Did This Scorching-Hot Planet Lose—and Regain—an Atmosphere?

About 40 light-years outside our solar system is a rocky planet so close to its host star that it takes about one and a half Earth days to complete a full orbit. The surface reaches an average temperature upwards of 530 Kelvin (on par with your oven’s broiler), and scientists believe the mantle is at most a few hundred meters thick and cracked like an eggshell.

It’s known as GJ 1132 b, but it may as well be the pits of hell. And in spite of the odds, a team of exoplanet researchers think that it might have an atmosphere—its second one, to be precise. In a paper published last Friday in The Astronomical Journal, a team of astrophysicists, geophysicists, and atmospheric chemists announced the detection of an atmosphere that’s roughly 99 percent molecular hydrogen, with trace amounts of methane, acetylene, and hydrogen cyanide floating above its pockmarked surface.

The thing is, no one really thinks this planet should still have an atmosphere, even those researchers. “It should have lost everything,” says Raissa Estrela, the paper’s coauthor, who researches exoplanet atmospheres at NASA’s Jet Propulsion Laboratory. In fact, a second team of exoplanet researchers submitted an independent analysis of the same data at roughly the same time that casts doubt on whether this atmosphere actually exists.

GJ 1132 b likely began its life as a sub-Neptune planet—a class of gaseous planets that the Kepler Space Telescope has shown to be the most common in our galaxy. They range from 1.5 to 3 times the size of Earth. This one was believed to be enveloped in a thick atmosphere of hydrogen and helium swirling around a dense, rocky core. But because of the planet’s proximity to its host star, researchers believe this gas envelope was burned away by intense ultraviolet radiation during the first 100 million years of its life.

In theory, all that should remain on this planet is a barren, irradiated rocky surface—but recent observations from the Hubble Space Telescope might tell a different story. Over the course of 20 orbits and 24 hours of observation time, a team of astronomers used the telescope’s imaging spectrograph to catch signatures of the light absorbed in the planet’s atmosphere as it transited its host star.

In GJ 1132 b’s case, the resulting spectrum indicated the presence of molecular hydrogen. For a planet that receives about 19 times as much solar radiation as Earth does, this result was perplexing. Because it’s so light, hydrogen escapes a planet’s gravitational pull very easily. When hydrogen molecules are heated, they expand and rise in the atmosphere, eventually reaching a velocity high enough to escape the grasp of smaller planets. The intense heat from its M-dwarf star should have left the planet a barren husk.

“That really raised the question: What is the origin of the atmosphere that we see?” asks Mark Swain, an astrophysicist at JPL and the paper’s lead author. “That led us to this detective work and an investigation into the possibility of regenerating the atmosphere from the mantle.” In other words, they suspected that after the planet lost its first atmosphere, it grew a second one.

Swain and Estrela turned to two papers, published in 2018 and 2019, that found that in the early days of a sub-Neptune planet’s lifecycle, when it's still hanging on to its primordial atmosphere, the pressure and temperature near the molten surface is high enough that a substantial amount of the hydrogen floating around in the atmosphere is absorbed into an ocean of magma. As the planet cools and its thick atmosphere is burned away, much of this additional hydrogen could be trapped beneath the solidifying surface. “The theory describing this is actually very new,” says Swain. “I was not up to speed on it until we started to interpret this.”

But if the surface had already cooled, how was this huge hydrogen reserve escaping? The 2018 paper from researchers at Grenoble Alpes University in France calculated the orbital configuration of the planet. They found that it actually has a pronounced eccentricity, the measure of how much a planet’s orbit deviates from a perfect circle—basically how squashed its elliptical orbit is. GJ 1132 b’s eccentricity is on par with Mercury’s, which receives twice as much solar radiation at its perihelion, or the point where it's closest to the sun, as it does when it’s furthest from it. The gravitational pull from the star would tug on the planet, creating friction in the molten interior and distorting its shape. And that could make for a geologically active planet, one in which materials from below the surface are pushed up through it.

This same process happens on Jupiter’s moon Io, where the surface is dotted with over 400 volcanoes—the most geologically active place in our solar system. If GJ 1132 b is also volcanically active, this volatility could be behind the planet's new atmosphere. Paul Rimmer, an atmospheric chemist at the University of Cambridge and another author on the paper, trained a chemical computer model to reproduce the conditions observed in the planet’s atmosphere. “I looked at what the chemistry might look like near the very, very top of a volcano,” says Rimmer. “If you have a certain amount of carbon, hydrogen, oxygen, and nitrogen coming out, there are certain ways that they want to fit together.”

