Scientists at Kyushu University, in collaboration with Osaka University and the Center for Fine Ceramics, have developed a system that uses machine learning to accelerate the discovery of materials important to green energy technology.

Accelerated Discovery of Environmentally Friendly Energy Materials Using a Machine Learning ApproachUsing this innovative method, the researchers successfully identified and synthesized two new candidate materials suitable for use in solid oxide fuel cells. These devices can generate energy using environmentally friendly fuels such as hydrogen, which do not emit carbon dioxide. The results of their research, documented in the journal Advanced Energy Materials, can not only accelerate the search for revolutionary materials in the field of energy but also potentially extend to other areas of innovation.

In response to the escalating global warming crisis, scientists are actively exploring alternative methods of energy production that do not depend on fossil fuels. Speaking about the need for carbon neutrality, Professor Yoshihiro Yamazaki of Kyushu University’s Department of Materials Science and Technology, Interdisciplinary Energy Research Platform (Q-PIT), states: “Creating a hydrogen society is the way to carbon neutrality. while optimizing the production, storage, and transportation of hydrogen, we also need to improve the efficiency of hydrogen fuel cells in energy production.”

Solid oxide fuel cells require efficient conduction of hydrogen ions (protons) through a solid material known as an electrolyte to generate an electrical current. While existing research has mostly focused on oxides with perovskite structures, the researchers aimed to expand the search to non-perovskite oxides with similar proton conduction capabilities. Traditional trial-and-error methods for identifying proton-conducting materials with alternative crystal structures have limitations. To overcome these problems, the researchers used machine learning to analyze the properties of various oxides and dopants. After identifying the key factors affecting proton conductivity, the team predicted possible combinations.

Following these predictions, the researchers successfully synthesized two materials with unique crystal structures, both of which exhibit proton conductivity in the same experiment. Notably, one material had a selenite crystal structure, representing the first known proton conductor with such a formation. Another, with a eulithite structure, demonstrated a distinct high-speed proton conduction pathway compared to perovskites. Although the current performance of these oxide electrolytes is modest, the research team expects improvements through further research.

Professor Yamazaki concludes: “Our structure has the potential to greatly expand the search for proton-conducting oxides, thereby accelerating progress in solid oxide fuel cells—a promising step towards the realization of a hydrogen society.” In addition, the adaptive nature of this structure implies its potential application in various areas of materials science, potentially accelerating the development of a variety of innovative materials with minor adjustments.

Other posts

  • AI-Powered Human Resources
  • Emotion Recognition with Machine Learning
  • Google Introduces The Open Source Framework Firebase Genkit
  • AI Innovations Presented at Google I/O
  • Researching Genetic Algorithms
  • Yelp Introduces an Innovative AI Assistant to Facilitate Business Connections
  • The Intersection of AI and Machine Learning in Financial Services
  • Anomaly Detection Using Machine Learning Algorithms
  • Meta Debuts The Latest AI Chip In A Competitive Technology Sprint
  • An Innovative Machine Learning Approach for Analyzing Material Surfaces