“You can put the carbon nanofibers into cement to strengthen the cement,” said Jingguang Chen, a professor of chemical engineering at Columbia with a joint appointment at Brookhaven Lab who led the research. “That would lock the carbon away in concrete for at least 50 years, potentially longer. By then, the world should be shifted to primarily renewable energy sources that don’t emit carbon.”

As a bonus, the process also produces hydrogen gas (H₂), a promising alternative fuel that, when used, creates zero emissions.

Capturing or converting carbon

The idea of capturing CO₂ or converting it to other materials to combat climate change is not new. But simply storing CO₂ gas can lead to leaks. And many CO₂ conversions produce carbon-based chemicals or fuels that are used right away, which releases CO₂ right back into the atmosphere.

“The novelty of this work is that we are trying to convert CO₂ into something that is value-added but in a solid, useful form,” Chen said.

Such solid carbon materials — including carbon nanotubes and nanofibers with dimensions measuring billionths of a meter — have many appealing properties, including strength and thermal and electrical conductivity. But it’s no simple matter to extract carbon from carbon dioxide and get it to assemble into these fine-scale structures. One direct, heat-driven process requires temperatures in excess of 1,000 degrees Celsius.

“It’s very unrealistic for large-scale CO₂ mitigation,” Chen said. “In contrast, we found a process that can occur at about 400 degrees Celsius, which is a much more practical, industrially achievable temperature.”

The tandem two-step

The trick was to break the reaction into stages and to use two different types of catalysts — materials that make it easier for molecules to come together and react.

“If you decouple the reaction into several sub-reaction steps you can consider using different kinds of energy input and catalysts to make each part of the reaction work,” said Brookhaven Lab and Columbia research scientist Zhenhua Xie, lead author on the paper.

The scientists started by realizing that carbon monoxide (CO) is a much better starting material than CO₂ for making carbon nanofibers (CNF). Then they backtracked to find the most efficient way to generate CO from CO₂.

Earlier work from their group steered them to use a commercially available electrocatalyst made of palladium supported on carbon. Electrocatalysts drive chemical reactions using an electric current. In the presence of flowing electrons and protons, the catalyst splits both CO₂ and water (H₂O) into CO and H₂.

For the second step, the scientists turned to a heat-activated thermocatalyst made of an iron-cobalt alloy. It operates at temperatures around 400 degrees Celsius, significantly milder than a direct CO₂-to-CNF conversion would require. They also discovered that adding a bit of extra metallic cobalt greatly enhances the formation of the carbon nanofibers.

“By coupling electrocatalysis and thermocatalysis, we are using this tandem process to achieve things that cannot be achieved by either process alone,” Chen said.

Catalyst characterization

To discover the details of how these catalysts operate, the scientists conducted a wide range of experiments. These included computational modeling studies, physical and chemical characterization studies at Brookhaven Lab’s National Synchrotron Light Source II (NSLS-II) — using the Quick X-ray Absorption and Scattering (QAS) and Inner-Shell Spectroscopy (ISS) beamlines — and microscopic imaging at the Electron Microscopy facility at the Lab’s Center for Functional Nanomaterials (CFN).

On the modeling front, the scientists used “density functional theory” (DFT) calculations to analyze the atomic arrangements and other characteristics of the catalysts when interacting with the active chemical environment.

“We are looking at the structures to determine what are the stable phases of the catalyst under reaction conditions,” explained study co-author Ping Liu of Brookhaven’s Chemistry Division who led these calculations. “We are looking at active sites and how these sites are bonding with the reaction intermediates. By determining the barriers, or transition states, from one step to another, we learn exactly how the catalyst is functioning during the reaction.”

X-ray diffraction and x-ray absorption experiments at NSLS-II tracked how the catalysts change physically and chemically during the reactions. For example, synchrotron x-rays revealed how the presence of electric current transforms metallic palladium in the catalyst into palladium hydride, a metal that is key to producing both H₂ and CO in the first reaction stage.

For the second stage, “We wanted to know what’s the structure of the iron-cobalt system under reaction conditions and how to optimize the iron-cobalt catalyst,” Xie said. The x-ray experiments confirmed that both an alloy of iron and cobalt plus some extra metallic cobalt are present and needed to convert CO to carbon nanofibers.

