Zero Genie

Table of contents

A version of this story appeared in Science, Vol 379, Issue 6628.

In 2013, from a windowless office at the University of Chicago, Laura Kreidberg peered into the sky of a distant planet. Kreidberg, then an astronomy graduate student, parsed data from the Hubble Space Telescope, its gaze set on the exoplanet GJ1214b, roughly three times the size of Earth. GJ1214b is a popular target for astronomers seeking clues to the nature of alien worlds, and not only because it’s relatively close, a mere 48 light-years away. It also orbits its star every 1.5 days, and each lap exposes its atmosphere to inspection.

On every pass, the planet briefly eclipses a portion of the star’s face. A fraction of the star’s light filters through the planet’s atmosphere, where different molecules absorb the light at characteristic wavelengths. By layering Hubble observations of 15 eclipses, Kreidberg hoped to see how GJ1214b’s atmosphere gobbled up parts of the rainbow, creating distinctive dips in the light. Markers for water vapor would hint at the presence of oceans, whereas sulfur signatures could indicate volcanoes.

But to her disappointment, “the spectrum got flatter and flatter and flatter,” says Kreidberg, now at the Max Planck Institute for Astronomy. Light just wasn’t getting through. The planet’s impenetrable atmosphere could only indicate one thing, she concluded: A layer of aerosols high in the atmosphere was blocking the light. The chemical richness she wanted to capture was hidden beneath mysterious clouds or hazes.

Frustrated, she wrote up her results. The paper, published in Nature in 2014, sparked a reckoning for exoplanet astronomers. They had spent decades preparing to dissect exoplanet atmospheres with JWST, the massive new space telescope that began operations last year with six times the light-gathering power of Hubble. With its superior resolution and a wavelength range extending into the infrared, they hoped to detect potential exoplanet biosignatures: molecules that indicate life, such as oxygen or a combination of carbon dioxide and methane.

But they had mostly ignored the possibility that JWST’s view would be obstructed by suspended aerosols—either cloud droplets, formed when gases condense, or haze particles, formed through light-driven chemical reactions. Thanks to observations of GJ1214b and a handful of similarly muted exoplanets, astronomers now acknowledge that hazes and clouds are likely to veil most of JWST’s exoplanet targets—as they do every planet and moon with an atmosphere in the Solar System.

What we don’t want is to spend all our efforts building enormous, billion-dollar telescopes to get data we don’t understand.

• Peter Gao
• Carnegie Institution for Science

“What we don’t want is to spend all our efforts building enormous, billion-dollar telescopes to get data we don’t understand,” says planetary scientist Peter Gao of the Carnegie Institution for Science. Researchers have a decent understanding of how a handful of common planetary gases interact and condense into clouds. But hazes—which can be made up of any number of tens of thousands of complex molecules—are more mysterious. Without laboratory experiments to show how they arise, what they are, and how they interact with light, Gao says, “it’s like putting together a puzzle, but you’re missing a piece.”

Thankfully, a handful of haze researchers are prepared to supply that missing piece. One is Sarah Hörst, a planetary scientist at Johns Hopkins University who has spent years trying to understand methane-based hazes on Titan, an icy moon of Saturn, by re-creating them in the lab. Prompted by the wake-up call from the GJ1214b spectrum—what she calls “the most expensive flat line in the history of science”—Hörst modified her setup to create haze particles in warmer exoplanet conditions. At the behest of anxious JWST observers, Hörst and her colleagues this week released long-awaited measurements of how the particles absorb and scatter light—providing a critical update to outdated models and a guide to help observers make sense of spectra flattened and distorted by alien hazes. “We have to go to the laboratory to reproduce the conditions,” Kreidberg says. “That is really the key to solving this problem.”

THE EFFORT to simulate photochemistry in the lab traces back to the famed Miller-Urey experiment in 1952, which attempted to re-create prebiotic chemistry on early Earth. In the 1970s, Carl Sagan and Bishun Khare, planetary scientists at Cornell University, extended this work to other bodies in the Solar System. By subjecting nitrogen and methane gases to light and radiation, they produced sticky grains made up of long-chain carbon-based molecules. They called them “tholins,” after the Greek word for “muddy.” (The runner-up name was “star tar.”) These particles helped explain the diffuse blue glow around Pluto that NASA’s New Horizons mission captured in 2015, and their organic complexity helped motivate the search for life on Titan.

The scientists also used an array of instruments to study how the particles block light across a wide range of wavelengths—from x-rays all the way to microwaves. The work was “maybe a little too meticulous,” recalls Gene McDonald, then a postdoc in Sagan’s lab who’s now a biochemist at the University of Texas (UT), Austin. But because it was so thorough, the data remain valuable half a century later.

