CAREER: Many-Body Green's Function Framework for Materials Spectroscopy

With support from the Chemical Theory, Models and Computational Methods (CTMC) program in the Division of Chemistry, Dr. Tianyu Zhu of Yale University is developing high accuracy theoretical methods for simulating spectroscopic properties of solid state materials. Computational modeling of how light interacts with materials is important for advancing technological applications in optoelectronics design, solar energy conversion, catalysis, and in semiconductor development. However, current computational tools have limited accuracy and efficiency for investigating large scale many-electron materials, hindering our capability to tune and control their electronic properties and chemical reactivity. Dr. Zhu and his group will develop and leverage new ideas in quantum chemistry, condensed matter physics, and data science, to create a reliable and efficient toolbox for modeling light-matter interactions in complex materials. These new methods will be incorporated into the open-source PySCF software package to benefit the broader scientific community. Through this program, Dr. Zhu and his team will develop a hands-on computer game demonstration of organic light-emitting materials design through Yale University’s outreach programs for K-12 students. He will also create a summer computational chemistry workshop and summer research internships targeting underrepresented high school students, as well as organize a guest lecture series to demystify computational chemistry for undergraduate students in chemistry. This research is directed at developing a workable many body Green’s function based on electronic structure methods for simulating charged and excitonic excitations in condensed matter systems, which is crucial for understanding electron correlation physics and energy transfer dynamics in materials. Dr. Zhu and his team will formulate a Green’s function quantum embedding method that enables the use of correlated excited state quantum chemistry tools in simulating photoemission spectra of extended systems, such as at the level of coupled cluster and multi-reference theories. A two particle extension of this method will be further developed to capture electron-hole interactions in describing optical spectra. In addition, the Zhu group will develop a machine learning approach to enable highly efficient, many body Green’s function calculations of molecules and materials. Adopting the established framework, systematic benchmarks on the accuracy of excited-state quantum chemistry methods in predicting valence excitations in weakly and strongly correlated electron materials will be pursued. 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.