“The two work together sequentially,” said Liu, whose DFT calculations helped explain the process.

“According to our study, the cobalt-iron sites in the alloy help to break the C-O bonds of carbon monoxide. That makes atomic carbon available to serve as the source for building carbon nanofibers. Then the extra cobalt is there to facilitate the formation of the C-C bonds that link up the carbon atoms,” she explained.

Recycle-ready, carbon-negative

“Transmission electron microscopy (TEM) analysis conducted at CFN revealed the morphologies, crystal structures, and elemental distributions within the carbon nanofibers both with and without catalysts,” said CFN scientist and study co-author Sooyeon Hwang.

The images show that, as the carbon nanofibers grow, the catalyst gets pushed up and away from the surface. That makes it easy to recycle the catalytic metal, Chen said.

“We use acid to leach the metal out without destroying the carbon nanofiber so we can concentrate the metals and recycle them to be used as a catalyst again,” he said.

This ease of catalyst recycling, commercial availability of the catalysts, and relatively mild reaction conditions for the second reaction all contribute to a favorable assessment of the energy and other costs associated with the process, the researchers said.

“For practical applications, both are really important — the CO₂ footprint analysis and the recyclability of the catalyst,” said Chen. “Our technical results and these other analyses show that this tandem strategy opens a door for decarbonizing CO₂ into valuable solid carbon products while producing renewable H₂.”

If these processes are driven by renewable energy, the results would be truly carbon-negative, opening new opportunities for CO₂ mitigation.

This research was supported by the DOE Office of Science (BES). The DFT calculations were performed using computational resources at CFN and at the National Energy Research Scientific Computing Center (NERSC) at DOE’s Lawrence Berkeley National Laboratory. NSLS-II, CFN, and NERSC are DOE Office of Science user facilities.

“Every now and then we get an exceptional opportunity to glimpse back into deep time,” says first author Ethan Mooney, a paleontology graduate student at the University of Toronto who worked on the project as an undergraduate with paleontologist Robert Reisz at the University of Toronto. “These types of discoveries can really enrich our understanding and perception of these pioneering animals.”

Skin and other soft tissues are rarely fossilized, but the researchers think that skin preservation was possible in this case because of the cave system’s unique features, which included fine clay sediments that slowed decomposition, oil seepage, and a cave environment that was likely an oxygenless environment.

“Animals would have fallen into this cave system during the early Permian and been buried in very fine clay sediments that delayed the decay process,” says Mooney. “But the kicker is that this cave system was also an active oil seepage site during the Permian, and interactions between hydrocarbons in petroleum and tar are likely what allowed this skin to be preserved.”

The skin fossil is tiny — smaller than a fingernail. Microscopic examination undertaken by coauthor Tea Maho of the University of Toronto Mississauga revealed epidermal tissues, a hallmark of the skin of amniotes, the terrestrial vertebrate group that includes reptiles, birds, and mammals and which evolved from amphibian ancestors during the Carboniferous Period. “We were totally shocked by what we saw because it’s completely unlike anything we would have expected,” says Mooney. “Finding such an old skin fossil is an exceptional opportunity to peer into the past and see what the skin of some of these earliest animals may have looked like.”

The skin shares features with ancient and extant reptiles, including a pebbled surface similar to crocodile skin, and hinged regions between epidermal scales that resemble skin structures in snakes and worm lizards. However, because the skin fossil is not associated with a skeleton or any other remains, it is not possible to identify what species of animal or body region the skin belonged to.

The fact that this ancient skin resembles the skin of reptiles alive today shows how important these structures are for survival in terrestrial environments. “The epidermis was a critical feature for vertebrate survival on land,” says Mooney. “It’s a crucial barrier between the internal body processes and the harsh outer environment.”

The researchers say that this skin may represent the ancestral skin structure for terrestrial vertebrates in early amniotes that allowed for the eventual evolution of bird feathers and mammalian hair follicles.

The skin fossil and other specimens were collected by lifelong paleontology enthusiasts Bill and Julie May at Richards Spur, a limestone cave system in Oklahoma that is an active quarry. The unique conditions at Richards Spur preserved many of the oldest examples of early terrestrial animals. The specimens are housed at the Royal Ontario Museum.

The findings include a planet with radius and mass between that of the Earth and Neptune, with a potential orbit around its host star of 146 days. The star system also contains an outer, large companion, 100 times the mass of Jupiter.

