Flares from magnetized stars can forge planets’ worth of gold

Astronomers have discovered a previously unknown birthplace of some of the universe’s rarest elements: a giant flare unleashed by a supermagnetized star. The astronomers calculated that such flares could be responsible for forging up to 10% of our galaxy’s gold, platinum and other heavy elements.

The discovery also resolves a decades-long mystery concerning a bright flash of light and particles spotted by a space telescope in December 2004. The light came from a magnetar—a type of star wrapped in magnetic fields trillions of times as strong as Earth’s—that had unleashed a giant flare.

The powerful blast of radiation only lasted a few seconds, but it released more energy than the sun does in 1 million years. While the flare’s origin was quickly identified, a second, smaller signal from the star, peaking 10 minutes later, confounded scientists at the time. For 20 years, that signal went unexplained.

Now, a new insight by astronomers at the Flatiron Institute’s Center for Computational Astrophysics (CCA) in New York City has revealed that the unexplained smaller signal marked the rare birth of heavy elements such as gold and platinum. In addition to confirming another source of these elements, the astronomers estimated that the 2004 flare alone produced the equivalent of a third of Earth’s mass in heavy metals. They report their discovery in a paper published on April 29 in The Astrophysical Journal Letters.

“This is really just the second time we’ve ever directly seen proof of where these elements form,” the first being neutron star mergers, says study co-author Brian Metzger, a senior research scientist at the CCA and a professor at Columbia University. “It’s a substantial leap in our understanding of heavy elements production.”

Most of the elements we know and love today weren’t always around. Hydrogen, helium and a dash of lithium were formed in the Big Bang, but almost everything else has been manufactured by stars in their lives, or during their violent deaths. While scientists thoroughly understand where and how the lighter elements are made, the production locations of many of the heaviest neutron-rich elements—those heavier than iron—remain incomplete.

These elements, which include uranium and strontium, are produced in a set of nuclear reactions known as the rapid neutron-capture process, or r-process. This process requires an excess of free neutrons—something that can be found only in extreme environments. Astronomers thus expected that the extreme environments created by supernovae or neutron star mergers were the most promising potential r-process sites.

It wasn’t until 2017 that astronomers were able to confirm an r-process site when they observed the collision of two neutron stars. These stars are the collapsed remnants of former stellar giants and are made of a soup of neutrons so dense that a single tablespoon would weigh more than 1 billion tons. The 2017 observations showed that the cataclysmic collision of two of these stars creates the neutron-rich environment needed for the formation of r-process elements.

However, astronomers realized that these rare collisions alone can’t account for all the r-process-produced elements we see today. Some suspected that magnetars, which are highly magnetized neutron stars, could also be a source.

Metzger and colleagues calculated in 2024 that giant flares could eject material from a magnetar’s crust into space, where r-process elements could form.

“It’s pretty incredible to think that some of the heavy elements all around us, like the precious metals in our phones and computers, are produced in these crazy extreme environments,” says Anirudh Patel, a doctoral candidate at Columbia University and lead author on the new study.

The group’s calculations show that these giant flares create unstable, heavy radioactive nuclei, which decay into stable elements such as gold. As the radioactive elements decay, they emit a glow of light, in addition to minting new elements.

The group also calculated in 2024 that the glow from the radioactive decays would be visible as a burst of gamma rays, a form of highly energized light. When they discussed their findings with observational gamma-ray astronomers, the group learned that, in fact, one such signal had been seen decades earlier that had never been explained. Since there’s little overlap between the study of magnetar activity and heavy-element synthesis science, no one had previously proposed element production as a cause of the signal.

“The event had kind of been forgotten over the years,” Metzger says. “But we very quickly realized that our model was a perfect fit for it.”

In the new paper, the astronomers used the observations of the 2004 event to estimate that the flare produced 2 million billion billion kilograms of heavy elements (roughly equivalent to Mars’ mass). From this, they estimate that one to 10% of all r-process elements in our galaxy today were created in these giant flares. The remainder could be from neutron star mergers, but with only one magnetar giant flare and one merger ever documented, it’s hard to know exact percentages—or if that’s even the whole story.

“We can’t exclude that there could be third or fourth sites out there that we just haven’t seen yet,” Metzger says.

“The interesting thing about these giant flares is that they can occur really early in galactic history,” Patel adds. “Magnetar giant flares could be the solution to a problem we’ve had where there are more heavy elements seen in young galaxies than could be created from neutron star collisions alone.”

To narrow down the percentages, more magnetar giant flares need to be observed. Telescopes like NASA’s Compton Spectrometer and Imager mission, set to launch in 2027, will help better capture these signals. Large magnetar flares seem to occur every few decades in the Milky Way and about once a year across the visible universe—but the trick is to catch it in time.

“Once a gamma-ray burst is detected, you have to point an ultraviolet telescope at the source within 10 to 15 minutes to see the signal’s peak and confirm r-process elements are made there,” Metzger says. “It’ll be a fun chase.”