Turning carbon dioxide into a valuable resource

When Andrew Bocarsly published his first study on using electrochemistry to convert carbon dioxide to something useful, “it generated approximately zero interest,” he says with a chuckle. The year was 1994, and in those days, people didn’t talk about greenhouse gases and climate change, the Princeton University chemist says.

“I used to start every talk explaining in detail why reducing emissions and mitigating CO₂ was a good thing to do, because not everyone in the audience bought into the idea—even as recently as 10 years ago,” he says. Times have changed. “Now it’s a one-sentence throwaway. I get up and say, ‘CO₂ is having a negative impact on the environment, and we really need to do something about it.’ ”

Scientists and nonscientists alike have come to recognize that atmospheric CO₂ levels are rising quickly and that the greenhouse gas is adversely affecting the environment. That change in thinking leaves Bocarsly and other researchers at CO₂ conferences more time to discuss what to do with CO₂ to best address the problem. That’s another area where thinking has changed in a big way, he notes.

Just 3 or 4 years ago, Bocarsly says, there was a lot of tension at CO₂ meetings between researchers who argued for capturing CO₂ and sequestering it deep in the ground and those who said the best way to go is to use it to make chemicals and fuels. He no longer hears those either-or arguments. According to Bocarsly, the consensus among attendees at the 19th International Conference on Carbon Dioxide Utilization at Princeton in June 2022 was that so much CO₂ is in the atmosphere that no single approach—​sequestration or utilization—will solve the problem. “We need to do absolutely everything we can,” he says.

From 1982 to 2022, the average global atmospheric level of CO₂ increased by more than 20%, from approximately 340 ppm to 420 ppm. Most of the increase came from human activities, such as commercial shipping and making iron, steel, and cement. In 2021, human-caused global CO₂ emissions amounted to an estimated 39.3 billion metric tons, according to the World Meteorological Organization.

Pulling a portion of that CO₂ from the atmosphere and reducing it chemically to something valuable would be a big advantage relative to sequestering the gas below ground, says Laura Gagliardi, a computational chemist at the University of Chicago. Making a CO₂-derived polymer, for example, would consume CO₂ that’s already in the atmosphere and spare the environment the additional CO₂ that would be emitted if the polymer were produced in the standard way from petroleum—a win-win situation.

The conversion chemistry can be driven by electricity or heat. Both CO₂-reduction processes would be done in the presence of a catalyst to minimize the required energy input. Electrocatalysis can be greener than thermal catalysis, Gagliardi says. But just like the thinking about sequestering versus using CO₂, “to benefit society and lower greenhouse gas levels, we need to pursue both options,” she says.

The case for electrochemistry

Reducing CO₂ in an electrically driven reactor—an electrochemical cell—offers several advantages over driving the chemistry thermally. Electroreduction can have a smaller carbon footprint because the cells can be powered by renewable electricity, which in the case of wind and photovoltaic power is growing rapidly and is increasingly cost competitive.

“The case for doing it electrochemically is that it’s automatically electrified,” says Ted Sargent, a chemist and electrical engineer at Northwestern University. “Decarbonized energy is going to come in the form of electricity. All the options—solar, wind, hydroelectric, and nuclear—supply energy in that form,” he points out.

In addition, unlike thermal reactors, which typically reduce CO₂ by reacting it with hydrogen at high temperature and high pressure, electrochemical cells generally operate at room temperature and atmospheric pressure, says Peng Zhu, a graduate student working with Rice University electrochemist Haotian Wang.

So the cells can be relatively simple, small, and inexpensive compared with thermal reactors, which need to be large to be cost efficient. Further, the thermal reaction requires heat and a supply of hydrogen gas, both of which usually come from fossil fuel–driven processes that emit a lot of CO₂, though green hydrogen is becoming increasingly available.

The problem with electroreduction is that it suffers from low energy efficiency and insufficient control of the chemistry, which leaves major industrial commercialization years away. Bocarsly knows firsthand how difficult it can be to commercialize scientific ideas. In 2009, he cofounded Liquid Light, a Princeton spin-off that went on to invest $35 million in its technology for electroreducing CO₂. But by 2017, the company had gone quiet, and its assets and patent portfolio were acquired by Avantium.

Despite the setbacks, the electroreduction concept continues to attract new talent. Unlike in the mid-1990s, when few researchers were studying this chemistry, nowadays the field is buzzing with activity: scientists in academia and industry in several countries are scrutinizing every part of the cell. They aim to kick-start commercial development by improving the performance and economics of the process and by tailoring the cells to generate high-​value compounds.

“If we can figure out how to surgically use electrical power to perform reactions, then electrochemical reduction should provide us with the most efficient route for turning CO₂ into valuable products,” Sargent says.

