydrogen and helium were squashed by gravity, condensing into enormous balls of gas until, once the balls were dense enough, nuclear fusion suddenly ignited in their centers. The first stars were born.

The German astronomer Walter Baade categorized the stars in our galaxy into types I and II in 1944. The former includes our sun and other metal-rich stars; the latter contains older stars made of lighter elements. The idea of Population III stars entered the literature decades later. In a 1984 paper that raised their profile, the British astrophysicist Bernard Carr described the vital role this original breed of star may have played in the early universe. “Their heat or explosions could have reionized the universe,” Carr and his colleagues wrote, “… and their heavy-element yield could have produced a burst of pregalactic enrichment,” giving rise to later stars richer in heavier elements.

Carr and his co-authors estimated that the stars could have grown to immense sizes, measuring anywhere between a few hundred and 100,000 times more massive than our sun, because of the large volume of hydrogen and helium gas available in the early universe.

Those at the heavier end of the range, so-called supermassive stars, would have been relatively cool, red and bloated, with sizes that could encompass almost our entire solar system. Denser, more modestly sized variants of Population III stars would have shone blue hot, with surface temperatures of some 50,000 degrees Celsius, compared to just 5,500 degrees for our sun.

In 2001, computer simulations led by Norman explained how such large stars could form. In the present universe, clouds of gas fragment into lots of small stars. But the simulations showed that gas clouds in the early universe, being much hotter than modern clouds, couldn’t as easily condense and were therefore less efficient at star formation. Instead, entire clouds would collapse into a single, giant star.

Their immense proportions meant the stars were short-lived, lasting a few million years at most. (More massive stars burn through their available fuel more quickly.) As such, Population III stars wouldn’t have lasted long in the history of the universe — perhaps a few hundred million years as the last pockets of primordial gas dissipated.

There are many uncertainties. How massive did these stars really become? How late into the universe did they exist? And how abundant were they in the early universe? “They’re completely different stars to the stars in our own galaxy,” Bowler said. “They’re just such interesting objects.”


Because they are so far away and existed so briefly, finding evidence for them has been a challenge. However, in 1999, astronomers at the University of Colorado, Boulder predicted that the stars should produce a telltale signature: a specific frequency of light from helium-2. This unstable form of helium only contains two protons in its nucleus, while regular helium also has two neutrons. “The helium emission is not actually originating from within the stars themselves,” explained James Trussler, an astronomer at the University of Manchester; rather, it was created when energetic photons from the stars’ hot surfaces plowed into gas surrounding the star.

“It’s a relatively simple prediction,” said Daniel Schaerer of the University of Geneva, who expanded on the idea in 2002. The hunt was on. 

Finding the First Stars

In 2015, Schaerer and his colleagues thought they might have found something. They detected a possible hint of a helium-2 signature in a distant, primitive galaxy that might have been linked to a group of Population III stars. Seen as it appeared 800 million years after the Big Bang, the galaxy looked as if it might contain the first evidence of the first stars in the universe.

Later work led by Bowler disputed the findings. “We found evidence for oxygen emission from the source. That ruled out a pure Population III scenario,” she said. An independent group then failed to detect the helium-2 line seen by the initial team. “It wasn’t there,” Bowler said.

Could others fare better?

Astronomers pinned their hopes on JWST, which launched in December 2021. The telescope, with its enormous mirror and unprecedented sensitivity to infrared light, can peer more easily into the early universe than any telescope before it. (Because light takes time to travel here, the telescope sees faint, faraway objects as they appeared long ago.) The telescope can also do spectroscopy, breaking up light into its component wavelengths, which allows it to look for the helium-2 hallmark of Population III stars.

Wang’s team analyzed spectroscopy data for more than 2,000 of JWST’s targets. One is a distant galaxy seen as it appeared just 620 million years after the Big Bang. According to the researchers, the galaxy is split into two pieces. Their analysis showed that one half seems to have the key signature of helium-2 mixed with light from other elements, potentially pointing to a hybrid population of thousands of Population III and other stars. Spectroscopy of the second half of the galaxy has yet to be done, but its brightness hints at a more Population III-rich environment.

“We are trying to apply for observing time for JWST in the next cycle to cover the entire galaxy,” Wang said, in order to “have a shot of confirming such objects.”

The galaxy is a “head-scratcher,” according to Norman. If the helium-2 results stand up to scrutiny, he said, “one possibility is a cluster of Population III stars.” However, he’s unsure if Population III stars and later stars could mix together so readily.


Daniel Whalen, an astrophysicist at the University of Portsmouth, was similarly cautious. “It definitely could be evidence of a mixture of Population III and Population II stars in one galaxy,” he said. However, although this would be “the first direct evidence” of the universe’s first stars, Whalen said, “it’s not clean evidence.” Other piping hot cosmic objects can produce a similar signature of helium-2, including scorching disks of material that swirl around black holes.

Wang thinks his team can rule out a black hole as the source because they did not detect specific oxygen, nitrogen or ionized carbon signatures that would be expected in that case. However, the work still awaits peer review, and even then, follow-up observations will need to confirm its potential findings.

Hot on the Trail

Other groups using JWST are also hunting for the first stars.

Besides looking for helium-2, another search method, proposed by the astronomer Rogier Windhorst of Arizona State University and colleagues in 2018, is to use the gravity of giant clusters of galaxies to see individual stars in the early universe. Using a massive object like a cluster to warp light and magnify more distant objects (a technique known as gravitational lensing) is a common way astronomers obtain views of distant galaxies. Windhorst believed that even individual Population III stars approaching the edge of a heavy cluster “could in principle undergo nearly infinite magnification” and pop into view, he said.

Windhorst leads a JWST program that is attempting the technique. “I’m pretty confident that in a year or two we will have seen some,” he said. “We already have some candidates.” Similarly, Eros Vanzella, an astronomer at the National Institute for Astrophysics in Italy, is leading a program that’s studying a clump of 10 or 20 candidate Population III stars using gravitational lensing. “We are just playing with the data now,” he said.

And there remains the tantalizing possibility that some of the unexpectedly bright galaxies already seen by JWST in the early universe could owe their brightness to massive Population III stars. “These are exactly the epochs where we expect the first stars are forming,” Vanzella said. “I hope … that in the next weeks or months, the first stars will be detected.”


Latest Intelligence