The Seven-Hour Explosion: A Gamma-Ray Burst That Defies Everything We Know

Somewhere in a massive, dusty galaxy — one that may itself be in the middle of a catastrophic collision — something exploded. Or rather, something kept exploding, over and over again, for days. Light from that event has been traveling for nearly eight billion years. On July 2, 2025, it arrived at a detector aboard NASA’s Fermi Gamma-ray Space Telescope.

What Fermi recorded was unlike anything seen in over 50 years of gamma-ray astronomy: GRB 250702B, the longest gamma-ray burst ever detected.

And nobody knows exactly what caused it.

What Gamma-Ray Bursts Are (And Why They Usually Only Last Seconds)

Gamma-ray bursts are the most luminous electromagnetic events in the known universe — the brightest explosions since the Big Bang. A typical long gamma-ray burst happens when an extremely massive star, having exhausted its nuclear fuel, collapses in on itself to form a black hole. The infalling matter forms a disk, magnetic fields focus energy, and the black hole launches twin jets of plasma moving at very nearly the speed of light. As these ultrarelativistic jets punch through the collapsing star, they produce a blaze of gamma rays that can briefly outshine an entire galaxy.

The whole show, from first photon to last, usually lasts somewhere between a few seconds and a couple of minutes.

A few rare events called ultra-long gamma-ray bursts stretch this out to 1,000 seconds or more — perhaps when the progenitor is a much larger, puffier supergiant star, or when a newly formed magnetar keeps injecting energy into the jet. These are strange enough. But GRB 250702B belongs to a completely different league.

Seven Hours of Fury

The prompt gamma-ray emission from GRB 250702B lasted approximately 25,000 seconds — nearly seven hours. That’s not a rounding error. To put it in perspective: you could watch a full-length film, have dinner, and still have time for a leisurely walk before this burst was done.

“This is certainly an outburst unlike any other we have seen in the past 50 years,” said Eliza Neights, an astronomer at NASA’s Goddard Space Flight Center.

But the duration alone doesn’t capture how strange this event was. About 24 hours before Fermi detected the main burst, China’s Einstein Probe satellite had already noticed something unusual: a soft X-ray transient from the same location. Something was warming up — building toward the main event — an entire day in advance.

Then, hours after GRB 250702B, the source fired again: GRB 250702D and GRB 250702E, two additional bursts from the exact same location. Three explosions. One source. A pattern that hints at some kind of periodicity in whatever engine was driving this.

In total, researchers estimate the central engine powering this system was active for at least three days. No confirmed gamma-ray burst progenitor has ever behaved this way.

Reading the Universe With JWST

The explosion wasn’t visible in ordinary optical light — the host galaxy swathes the burst in dust. That’s part of what made untangling this event so difficult. It required a global armada of observatories: Fermi and Einstein Probe for gamma-rays and X-rays, NASA’s NuSTAR and the Chandra X-ray Observatory for detailed X-ray spectroscopy, the Very Large Array for radio signals, and the Hubble Space Telescope for the first optical look at the host.

But the crucial measurement came from the James Webb Space Telescope, whose Near Infrared Spectrograph (NIRSpec) can peer through cosmic dust with unprecedented sensitivity.

The Gompertz et al. team used JWST to identify narrow hydrogen emission lines in the host galaxy’s spectrum. They found a redshift of z = 1.036 ± 0.004, placing the source approximately 8 billion light-years from Earth. That means this explosion — whatever it was — happened when the universe was only about 5.8 billion years old, before Earth even existed.

At that distance, the gamma-ray energy released was staggering: at least 2.2 × 10⁵⁴ erg. For context, the Sun will radiate roughly 10⁵¹ erg over its entire 10-billion-year lifetime. GRB 250702B unleashed more than 2,000 Sun-lifetimes of energy in a matter of hours.

The Host Galaxy: Almost as Strange as the Burst

When astronomers finally resolved the host galaxy clearly — first with Hubble, then with Webb’s superior detail — they found something unusual. The galaxy is enormous by the standards of GRB hosts: around 10^10.66 solar masses (roughly 45 billion solar masses), dusty, highly asymmetric, and possibly in the middle of a major merger with another galaxy.

“In such vibrant and unprecedented detail, we see just one very large galaxy with a dust lane,” said Huei Sears, a postdoctoral researcher at Rutgers University who led follow-up Webb imaging. “The galaxy has such complex structure that it’s not 100% clear if there’s anything left to see of the explosion, but if there is, it’s really faint.”

The combination of an exotic burst in an exotic galaxy is telling. Normal long GRBs tend to occur in small, blue, star-forming dwarf galaxies. The host of GRB 250702B is nothing like that. Whatever process generated this explosion seems to thrive — or require — the kind of dense, dusty, merger-rich environment that massive galaxies provide.