Sagan and Khare showed how the haze particles scatter light in different directions, which would erode many of the distinctive absorption features in starlight passing through a planet’s atmosphere. But they found this effect depends on the size and composition of the particles. For example, Earth’s sky is blue because small nitrogen and oxygen molecules in the atmosphere preferentially scatter shorter, blueish wavelengths of light, whereas clouds appear white because their much larger water droplets scatter sunlight evenly.

Besides scattering light, tholins also absorb it at certain wavelengths, with each molecular bond soaking up a sliver of the spectrum, the researchers found. Even though an exoplanet haze might stomp out signs of atmospheric gases, a tholin’s own absorption signature offers clues. If it is one that has been studied in a lab, the atmospheric composition can be inferred based on the gases used in the experiment.

“Most of the time, people treat these particles as nuisances,” says Lia Corrales, an astronomer at the University of Michigan, Ann Arbor. In a paper posted this month to the preprint server arXiv, Corrales and colleagues show how even slight changes to experimental conditions can produce tholins with distinct optical properties. “Aerosols themselves have some of that [compositional] information because they have spectral shapes,” she says.

To glean that information from the spectra they observe, astronomers need to have a point of comparison: a predicted spectrum for a given mixture of gases, clouds, and haze. Scientists spin out forecasts for hypothetical atmospheres using computer programs such as the Planetary Spectrum Generator, which sometimes draws more than 1 million hits a month, according to its developer, Geronimo Villanueva at NASA’s Goddard Space Flight Center (GSFC). In addition to varying the atmospheric composition, users of the program can account for hazes or clouds by entering their optical properties.

The trick is knowing those properties. Databases exist for clouds, but not for different hazes. In fact, most modelers still rely on tholin properties from the Sagan-Khare experiments 40 years ago—which were designed for Titan, an environment that’s nothing like most exoplanet atmospheres. Some instead opt to use data for ordinary soot, which has well-studied properties but is also unlikely to be found on an exoplanet. “We basically have a menu of two to choose from, and beyond that, we just make things up—we guess,” says Eliza Kempton, an astronomer at the University of Maryland, College Park. A wrong guess could mean misinterpreting the atmosphere’s makeup. The knowledge gap about exoplanet-relevant hazes has been “the elephant in the room” for years, Villanueva says.

NIKOLE LEWIS, an astronomer at Cornell, says efforts to fill the gap have been slowed by a disconnect between exoplanet and planetary science. Because of the way NASA funding is siloed, “the same infrastructure that we use to explore planets in our own Solar System is not immediately transferable to exoplanets,” she says. “We just had to basically grow the momentum to be able to do these types of experiments.”

And so, Lewis grew it herself. As disappointingly flat exoplanet spectra trickled in, Lewis turned to Hörst, a close friend from graduate school who had just started to build her own tholin chamber for studies of Titan. Over the course of a year, Lewis worked to sell Hörst on modifying her setup to use the different gas mixtures and higher temperatures expected for looming exoplanet targets. Finally, in 2016, the team received funding from a new NASA exoplanet research program. They were ready to make some exo–star tar. “We knew how hard these experiments were going to be, and there was a little bit of buyer’s remorse there,” Hörst says with a chuckle. “The things we do for the exoplanet people.”

The first challenge was figuring out the ingredients of atmospheres that haven’t yet been observed. To start off, Hörst and Lewis narrowed their scope to super-Earths, exoplanets two to 10 times the mass of our planet. (As the most common type of planet in the Milky Way, super-Earths are a primary target for JWST.) Then, they ran models to predict the atmospheric gases that might be present for a range of temperatures and bulk compositions. They were left with 12 archetypes—hypothetical combinations of gases such as hydrogen, water vapor, and carbon dioxide that should represent genuine exoplanet conditions.

Then came the experimental hurdles. First, they had to rig up a way to simultaneously mix eight or nine gases—many of which are toxic, flammable, or otherwise hard to handle. Next, rather than cooling the gases down as for Titan, they had to be able to heat the mixture to more than 520°C. Worst of all were the water-rich planets. They might be sanctuaries for life, but they are a pain for research scientist Chao He, who treks to the lab repeatedly through the night to replace stashes of dry ice. The cold helps the water vaporize into the chamber at the right pressures. “I have to babysit the experiment,” He says. “It’s not pleasant.”