This is a rare discovery, with exoplanets smaller and lighter than Neptune and Uranus being notoriously hard to detect, with only a few being identified to this day. Such rare systems are particularly interesting to better understand planetary formation and evolution; they are thought to be a key step for the detection of Earth-like planets around stars.

The new planetary system is discovered around the star HD88986. This star has a similar temperature to the Sun with a slightly larger radius and is bright enough to be seen by keen observers at dark sky sites across the UK, such as Bannau Brycheiniog National Park (Brecon Beacons).

This study, published in the journal Astronomy & Astrophysics, is led by Neda Heidari, an Iranian postdoctoral fellow at the Institut d’astrophysique de Paris (IAP). In the UK, Thomas Wilson, a senior research fellow at the University of Warwick, co-led the analysis of satellite data including searching for new planets. The team also includes researchers at 29 other institutes from nine countries including Switzerland, Chile, and the USA.

A cold, Neptune-like exoplanet

The planetary system includes a cold planet smaller than Neptune, a so-called sub-Neptune, HD88986b. This planet has the longest orbital period (146 days) among known exoplanets smaller than Neptune or Uranus with precise mass measurements.

Neda Heidari, IAP, explained: “Most of the planets we’ve discovered and measured for their mass and radius have short orbits, typically less than 40 days. To provide a comparison with our solar system, even Mercury, the closest planet to the Sun, takes 88 days to complete its orbit. This lack of detection for planets with longer orbits raises challenges in understanding how planets form and evolve in other systems and even in our solar system. HD88986b, with its orbital period of 146 days, potentially has the longest known orbit among the population of small planets with precise measurements.”

HD88986b was detected using the SOPHIE — a high-precision spectrograph (a machine that analyses wavelengths of light from exoplanets) at the Haute-Provence Observatory, France. SOPHIE detects and characterises exoplanets using the ‘radial-velocity method’; measuring tiny motion variations of the star induced by planets orbiting it.

These observations revealed the planet and allowed the team to estimate its mass to approximately 17 times that of the Earth.

Complementary observations obtained with NASA’s space telescope Transiting Exoplanet Survey Satellite (TESS) and the European Space Agency’s (ESA) space telescope CHaracterising ExOPlanet Satellite (CHEOPS) indicate that the planet probably “transits” in front of it host star. This occurs when its orbit passes on the line of sight between the Earth and the star, partially occulting the star — causing a decrease in its brightness that can be observed and quantified.

These observations by both satellites allowed the team to directly estimate the diameter of the planet as about twice that of the Earth. The findings of the study rely on more than 25 years of observations, also including data from ESA’s Gaia satellite and the Keck Telescope in Hawaii.

Moreover, with an atmosphere temperature of only 190 Celsius degrees, HD88986b provides a rare opportunity for studying the composition of the so-called “cold” atmospheres, as most of the detected atmospheres for exoplanets are above 1,000 Celsius degrees.

Due to the wide orbit of the sub-Neptune HD88986b (as large as 60% of the Earth-Sun distance), HD88986b probably underwent rare interactions with other planets that may exist in the planetary system, and weak loss of mass from the strong ultraviolet radiation of the central star. It may therefore have retained its original chemical composition, allowing scientists to explore the possible scenarios for the formation and evolution of this planetary system.

Thomas Wilson, Department of Physics, University of Warwick, said: “HD88986b is essentially a scaled-down Neptune, between the orbits of Mercury and Venus. It becomes one of the best studied small, cold exoplanets paving the way for studying its atmosphere to understand the similarity to our own planet Earth. It also orbits a star with a similar temperature to the Sun making it a precursor to the Earth-like planets to be found by the PLATO space telescope, in which Warwick plays a leading role.”

A second, outer companion

The astronomers also revealed a second, outer companion around the central star. This exoplanet is particularly massive (more than 100 times the mass of Jupiter), and its orbit has a period of several tens of years. Further observations are needed to understand its nature and better determine its properties.

Thomas Wilson added: “We collected data from telescopes pointing at HD88986 for over 25 years making this one of the longest-studies exoplanet systems. This wealth of data revealed a second outer companion more massive than Jupiter that may have been important for the formation of the Neptune-like planet in a similar way to Jupiter in our own Solar System.”