A Mystery With Many Suspects

Researchers have published a remarkable suite of papers attempting to explain GRB 250702B. None of the standard models work cleanly. Here are the leading suspects:

The Helium Merger. In this scenario, a stellar-mass black hole (about 3 times the mass of the Sun) and a helium star — the stripped remnant of a massive star that lost its outer layers — orbit each other in a tight binary. Eventually the black hole spirals inward, falling into the helium star, and proceeds to consume and explode it from the inside out. The helium star expands as it is accreted, fueling a jet that persists for hours. Eliza Neights and colleagues, publishing in Monthly Notices of the Royal Astronomical Society, argue this “helium merger” model best explains the hard gamma-ray spectrum, sub-second variability, and extreme duration — and suggests GRB 250702B may be related to some ultra-long GRBs, some collapsars, and certain gravitational-wave progenitors.

The Micro-Tidal Disruption Event. Rather than a merger, perhaps a stellar-mass black hole or neutron star — kicked out of its birth position by a supernova — caught up with a companion star and shredded it through tidal forces. Beniamini, Perets & Granot (2025) propose three variants of this “micro-TDE” scenario, all of which can reproduce the curious day-long delay between the X-ray precursor and the main gamma-ray flare. Fallback accretion of the disrupted stellar material would naturally produce an engine that runs for days.

The Atypical Collapsar. A more extreme version of the standard model: a massive supergiant star collapsing, but one unusually large and slow to swallow, extending the jet activity far beyond normal bounds. Comprehensive gamma-ray analysis by Zhang et al. using data from Insight-HXMT, GECAM, and Fermi argues the X-ray decay after the main burst follows exactly the power-law expected from “fallback accretion” onto a black hole — consistent with a collapsar.

Intermediate-Mass Black Hole Tidal Disruption. Perhaps a black hole weighing thousands to millions of solar masses — larger than a stellar remnant but smaller than a galaxy’s central supermassive black hole — disrupted a passing star. The off-nuclear position of the burst within its galaxy would be consistent with this. However, Carney et al. note that several properties of the afterglow are inconsistent with known relativistic TDEs around supermassive black holes.

“A lot of the studies on this explosion provide different, and sometimes contradictory, explanations,” Sears said. “It’s still early in our understanding of what really happened.”

That’s not a failure of science — it’s science working exactly as it should.

Why This Matters

GRB 250702B is more than just a record-breaker. It may be evidence of an entirely new astrophysical phenomenon — a progenitor channel that has never been confirmed before.

The helium merger model, if correct, has tantalizing implications. Such a system, where a black hole falls into and consumes a companion star, would be a distant cousin of the compact binary mergers that LIGO detects as gravitational waves. It would mean some long GRBs and some gravitational-wave sources share an evolutionary ancestry — that the same binary star systems that end their lives colliding can sometimes end them in this slower, stranger, inside-out fashion.

“This gives us a unique chance to study the extremes of how stars and black holes evolve,” Sears said. “GRB 250702B could even be the discovery of something unexpected and new.”

The rate estimate from JWST confirms just how rare this is: GRB 250702B-type events occur at least 1,000 times less frequently than ordinary long gamma-ray bursts, and more than 100,000 times less frequently than core-collapse supernovae. In the lifetime of the Fermi telescope, this may be the only one we ever see.

The Collaboration Behind the Discovery

One detail worth savoring: this discovery required the simultaneous operation of instruments on at least a dozen different spacecraft and ground-based observatories, coordinated across many institutions and countries. Fermi first spotted the burst. Einstein Probe had the precursor. Swift and NuSTAR tracked the X-ray afterglow over weeks. The Very Large Array and Gemini and Keck and Magellan and Chandra all contributed. JWST delivered the final, definitive distance measurement.

“Only through the combined power of instruments on multiple spacecraft could we understand this event,” said Eric Burns, an astrophysicist at Louisiana State University.

This is what modern astronomy looks like: not one heroic telescope, but a whole civilization’s worth of instruments working in concert, each one contributing a different kind of light to the same extraordinary puzzle.

---

GRB 250702B is a reminder of something I find endlessly thrilling about physics: the universe is bigger, stranger, and more creative than our models. Every time we think we’ve catalogued all the ways that stars can die, the sky produces something we’ve never seen before. Eight billion years ago, in a galaxy tumbling into another galaxy, something happened that we still can’t fully explain.

The light from that moment spent billions of years traveling to reach us. On July 2, 2025, a small spacecraft named after Enrico Fermi caught it.

Now we just have to figure out what it was.

---

Key papers:

• Neights et al. 2025: “GRB 250702B: Discovery of a Gamma-Ray Burst from a Black Hole Falling into a Star” — MNRAS, DOI: 10.1093/mnras/staf2019
• Gompertz et al. 2025: “JWST Spectroscopy of GRB 250702B” — arXiv: 2509.22778
• Carney et al. 2025: “Optical/Infrared Observations of the Extraordinary GRB 250702B” — ApJL 994(2):L46, DOI: 10.3847/2041-8213/ae1d67
• Beniamini, Perets & Granot 2025: “Ultra-long Gamma-ray Bursts from Micro-Tidal Disruption Events” — arXiv: 2509.22779