When He flips a valve, the gas mixture races down pipes, twirls around a heating coil, and flows through a stainless steel chamber the size of a water bottle. There, the gas is held at light pressures meant to simulate the thin air at the top of an atmosphere. He floods the chamber with ultraviolet (UV) light, to replicate starlight, or radiation from a plasma discharge source, to mimic solar storms and cosmic rays. After 3 to 5 days, he pumps out the gas and cracks open the chamber within a glovebox. With a spoon, He scrapes tholins off the walls into a vial or dish.

The team was concerned the modifications might not yield enough hazes to study—or that they would destroy the setup entirely. “This was a giant leap of faith on Sarah’s part,” Lewis says. “And it paid off.”

Using microscopes, spectrometers, and other instruments, the researchers measured the size, optical properties, and density of the particles. It was immediately obvious that the samples, gold and olive green in color, were radically different from Sagan and Khare’s brownish Titan tholins. Some experiments generated haze without adding methane, long thought to be an essential ingredient for these large organic molecules. Another surprise came after Sarah Moran, then Hörst’s graduate student who’s now an astronomer at the University of Arizona, brought samples to the Institute for Planetary Sciences and Astrophysics in Grenoble, France. (She prayed she wouldn’t get stopped by customs for the “weird, strangely colored powders” in her suitcase.) There, Moran found that the tholins appeared to contain amino acids and nucleobases—primary building blocks of life.

That may be no coincidence—some researchers believe hazes could be critical to kick-starting life. In 2017, Giada Arney, an astrobiologist at GSFC, modeled the behavior of hazes on early Earth—which she deems “the most alien planet we have geochemical data for.”

At the time, billions of years ago, Earth’s atmosphere was thought to be mostly nitrogen with some carbon dioxide, methane, and water vapor. That recipe would give rise to organic hazes similar to Titan’s and water clouds like those on Earth today. Arney ran climate models to simulate the effects of these aerosols on Earth-like planets. She found that, in addition to hosting complex organic chemistry, some hazes reflect light and help keep a planet from overheating, whereas others reduce UV rays on the surface by up to 97%. Both effects could make an exoplanet more habitable.

She also found certain hazes only form when the atmosphere contains so much methane that living organisms must be making it. Such a haze could itself be a biosignature, Arney says—and one that JWST might be able to detect. But not without “a good library of optical properties of hazes,” she says.

AT LONG LAST, she’s getting her wish. On 10 January, Hörst and colleagues posted a study to arXiv that unveils optical properties across the entire JWST wavelength range for two of their exoplanet tholins—generated in the warm, water-rich atmospheres that produced the most haze.

The team plans to cook up several more haze analogs, coordinating its experimental conditions with incoming JWST observations and cataloging the most abundant tholins. Exoplanet astronomers are yearning to get their hands on the data, Gao says. “We’ve been waiting for this for decades, really,” he says. “Sarah’s work will be immediately used by groups around the world.”

Still, Hörst’s experiments will only explore a fraction of possible planetary conditions. Thankfully, she’s not the only player in the tholin game. At NASA’s Jet Propulsion Laboratory, Benjamin Fleury is exploring hazes that might form on much hotter exoplanets, the giant, close-in “hot Jupiters” that reach temperatures of more than 1200°C. At LATMOS, a research institute near Paris, scientists modified Titan haze experiments to approach conditions that might be found on an Earth-like exoplanet—by adding oxygen, for example.

Some researchers are focused instead on exoplanet clouds, including Alexandria Johnson, an atmospheric scientist at Purdue University who grows and studies clouds in a chamber. She wonders how clouds and hazes might interact on an exoplanet—whether, for example, haze particles might promote cloud formation by providing “nucleation sites” for gases to condense. The experimental diversity is critical, says Thomas Gautier, a planetary scientist on the LATMOS team. “Nobody’s trying to claim that we do it better than the other—we do it differently, and that’s a good thing.”

Partitioning the work only pays off if everyone’s on the same page, though. Xinting Yu at UT San Antonio recently initiated the first cross-laboratory tholin comparison study—an effort that will standardize production and measurement techniques and boost confidence in the results. Meanwhile, Ella Sciamma-O’Brien at NASA’s Ames Research Center is building a centralized database for the different teams to house their optical measurements.

The collaborations will come in handy as new JWST observations pour in—including ones of GJ1214b, the planet that started it all. In July 2022, Kempton, Kreidberg, and colleagues got about 40 hours of time on the telescope to look at it. They are still analyzing the observations, but—thanks to JWST’s infrared vision and help from these experimental hazes—Kreidberg hopes to decode the atmosphere that has long evaded her.

“This planet is kind of like a white whale—we’ve been chasing after its secrets for over a decade now,” Kreidberg says. “I think this time we finally caught it.”