On Earth, volcanoes mainly eject carbon dioxide, water, and sulfur. But Rimmer found that volcanoes on GJ 1132 b are likely ejecting this buried hydrogen, along with methane and hydrogen cyanide—two gases that are typically not found in equal abundance on rocky, terrestrial planets. “It was a very, very unusual sort of chemistry compared to what you would expect to find on Earth,” he says.

But there is at least one small pocket in the Earth’s mantle where we’ve discovered similar conditions. In 2016, a mining company found an exceedingly rare mineral called tistarite beneath Mount Carmel in northern Israel. Geologists determined that it was ejected from a volcano during the Cretaceous period, and was initially formed in magma with hardly any oxygen. “It’s very rare on Earth, but this would be all over the planet on GJ 1132 b,” says Rimmer. This unique volcanism could theoretically produce methane and hydrogen cyanide in equal amounts, he says, but it’s all still very conceptual. Rimmer notes there’s still more work to be done to study the geochemistry of this planet and others like it to determine whether this chemistry is plausible.

Sukrit Ranjan, a planetary scientist at Northwestern University who previously worked with Rimmer to model phosphine in Venus’s atmosphere—a hotly contested recent claim—says these findings are incredibly exciting. We have plenty of examples in our own solar system of planets that have rich hydrogen atmospheres, he notes, but we’ve never before observed a rocky planet dominated by hydrogen. “That’s not something that was predicted ahead of time,” says Ranjan. “For the most part, folks assume that if you have an H₂[hydrogen]-dominated atmosphere it should be lost relatively early on in the planet’s history and you probably won’t be able to regenerate and maintain it.”

Laura Kreidberg, who directs research on exoplanet atmospheres at the Max Planck Institute, would like to see an independent analysis of the data before jumping to conclusions. “There are a lot of little decisions in the data processing that can produce unexpected bumps and wiggles,” says Kreidberg. “I’d like to see the spectrum reproduced by another team using independent methods to see if they get the same thing.”

In fact, that process is already underway. Last week, another research team led by Lorenzo Mugnai, an astrophysicist at the Sapienza University of Rome, released a separate paper that independently analyzes the same Hubble data on GJ 1132 b. But when Mugnai’s team crunched the data, they found that the planet’s spectrum to be relatively flat—in other words, it had no detectable atmosphere. “It’s very hard to be sure of the cause of the differences, because it’s a very difficult analysis,” Mugnai says. “We know the devil is in the details.”

The two teams are having regular meetings to figure out what led to such a dramatic discrepancy in their results, but Mugnai and Swain both think the problem could lie in how they account for the variation in sunlight as the planet moves in front of its star, a parameter known as limb darkening. “A star is not uniform in brightness from the center to the edge,” Swain says. “When the planet is close to one edge or another, it appears to block less light, because part of the star it’s covering up is dimmer on average than the rest of the star.”

To correct for this effect, researchers need to process their data with a model that can take into account the dimming and brightening of the star. Both teams used the same model, but with different coefficients. They’re now planning on swapping methods to see if they can replicate the results of the other team.

Even so, Darius Modirrousta-Galian, the coauthor of Mugnai’s paper, thinks it’s highly unlikely that GJ 1132 b has been able to retain enough hydrogen to produce a second atmosphere because it’s so close to its host star. Exoplanet researchers are still uncertain of how influential stellar radiation can be in the formation of atmospheres. “The approach we take is that actually stellar irradiation is so strong, and it causes winds on the planet to have supersonic velocities and extreme particle velocities, that the atmosphere basically boils off,” he says.

Modirrousta-Galian says the amount of hydrogen in the primordial envelope that would be required to overcome this loss and to make a second atmosphere would be several times the mass of the planet. “We have no problem within our model that the planet could have been born with a hydrogen atmosphere,” he says. “The conclusion we arrived at is that we just don’t have one now.”

Still, more research—and ideally new observations by the James Webb Space Telescope, set to launch on October 31—is needed to verify, or further complicate, either of the teams’ results. If GJ 1132 b does prove to have a hydrogen atmosphere, it could open up new avenues of exploration for planetary scientists. For one thing, these atmospheres would be much easier to analyze than those of small planets with denser envelopes made of heavier elements. Hydrogen’s low molecular weight contributes to a broader, puffier atmosphere for light to shine through. And that makes for a stronger spectrographic signature that’s easier to read from Earth.

Both teams are pushing the limits of what’s possible with the Hubble Space Telescope, which was launched in the year 2000, two years before astronomers discovered the first known exoplanet. At 1.16 times the size of Earth, GJ 1132 b is the smallest planet that has ever had a published transmission spectrum, Swain notes. “I think the exciting thing here is gaining a better understanding of what details really matter to the study of small planets,” he says.