New observations suggest that stars began forming just 250 million years after the Big Bang — a record-breaker that will likely open a new line of cosmological inquiry.

Astronomers peering back into time suggest that the cosmic dark ages — before the universe hosted its sea of twinkling lights — might have lasted no more than 250 million years. The team presented their results in the journal Nature today.

Takuya Hashimoto (Osaka Sangyo University, Japan) and his colleagues used the Atacama Large Millimeter/submillimeter Array (ALMA) to peer at a galaxy whose light was emitted 550 million years after the big bang, picking up a long-sought signal: oxygen. It’s the most distant galaxy for which astronomers have been able to detect individual elements — and that single element has a big story to tell.

Earliest star formation in a far-away galaxy
The galaxy cluster MACS J1149.5+2223 features in this image taken with the Hubble Space Telescope. The inset shows ALMA's view of oxygen (green) in the galaxy MACS1149-JD1, whose light has traveled 13.28 billion light-years to Earth.
Inset: ALMA (ESO / NAOJ / NRAO) / Hashimoto et al.; Background: NASA / ESA Hubble Space Telescope / W. Zheng (JHU) / M. Postman (STScI) / CLASH Team

Because only hydrogen, helium, and a little lithium emerged from the Big Bang, the young universe was pristine. It wasn’t until the first generation of stars exploded, breathing carbon, oxygen and other heavy elements into the cosmos, that the universe’s inventory of elements increased. So, the detection of oxygen 550 million years after the Big Bang suggests that a generation of stars had already formed and died by this point.

Hashimoto and his colleagues estimate that the first generation of stars would have formed 250 million years after the Big Bang — even though Planck measurements of the cosmic microwave background indicate that star formation wouldn't have been prevalent in this epoch. That said, the results are in line with a tentative result from the EDGES experiment, which found a signal from stars forming just 180 million years after the Big Bang. The EDGES result still awaits confirmation from other groups performing similar experiments.

For decades, teams of scientists have been racing to find the signatures of the first stars — and it’s more than cosmic curiosity. “There is renewed optimism we are getting closer and closer to witnessing directly the birth of starlight,” says coauthor Richard Ellis (University College London) in a press release. “Since we are all made of processed stellar material, this is really finding our own origins.”

An Explosive Start

What’s more, the team also used infrared data taken with the Hubble Space Telescope and the Spitzer Space Telescope to glean the number of stars within the galaxy. Typically, galaxies form a small number of high-mass stars and a large number of low-mass stars. The high-mass stars die first, exploding as supernovae a few million years after they form. But the lowest-mass stars can survive for trillions of years — much longer than the age of the universe.

But Hashimoto and his colleagues saw that the galaxy contained even fewer few high-mass stars than expected — meaning that the star formation kicked-off strong, tapered off, and then started forming stars again. That’s the opposite of predictions from simulations of the early universe. The star formation rate was expected to increase with time at these early epochs, starting out slow and then growing exponentially — at least for high-mass galaxies like the one Hashimoto’s team detected.

“It may mean that we don’t really understand the first generation of galaxies sufficiently well,” says coauthor Erik Zackrisson (Uppsala University, Sweden). “There might be some ingredient that is missing from the simulations.”

Discovering that missing ingredient will be the goal of future work, but Zackrisson has a few ideas. It could be that the very first generation of stars produced far more powerful supernovae than theorists suspect. Or perhaps this particular galaxy hosts a ravenous supermassive black hole. Both would unleash powerful winds that would push gas away from the galaxy and suppress further star formation.

Rychard Bouwens (Leiden University, The Netherlands), who was not involved in the study, argues that the paper’s conclusions are reliable yet uncertain only because the team is peering so far back into the universe’s history. “It’s always this way when you’re at the cutting edge,” he says. “It might be providing us with important clues to what happened at very early times in the universe, but we can't be sure until we observe more objects.”

Both Bouwens and Zackrisson are excited for the 2020 launch of the James Webb Space Telescope, which will be able to directly image galaxies at these early times. Not only that, but Webb is expected to image hundreds if not thousands of these young galaxies. As such, Bouwens compares the current observations to standing on a ship that is enveloped in a thick fog. Although you might be able to see the hazy outline of a lighthouse with Hubble or Spitzer, the James Webb Space Telescope will act like rays of sunlight — forcing the fog to clear and allowing us to see the lighthouse clearly, as well as the rocky coast behind it.


T. Hashimoto et al. “The onset of star formation 250 million years after the Big Bang.” Nature, 2018 May 17.


Image of millerf1


May 17, 2018 at 7:47 am

Could the presence of oxygen mean that life developed during the cooling period after the big bang?
A recent speaker at Fermilab caused me to realize that the cold dark period was not entirely cold and dark during the whole period. There was a transition from super hot and dense to cool as the inflation ended and star formation began. There could be a period that life could exist in the free space clouds?
As stated elsewhere I think such models are oversimplified. The existence of early quasars also indicates that things were going on much earlier.

You must be logged in to post a comment.

Image of Robert-Black


May 18, 2018 at 4:21 pm

I read in another article that early stars (starting out as all hydrogen & helium) were much larger than stars formed later. That would also play into how long they would last before going Supernova, I'd think. Such a low metal universe would have been difficult to live in, since our own bodies use significant quantities of elements heavier than helium.

You must be logged in to post a comment.

Image of Deryk Houston

Deryk Houston

May 24, 2018 at 7:09 pm

It might also mean that the entire Big Bang theory is completely wrong. This seems the most likely conclusion, considering that we really have no idea how light changes as it travels so many billions of miles through a universe jam packed with all sorts of high energy particles and materials flying in every direction. For years, I've watched as astronomers twisted and bent the math, the data and just about everything else including throwing out date that doesn't match the accepted theory, rather than ask...... what else could this data be telling us if the big bang theory is wrong?

You must be logged in to post a comment.

Image of fif52


December 4, 2018 at 7:25 am

Is there any hydrogen involved in this green cloud?

You must be logged in to post a comment.

You must be logged in to post a comment.