What to make

To make a dent in atmospheric CO₂ levels, researchers set their sights on products such as transportation fuels, which have a large carbon footprint, in part because they are made and consumed in enormous volumes. In 2021, for example, airlines based in the US used roughly 50 billion L of jet fuel, according to the US Department of Transportation.

Much of the required power for passenger cars, delivery trucks, and buses—and perhaps eventually 18-wheelers—can probably be supplied by batteries, says Alex Bell, a specialist in catalysis and sustainable chemistry at the University of California, Berkeley. But for the foreseeable future, the rail, aviation, and maritime shipping industries will depend on hydrocarbon fuels because they supply 30–40 times as much energy by weight as batteries, Bell explains.

Fuels for those industries could be made by converting CO₂ and water in an electrochemical cell to a mixture of carbon monoxide and hydrogen, known as synthesis gas or syngas. Syngas can then be converted readily to a variety of fuels and other compounds via Fischer-Tropsch synthesis, a well-established carbon-​carbon coupling process. That’s the way Berkeley-based Twelve (formerly Opus 12) is gearing up to make most of its products, according to Etosha Cave, the company’s chief science officer.

The start-up, which Cave says has raised $200 million and has more than 200 employees, has partnerships with Alaska Airlines and Microsoft to commercialize its process for making jet fuel and with Virgin Voyages for producing marine fuel. The company has also used CO₂-derived syngas to make polycarbonate sunglass lenses and other polymer products.

Many researchers are making bigger and more valuable molecules—ones with more than one carbon atom—directly in electrochemical cells. “For a long time, we were limited to C₁ products such as CO, methanol, and methane. Now you see a lot of research on C₂ and larger compounds,” Bocarsly says.

Common targets include ethanol and ethylene—C₂ compounds. Sargent explains that his group works on these molecules because ethanol is widely used in automobile fuel. Plus, companies already have methods for turning the alcohol into sustainable aviation fuel. And ethylene, which normally has a large carbon footprint, can be converted to many products, including aviation fuel and polymers.

Brian Seger, a physics professor at the Technical University of Denmark, says that SelectCO2, a European Union–based consortium he coordinates, focuses on the same products for the same reasons. The group includes partners from academia and national laboratories, as well as from large companies such as De Nora, an Italian supplier of equipment and materials for industrial electrochemistry.

Deciding which products to target—molecules with one, two, three, or more carbon atoms—mainly comes down to economics, which includes the price of the electricity, the value of the products, and the efficiency of the process. In the case of ethylene, the energy efficiency of electroreduction is about 25%, Sargent says. For the process to be commercially viable, “we need to be in the 50–60% ballpark.”

Better catalysts to steer the ship

Getting there calls for tweaking every part of the cell.

CO₂ enters an electrochemical cell on the cathode side, where it interacts with a catalyst, often a particulate material supported on that electrode. Most work on electrochemical cells is focused on customizing the catalyst because it’s the secret sauce that kicks off the reaction, controls the energetics, and guides reactants to form products. Small changes in catalyst composition can have a strong effect on cell performance and product distribution.

In one study along those lines, Sargent’s group teamed up with Zachary Ulissi and coworkers at Carnegie Mellon University and used quantum methods coupled with machine learning to search for copper alloy catalysts for making ethylene. Researchers have long known that pure copper outperforms all other single-element catalysts when it comes to making multicarbon products. But perhaps this team could find a bimetallic catalyst that does it even better.

It did. The computations pointed to copper-aluminum alloys, so the team made and tested a series of them. It found that the faradic efficiency—a measure of how well the electrons drive the desired reaction, in this case, CO₂ to ethylene—was 80%. That’s up from 66% for pure copper (Nature 2020, DOI: 10.1038/s41586-020-2242-8).

In a related study, Sargent’s group and scientists at the University of Science and Technology Beijing looked for ways to tailor copper to make catalysts that cause the reduction reaction to favor multicarbon alcohols over ethylene. They found that copper decorated with barium oxide forms ethanol and 1-propanol (a C₃ compound) in a ratio of 3:1, which is 2.5 times as selective as pure copper (Nat. Catal. 2022, DOI: 10.1038/s41929-022-00880-6).

Bocarsly’s group also studies two-​component catalysts, searching for ones that steer the electrochemistry to make relatively large molecules. Through what the Princeton chemist describes as “a series of accidents and mistakes” involving trace-metal impurities, the group discovered that a mixed chromium oxide–​gallium oxide material catalyzes the production of 1-butanol. But to make the C₄ alcohol, the Cr-Ga catalyst has to be spiked with a tiny bit of nickel. Steve Cronin, a postdoctoral scholar working with Bocarsly, presented the unpublished work at the CO₂ meeting at Princeton in 2022.

