Gas giants are enormous planets made primarily of hydrogen and helium. They may contain dense central cores, but unlike Earth, they do not have solid surfaces you could stand on. In our solar system, Jupiter and Saturn are classic examples. Beyond our neighborhood, astronomers have identified many gas giant exoplanets, some far larger than Jupiter. The most massive of these worlds begin to resemble brown dwarfs, substellar objects sometimes called "failed stars" because they do not fuse hydrogen.
This overlap raises a major question in astronomy. How exactly do these massive planets form? One possibility is core accretion, the same process believed to have created Jupiter and Saturn. In this scenario, a solid core slowly builds up inside a disk of dust and ice, gathering rocky and icy material until it becomes massive enough to pull in surrounding gas. Another possibility is gravitational instability, where a swirling cloud of gas around a young star collapses quickly under its own gravity, forming a large object more like a brown dwarf.
A research team led by the University of California San Diego set out to investigate this mystery using data from the James Webb Space Telescope (JWST). By studying the HR 8799 star system, they uncovered evidence that offers a surprising answer. Their findings were published in Nature Astronomy.
The HR 8799 System and Its Super Jupiters
HR 8799 lies about 133 light years away in the constellation Pegasus. It hosts four massive planets, each between five and ten times the mass of Jupiter. These worlds orbit at distances ranging from 15 to 70 astronomical units, meaning even the closest planet is 15 times farther from its star than Earth is from the sun. The planet masses range from 5−10 MJup, so even the smallest of them outweighs Jupiter by a factor of five.
In some ways, HR 8799 resembles a scaled-up version of our solar system, which also features four outer giant planets stretching from Jupiter to Neptune. However, the sheer size of the HR 8799 planets and their wide orbits puzzled scientists. Earlier models based on our solar system suggested that planets forming through core accretion would not have enough time to grow so massive before the young star dispersed the surrounding disk of gas.
JWST Spectroscopy Reveals Sulfur Clues
To dig deeper, astronomers used spectroscopy, a technique that analyzes light to reveal the chemical makeup and physical properties of distant planets. Before JWST, researchers relied on ground-based telescopes to measure molecules such as water and carbon monoxide in exoplanet atmospheres. Over time, scientists realized that carbon and oxygen-based molecules are not ideal for tracing how planets form because their origins are difficult to pinpoint.
Instead, the team focused on more stable materials known as refractory elements. These elements, including sulfur, exist in solid form within the protoplanetary disk where planets take shape. Detecting sulfur in a gas giant atmosphere points strongly toward formation through core accretion.
"With its unprecedented sensitivity, JWST is enabling the most detailed study of the atmospheres of these planets, giving us clues to their formation pathways. With the detection of sulfur, we are able to infer that the HR 8799 planets likely formed in a similar way to Jupiter despite being five to ten times more massive, which was unexpected," stated Jean-Baptiste Ruffio, a research scientist at UC San Diego and first co-author of the paper.
HR 8799 is relatively young at around 30 million years old (for reference, our solar system is about 4.6 billion years old). Younger planets still retain heat from their formation, making them brighter and easier to analyze with spectroscopy.
JWST's high resolution spectrograph allows scientists to examine exoplanet light without interference from molecules in Earth's atmosphere. For the first time, astronomers detected detailed signatures of several rare molecules in the atmospheres of the system's three inner gas giants that had previously gone unseen.
Detecting Hydrogen Sulfide on Distant Worlds
Extracting this information was challenging. The planets are roughly 10,000 times fainter than their host star, and JWST was not originally optimized for such extreme contrasts. Ruffio developed new data analysis techniques to isolate the planets' faint signals. Jerry Xuan, a 51 Pegasi b Fellow at UCLA, built sophisticated atmospheric models to compare with the telescope's spectra and determine whether sulfur was present.
"The quality of the JWST data is truly revolutionary and existing atmospheric model grids were simply not adequate. To fully capture what the data were telling us, I iteratively refined the chemistry and physics in the models," he said. "In the end, we detected several molecules in these planets -- some for the first time, including hydrogen sulfide."
Clear signs of sulfur were found on the third planet, HR 8799 c, and the researchers believe it likely exists on the other two inner planets as well. The team also discovered that these planets contain higher amounts of heavy elements such as carbon and oxygen compared to their star, additional evidence that they formed as planets rather than as brown dwarf-like objects.
Rethinking Planet Formation Models
"There are many models of planet formation to consider. I think this shows that older core accretion models are outdated," stated UC San Diego Professor of Astronomy and Astrophysics Quinn Konopacky, another co-author of the paper. "And of the newer models, we are looking at ones where gas giants can form solid cores really far away from their star."
So far, HR 8799 remains the only directly imaged system known to contain four massive gas giants. However, other systems have been found with one or two even larger companions whose origins remain uncertain.
"I think the question is, how big can a planet be?" Ruffio stated. "Can a planet be 15, 20, 30 times the mass of Jupiter and still have formed like a planet? Where is the transition between planet formation and brown dwarf formation?"
Researchers continue to explore these questions, studying one star system at a time.
Partial list of authors: Jean-Baptiste Ruffio, Eve J. Lee and Quinn Konopacky (all UC San Diego); Jerry W. Xuan (California Institute of Technology and UCLA); Dimitri Mawet, Aurora Kesseli, Charles Beichman, Geoffrey Bryden and Thomas P. Greene (all California Institute of Technology); and Yayaati Chachan (UC Santa Cruz). Full list of authors appears in the paper.
This work was supported, in part, by the National Aeronautics and Space Administration (80NSSC25K7300 and FINESST Fellowship award 80NSSC23K1434). Any opinions, findings, and conclusions or recommendations expressed in this work are those of the author(s) and do not necessarily reflect the views of the National Aeronautics and Space Administration.
