Researchers from Kyushu University, in collaboration with Osaka University and the Fine Ceramics Center, have made a significant breakthrough in the field of green energy technology. They have developed a framework that utilizes machine learning to expedite the discovery of materials for sustainable energy applications. This new approach has led to the identification and successful synthesis of two novel candidate materials for use in solid oxide fuel cells, which can generate clean energy without carbon dioxide emissions.
The researchers’ findings, detailed in the journal Advanced Energy Materials, not only have implications for boosting the efficiency of hydrogen fuel cells but also hold the potential to accelerate the search for innovative materials in various sectors beyond energy.
Addressing the urgent need for carbon neutrality, Professor Yoshihiro Yamazaki from Kyushu University’s Department of Materials Science and Technology explained the significance of their research. Creating a hydrogen society is one way to combat climate change, but in addition to optimizing hydrogen production, storage, and transportation, it is crucial to enhance the power-generating efficiency of hydrogen fuel cells.
Solid oxide fuel cells require a solid material, known as an electrolyte, to efficiently conduct hydrogen ions and generate an electric current. To date, research on new electrolyte materials has primarily focused on oxides with specific crystal arrangements, called perovskite structures.
Proton-conducting oxide materials were first discovered in a perovskite structure, and since then, high-performing perovskites are constantly being reported, said Professor Yamazaki. However, we aim to expand the discovery of solid electrolytes to non-perovskite oxides that also exhibit excellent proton conduction capabilities.
Traditional methods of discovering proton-conducting materials with different crystal structures through trial and error have inherent limitations. Adding small amounts of another substance, known as a dopant, to the base material is necessary for an electrolyte to acquire proton conductivity. With numerous candidate base and dopant materials, each possessing varying atomic and electronic properties, finding the optimal combination to enhance proton conductivity becomes a time-consuming task.
To overcome these challenges, the researchers employed machine learning techniques to calculate the properties of different oxides and dopants. By analyzing the data, they identified the key factors influencing proton conductivity and accurately predicted potential combinations.
Guided by these factors, the team synthesized two promising materials, both exhibiting unique crystal structures, and evaluated their ability to conduct protons. Remarkably, both materials demonstrated proton conductivity in just a single experiment.
Of particular interest is one material with a sillenite crystal structure, making it the first-known proton conductor with such a structure. The other material, with a eulytite structure, possesses a distinct high-speed proton conduction pathway not observed in perovskites. Although the current performance of these oxides as electrolytes is low, further exploration and refinement hold the promise of improving their conductivity.
Professor Yamazaki emphasized the potential of their framework in expanding the search space for proton-conducting oxides and thereby accelerating advancements in solid oxide fuel cells. This breakthrough brings us closer to realizing a hydrogen society. Furthermore, with minor modifications, this framework can have applications in other areas of materials science, facilitating the development of various innovative materials.
The discovery of new, efficient, and sustainable materials is crucial for the transition towards greener energy sources. The use of machine learning to expedite the search for these materials represents a significant step forward in the quest for a carbon-neutral future. This research opens up new possibilities for the optimization of hydrogen fuel cells and holds promise for exploring innovative materials in a wide range of industries beyond the energy sector.
