Title: Materials design and discovery of high-melting-point materials from ab initio and deep learning
Abstract: High-performance refractory materials play an important role in applications ranging from gas turbines to heat shields for hypersonic vehicles. In search of high-melting-point materials, Hong develops computational methods that are both accurate and fast via ab initio and deep learning.
Hong has built an accurate and cost-effective method for ab initio melting temperature calculation. The method is implemented in the SLUSCHI package, an automated tool for DFT melting point calculation. Employing the method and the tool, Hong discovered the material with the world’s highest melting temperature and dozens of refractory materials of various types. Next, Hong constructs a melting temperature database that contains thousands of high-melting-point materials from both experiment and computation. Based on the database, Hong then builds a graph neural networks model that rapidly predicts melting temperature from the input of only chemical formula.
The ab initio package and the graph neural networks model are publicly available. The combination and complement of an accurate ab initio method and a rapid deep learning model accelerate and facilitate computational materials design and discovery of high-melting-point materials.
Bio: Qi-Jun Hong is an Assistant Professor of materials science and engineering at Arizona State University. Hong received his Ph.D. in Chemistry titled “Methods for melting point calculations” from the California Institute of Technology in 2015. Before joining ASU, Hong was a Post-doctoral Research Associate at Brown University and then an Applied Scientist in machine learning at Amazon.com, Inc.