Design, operation, and durability

Customizing catalysts is one way to improve cell performance. Another way is to redesign the cell. That’s what Zhu, Wang, and coworkers at Rice have been doing.

Liquid products offer advantages over gases because they can be transported and stored more easily and can have higher energy densities, a key consideration for fuels. But liquids generally accumulate in the cell’s electrolyte solution and must be separated and purified, which is expensive. So the Rice team replaced the traditional liquid electrolyte, which transports ions between the cathode and anode, with a solid one—a porous, ion-conducting sulfonated copolymer.

To test the design, the team set up a cell with a bismuth nanoparticle catalyst that converts CO₂ to formic acid, which is used in large volume as a cleaner and in chemical and textile manufacturing. The reaction formed formate and hydrogen ions, which combined in the solid electrolyte, generating formic acid molecules. The team flowed a stream of inert gas through the electrolyte and collected the condensed product in nearly 100% purity (Nat. Commun. 2020, DOI: 10.1038/s41467-020-17403-1).

Meenesh Singh, a chemical engineer at the University of Illinois Chicago, also designs and operates electrochemical cells in unconventional ways. Many researchers in this field use a type of cathode known as a gas-diffusion electrode (GDE), which generally consists of a porous carbon cloth coated on one side with a catalyst. The design of GDEs assists CO₂ in flowing freely toward the catalyst layer, where the gas can react and form products. But in these types of cells, the product stream often contains gaseous molecules such as ethylene mixed with large quantities of unreacted CO₂, necessitating costly purification.

To address the problem, Singh and coworkers designed a liquid-diffusion electrode consisting of a 3D copper mesh with pores up to 100 µm in diameter. The relatively large pores enable a liquid electrolyte saturated with CO₂ to flow freely in the liquid state to the catalyst layer. As gaseous ethylene forms, it automatically separates from the liquid phase, avoiding the need for further purification.

Conventional cells have another shortcoming: the copper catalyst gradually degrades, which leads to poor performance and low selectivity for ethylene.

Singh explains that the active form of the catalyst is a copper oxide. But in a twist of bad luck, the cell voltage required for reducing CO₂ also reduces copper oxide, slowly turning it into catalytically inactive copper metal. Singh’s solution is to oscillate the potential, briefly switching from a small negative voltage, which generates ethylene, to a small positive voltage, which regenerates the oxide catalyst (Cell Rep. Phys. Sci. 2022, DOI: 10.1016/j.xcrp.2022.101053).

Ionic liquid electrolytes

Another component of the cell that may have room for improvement is the electrolyte. Aqueous solutions—alkaline or acidic—are standard. But they’re not the only option. Buxing Han and coworkers at the Chinese Academy of Sciences have evaluated a large number of ionic liquids as electrolytes for the electroreduction of CO₂.

Ionic liquid electrolytes can provide many advantages relative to aqueous and organic electrolytes, Han says. They can have higher electrical conductivity and greater electrical and thermal stability. And they can be used over a wider electrochemical window, or voltage range. Ionic liquids may be more expensive than standard electrolytes, he acknowledges, but they can be used repeatedly and may make product separation easier.

Qinggong Zhu, a researcher who works with Han, points out that CO₂ is highly soluble in imidazolium-based and other types of ionic liquids, which favors high reaction rates. In contrast, CO₂’s solubility in aqueous solutions is quite low, which leads to poor CO₂-conversion efficiency and sluggish kinetics, she says.

Another attractive feature of ionic liquids is that they can be tailored synthetically to customize their properties, says Xinchen Kang, a researcher who works with Han and Zhu. In this case, ionic liquids can be styled to activate CO₂, enhance interactions between the electrode and reaction intermediates, and influence the reaction pathway. Han and coworkers recently compiled an extensive review of ionic liquids used for CO₂ electroreduction (Innovation 2020, DOI: 10.1016/j.xinn.2020.100016).

Large-scale commercialization of CO₂ electroreduction isn’t going to happen overnight. But enthusiasm for the technology is growing quickly. “This is undoubtedly one of the most challenging and promising applications of electrochemistry,” says Anna Ramunni, a research group leader at De Nora. “Once it’s developed, it will be a pillar of decarbonization.”

“Ten to 20 years ago, people were very skeptical that we would ever be able to convert CO₂ to something useful that wasn’t so expensive that no one would be interested in buying,” Bocarsly says. Times have changed. “People are starting to think it’s commercially viable.”

You can quibble about the low efficiency of one multicarbon product or another, he says. “But that means you already know how to make it. There’s no question this is going to happen.”