Innovative Process Boosts Clean Hydrogen Production

(Stanford Report) Using a novel heating method that cuts carbon emissions, engineers dramatically improved hydrogen production efficiency. The process also comes with a surprising side benefit.

Producing hydrogen is essential for some of the world’s biggest industrial and agricultural needs, but it also comes with a steep environmental cost: most hydrogen made today is generated through processes that release large amounts of CO2. This is due to a combination of high-heat requirements and direct emissions resulting from the process. Now, Stanford researchers have developed a new, highly efficient way to make hydrogen that drastically reduces CO2 emissions and comes with an unexpected and potentially valuable byproduct.

The method, described in a new study published in Science, uses a process called methane pyrolysis, which breaks methane down into hydrogen gas and solid carbon rather than the gaseous CO2 released by conventional hydrogen production.

“Today we take natural gas and produce hydrogen from it, but you’re also producing CO2,” said co-first author Henry Moise, a researcher in Stanford’s Department of Chemical Engineering in the School of Engineering. “Pyrolysis is a similar process, but instead of producing CO2, you produce solid carbon.”

Scaling challenges

Methane pyrolysis has been studied for years, but researchers have struggled to move it beyond the lab due to two major obstacles: removing the solid carbon from the reactor and getting enough heat in the reactor to carry out the process. The new paper, co-first-authored by Moise and German visiting student Sebastian Moll, focuses on solving the latter challenge.

“If you want to do pyrolysis at the scale required to fulfill hydrogen markets, then you have to create very large reactors,” said senior author Matteo Cargnello, an associate professor of chemical engineering at Stanford Engineering. “To heat up these reactors to very high temperatures, like 1,000 degrees Celsius, your heating methods have to be very efficient.”

Today’s industrial reactors are heated from the outside, which becomes a problem at scale. “You can imagine that once you heat something really big from the outside, it’s hard for the heat to penetrate all the way to the middle of the reactor,” Cargnello said.

To tackle the issue, Moise and his colleagues developed a way to heat the reactor from within. Rather than burning natural gas, which produces CO2, the team found that placing a burner inside the reactor to selectively combust a portion of hydrogen produces mostly water, avoiding direct carbon emissions while efficiently driving the reaction forward.

The payoff, according to Moise, is a roughly tenfold gain in efficiency. In other words, using the same amount of energy, a reactor using this new method of autothermal heating produces the same amount of material as 10 other reactors not using conventional external heating methods.

“It goes back to this idea of efficiency. You can reduce CO2 emissions by avoiding making them in the first place, but you can also reduce CO2 emissions by being more efficient about how you use your energy,” Moise said. “So there’s multiple ways to make a more sustainable process.”

Unexpected benefits

Beyond the gains in efficiency, the researchers found an unexpected benefit to the new process: the solid carbon produced by the reaction turned out to be high-quality graphite, a material used in batteries, electrodes, and other technologies.

“It’s not that we wouldn’t anticipate some higher quality carbon, but it was just such a high degree of graphitization and high-quality carbon,” Moise said, calling it the biggest surprise of the project.

Cargnello underscored that the unexpected quality of the carbon byproduct could offer a path toward domestic production of a material the U.S. currently imports from abroad. At the same time, he cautioned that the graphite produced isn’t pure enough yet for demanding applications like battery materials. “This is a big step forward,” he said, “but there are still other steps and more research that needs to be done.”

The researchers underscored the importance of developing sustainable processes for producing hydrogen, a foundational molecule that most people don’t realize they rely on. “At least a few percentage points of the GDP depend on hydrogen, and we’re going to keep making it whether it’s dirty or not,” Moise added.

He pointed to fertilizer as one example: ammonia production depends on hydrogen, and ammonia-based fertilizer is estimated to help feed roughly half the world’s population. Hydrogen also helps remove sulfur from gasoline during refining and helps produce important products like methanol. “It’s so ubiquitous, it kind of becomes invisible,” he said.

The researchers hope the new process will unlock the possibility of having cheap and clean hydrogen production and bring sustainability to a product that underpins many important products used in everyday life.

Cargnello emphasized that the project was a collaborative effort, highlighting contributions from Eric McFarland’s team at the University of California, Santa Barbara, who provided data on larger-scale reactors; co-author Arun Majumdar for encouraging the team to pursue the problem; and Moll, who spent six months in the lab working alongside Moise. Additionally, the multiple funding sources were foundational for the success of the project, including the Kavli Foundation, the Carbon Hub at Rice University, the Natural Gas Initiative at Stanford, and the CO2 Research Center at Aarhus University in Denmark.

The need for collaboration – and funding – continues, added Cargnello, as the next steps will be big ones, including scaling up the research and verifying its use for mainstream hydrogen production.

For more information

Additional Stanford co-authors include PhD students Joshua Martinez-Navarro and Sai Varanasi, and former postdoctoral scholar Kun Xu. Cargnello is also a senior fellow at the Precourt Institute for Energy and a member of Bio-X. Arun Majumdar is the Chester Naramore Dean of the Stanford Doerr School of Sustainability and the Jay Precourt Provostial Professor. He is also a professor of mechanical engineering in the School of Engineering and of energy science and engineering in the Doerr School of Sustainability, and a professor of photon science at SLAC, and a senior fellow at the Stanford Woods Institute for the Environment and at the Precourt Institute for Energy.

Additional authors are from the University of California, Santa Barbara, and Karlsruhe Institute of Technology in Germany.

This research was primarily supported by the Kavli Foundation, by the Carbon Hub at Rice University, and by the Natural Gas Initiative at Stanford. Additional funding was provided by the Novo Nordisk Foundation and CZero Inc. Part of this work was performed at nano@stanford. The Natural Gas Initiative is part of the Precourt Institute for Energy’s Center for Fuels of the Future, an industrial affiliate program. Stanford industrial affiliate programs are funded by membership fees from companies. The Precourt Institute for Energy is in the Stanford Doerr School of Sustainability. View current Stanford Doerr School of Sustainability affiliate members.

Competing interests: A provisional patent with the findings reported in this work was filed by Stanford University. One Author (EM) has a patent in autothermal pyrolysis (WO/2025/175109) and a financial interest in one of the funding agents, CZero Inc. READ MORE