Researchers from the University of Toronto have used artificial intelligence (AI) to generate a “recipe” for an exciting new catalyst needed to produce green hydrogen fuel. As the effects of climate change become increasingly apparent, research like this could open the door to green hydrogen fuel, which could be used for transportation, residential, and commercial heating. Producing hydrogen fuel involves a complicated process.
Scientists take water and pass electricity from renewable sources through metal electrodes submerged in the water. These electrodes are coated in a catalyst that speeds up the splitting of water into hydrogen gas and oxygen gas, with the hydrogen gas subsequently used for fuel. Historically, iridium oxide was the preferred catalyst due to its ability to withstand the harsh acidic conditions of this reaction.
However, iridium is both scarce and expensive, making it an unsustainable source for large-scale hydrogen production. Ruthenium-based catalysts emerged as an alternative due to their abundance and lower cost, but they were found to be unstable during the reaction. To address this issue, the University of Toronto team created an AI program to find an optimal alloy combination for the catalyst more quickly.
This program analyzed over 36,000 different metal oxide combinations through virtual simulations—a task that would have been prohibitively time-consuming using traditional trial-and-error lab methods.
AI expedites green hydrogen catalyst discovery
The Canadian Light Source (CLS) at the University of Saskatchewan played a key role in this research, allowing scientists to examine material properties that are not visible through regular microscopes.
“We’re talking about hundreds of millions or billions of alloy candidates, and one of them could be the right answer,” mentioned Jehad Abed, a member of the AI development team. After processing the data, the AI program recommended a combination of ruthenium, chromium, and titanium. Testing in the lab showed that this new catalyst was effective.
Using the CLS, the team observed how this new alloy works, gaining insights into its atomic arrangement and behavior during the water-splitting reaction. Most importantly, they found that the new catalyst significantly reduces the risk of ruthenium dissolution and maintains its structural integrity better than other candidates. The new alloy showed remarkable performance, being 20 times more stable and durable than benchmark metals.
“The computer’s recommended alloy performed 20 times better than our benchmark metal in terms of stability and durability,” said Abed. While the research marks a major success, there is still much work to be done before this new ruthenium, chromium, and titanium alloy can be used for large-scale hydrogen production. Extensive testing under real-world conditions will be necessary.
This work exemplifies how AI can expedite the search for solutions to complex climate-related challenges.
