Collaborative Research: DMREF: Machine Learning Algorithm Prediction and Synthesis of Next Generation Superhard Functional Materials

NON-TECHNICAL SUMMARY The goal of this project is to discover new materials that possess the properties needed to enable the technologies of the future. Many materials have outstanding properties that make them desirable for some applications but are deficient in other properties that limit their use. A well-known example is diamond, which is the hardest known material but is also an electrical insulator. Is there a material yet to be discovered that could satisfy the need for a superhard material that also has the high electrical conductivity of a metal or other useful properties? This project will combine diverse areas of expertise to search for new superhard materials that also possess other desirable properties that enable them to fulfill uniquely demanding technological requirements. Both three-dimensional and two-dimensional forms of these materials will be synthesized. A feedback loop between experiment and theory will be used to characterize the materials, rationally design those with desired properties, and optimize the synthesis protocols. Students will be trained in an interdisciplinary collaborative team of theoreticians and experimentalists whose expertise includes chemistry, physics, and materials science and engineering. TECHNICAL SUMMARY This project will combine newly developed crystal structure prediction with experiments for the discovery and synthesis of novel superhard materials, including those with additional functionality. The theoretical effort will pioneer a novel method for crystal structure calculations that merges a priori evolutionary algorithms with machine learning techniques. The experimental effort will employ high pressure-temperature synthesis methods along with techniques such as chemical vapor deposition to create metastable materials starting with thin films. The project will focus on the creation of new materials consisting of the light elements boron, carbon, and nitrogen, either alone or combined with heavier metallic elements. The new materials will be characterized by a variety of methods, including advanced spectroscopic techniques such as those now available at synchrotron facilities. The theoretical techniques will guide the choice of target materials for synthesis and the most promising precursor materials and synthesis pathways. The developed computational tools will be made available under open source licenses, thereby contributing towards cyberinfrastructure. Student training in theory and experiment will advance STEM initiatives through summer research experiences and undergraduate course development, including those targeting underrepresented groups. 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.