DMREF: AI-Accelerated Design of Synthesis Routes for Metastable Materials

点击翻译按钮获取中文摘要

NON-TECHNICAL SUMMARYOne of the current grand challenges in materials science and physics is the control and processing of matter away from equilibrium. This project aims to identify design rules and novel pathways to synthesize desired materials that are metastable and survive for long times at ambient conditions. This research can dramatically expand the materials design space to enable future applications – one of the goals of the Materials Genome Initiative (MGI). A motivating example is the challenge of room-temperature superconductors. That is, recently discovered high-pressure hydrides have reached the longstanding goal of room temperature superconductivity. However, they are difficult to implement in practical technologies because they decompose when they return to ambient pressure. Similar problems are encountered with other materials, such as magnets and superhard systems. In this work, metastable materials will be developed to address this critical need as they offer a promising way forward on this important front.TECHNICAL SUMMARYThe project addresses the science of nonequilibrium processes by searching for and identifying the design rules for synthesis pathways of metastable materials. The effort will establish methods to transform amorphous precursor materials into desired phases that are kinetically stable at ambient conditions, preventing their transformation to the thermodynamic ground state. It will combine materials informatics and machine-learning for data mining and structure prediction with the application of external pressure via diamond anvil cells, temperature via fast laser heating, and high magnetic field to transform amorphous precursor materials into desired phases that are metastable at ambient conditions. The research is expected to lead to new metastable synthesis methods, machine learning approaches for structure prediction and characterization of their thermodynamics and kinetics, and metastable materials with a wide variety of desirable properties such as superconductivity, magnetism, and superhardness. The project will train four junior researchers and teach them scientific techniques and professional skills beyond their research projects through interaction with the broader team. The proposal will also develop and provide in-school testing of a new K-12 experimental kit on crystal growth for a Lending Library of Experiments. The team will train science teachers from across the southeastern U.S. in its use through workshops, and it will make the kits available to them by mail.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.