Kyushu University, in collaboration with Osaka University and the Fine Ceramics Center, has harnessed the power of machine learning to accelerate the discovery of green energy materials. By utilizing this innovative approach, the scientists have successfully identified two materials for solid oxide fuel cells, a technology that enables carbon dioxide-free energy generation using hydrogen fuels.
In the quest to combat climate change, researchers worldwide are striving to develop energy solutions that reduce reliance on fossil fuels. Professor Yoshihiro Yamazaki from Kyushu University’s Department of Materials Science and Technology, Platform of Inter-/Transdisciplinary Energy Research (Q-PIT), emphasizes the importance of creating a hydrogen society as part of the path toward carbon neutrality. He notes that apart from optimizing the production, storage, and transportation of hydrogen, it is crucial to enhance the power-generating efficiency of hydrogen fuel cells.
For solid oxide fuel cells to generate electricity, the efficient conduction of hydrogen ions (protons) through a solid material, known as an electrolyte, is essential. Until now, the research has primarily focused on oxides with perovskite crystal structures as potential electrolyte materials. However, Professor Yamazaki and his team sought to expand the search for solid electrolytes beyond perovskite oxides in order to discover alternative crystal structures with efficient proton-conducting capabilities.
The traditional trial-and-error approach to finding proton-conducting materials with different crystal structures has its limitations. Typically, small amounts of a dopant, an additional substance, need to be introduced to the base material to impart proton conductivity. With numerous base and dopant candidates, each possessing unique atomic and electronic properties, identifying the optimal combination can be a time-consuming and arduous process.
In light of these challenges, the researchers synthesized two promising materials, each with its own distinct crystal structure. The conductivity of protons in these materials was then assessed through experiments. Surprisingly, both materials exhibited proton conductivity after just a single experiment.
This breakthrough in material discovery showcases the potential of machine learning to expedite the development of eco-friendly technologies. By leveraging this technology, scientists can efficiently explore a wider range of material compositions and crystal structures, enhancing the overall progress in green energy research.
The study conducted by the scientific team from Kyushu University, Osaka University, and the Fine Ceramics Center has been published in the Advanced Energy Materials journal. This research not only accelerates the search for efficient proton-conducting materials but also contributes significantly to the advancement of solid oxide fuel cells, offering a promising avenue for sustainable energy generation without carbon dioxide emissions.
As the world continues to combat climate change, such groundbreaking discoveries and technological advancements will play a pivotal role in transitioning toward a greener and more sustainable future. The integration of machine learning and materials science offers immense potential for expedited innovation, propelling the development of eco-friendly technologies that can drive us closer to a carbon-neutral society.
