Digital Materials Discovery

University of Sheffield Collaborating Faculties: Faculty of Engineering, Royce Institute

Overview: This project uses AI to accelerate the discovery of new materials crucial for green technologies. It explores both data and modelling aspects to predict material composition and properties, through two case studies: permanent magnetic materials and novel corrosion-resistant coatings. The project uses machine learning algorithms for predictions, drawing on both existing and self-generated databases enriched by natural language processing. It addresses inherent challenges like data scarcity and imbalance through data augmentation and the integration of machine learning with physics-based models. The project will deliver high-quality methodologies, open-source code, and a user-friendly interface to broaden the application of these predictive capabilities. This has the potential to revolutionize materials discovery by enabling the software to autonomously predict material properties from specified compositions, paving the way for significant breakthroughs.

Motivation: The demand for sustainable technologies requires the discovery of novel materials with superior properties. Traditional methods for materials discovery are time-consuming and resource-intensive due to their reliance on physical experiments and iterative testing, limited by both practical laboratory setups and human imagination. This project bridges this gap by harnessing the power of AI, using machine learning algorithms, digital databases, and computational tools. This significantly reduces the time it takes to translate a concept into a material with desired properties. By enabling autonomous prediction of material properties, this project has the potential to revolutionize the field of materials science. In the short term, it can accelerate the discovery cycle, leading to faster development of solar cells and energy-efficient materials. In the long term, this project could pave the way for entirely new materials with unforeseen properties, ultimately propelling advancements across diverse fields like aerospace, medicine, and electronics.