CHENNAI: Researchers at IIT-Madras have developed an artificial intelligence platform to accelerate discovery of high-performance, sustainable metallic alloys for industries, including electric vehicles, aerospace, renewable energy, and marine infrastructure.
The researchers used large language models (LLMs) to extract and organise decades of scientific knowledge from more than 10,000 research papers, creating two databases with over 1,85,000 structured records. The databases are freely available through Alloy Tattvasar and GitHub.
The platform addresses a key challenge in materials research, where valuable experimental information is scattered across scientific papers, tables, and figures. Its automated pipeline extracts alloy compositions, manufacturing processes, testing conditions, and more than 350 material properties, enabling systematic comparison of materials.
"Instead of spending years manually collecting data from thousands of publications, our framework automatically builds structured databases that can be used to identify sustainable materials much faster," said Rohit Batra, assistant professor, Department of Metallurgical and Materials Engineering.
The system also incorporates environmental, economic, and social indicators, enabling researchers to assess alloys on performance as well as sustainability. It uses retrieval-augmented generation to retrieve relevant examples during extraction, improving the accuracy of information drawn from scientific text and tables.
The researchers demonstrated the platform's utility by identifying promising high-entropy alloys for lightweight structural applications in automobiles and aerospace, soft magnetic materials for electric motors and transformers, and corrosion-resistant alloys for marine infrastructure, offshore engineering, chemical processing, and energy systems.
The study was carried out by Aravindan Kamatchi Sundaram, Mohit Chakraborty, Sai Mani Kumar Devathi, and doctoral scholar B Pabitramohan Prusty under Batra's guidance. It received funding from the Anusandhan National Research Foundation, DRDO's Directorate of Industry and Academia and IIT-Madras's Wadhwani School of Data Science and AI, with computational support from the Robert Bosch Centre for Data Science and AI.
Published in Advanced Science, the research could help reduce duplicate experiments, support greener manufacturing and reduce dependence on critical materials.
The team plans to extend the platform to extract information from figures and microstructural images and apply the framework to polymers, ceramics and composites.