D3SC: Developing Data-Driven, Automated Methodology to Understand and Control Light-Driven Catalytic Processes

点击翻译按钮获取中文摘要

With the support of the Chemical Catalysis program in the Division of Chemistry, Professor Stefan Bernhard of Carnegie Mellon University will study catalytic processes that are driven by light. Continued development of a sustainable society requires advances in molecular sciences that will pave the way to the solar-powered, efficient, safe and precise molecular engineering of new functional molecules, materials, and fuels. One of the major obstacles on this path stems from the inherent difficulty of extending current chemical theories to accurately describe the interplay of matter and energy in light-driven chemical transformations. The proposed work utilizes an investigative approach that seeks to accelerate chemical discovery and enhance control of chemical reactions by developing new tools integrating high-throughput experiments, quantum chemistry, and data science. The Bernhard lab involves a diverse group of researchers to accomplish these research tasks. Stefan Bernhard and his research team will be involved in outreach activities for non-scientists to educate the public on energy issues and instill excitement for science, in general.With the support of the Chemical Catalysis program in the Division of Chemistry, Professor Stefan Bernhard of Carnegie Mellon University will study the complex interplay of reagents and catalysts in light-driven chemical processes. The funded research will create a deeper understanding by using new tools integrating high-throughput experiments, quantum chemistry, and data science. The created artificial intelligence resulting from a merger of chemical theory and chemical data will allow the capture of the effects of complex many-body electronic interactions that elude current algorithms. To gather the data needed to drive this effort, the Bernhard team will develop massively parallel, automated chemical reactors capable of sensing and recording the progress on a variety of reaction types in parallel and in real time. Increasingly complex photochemical processes ranging from simple energy transfers to light-driven polymerizations will be investigated and their observed reactivity patterns will be modeled, compared and analyzed. The initial phase of the work will focus on photocatalytic reactions catalyzed by iridium complexes. These are expected to provide an excellent trial system due to the ability to easily synthesize structures that span a vast chemical space, but where the structure-activity relationships have been difficult to model and predict. Later efforts will increase the chemical and structural diversity of the photocatalyst by inclusion of a much wider variety of ligand frameworks and involving central ions with a range of electronic structures. Quantum chemical calculations on the photocatalysts will be the basis of a feature learning-based approach to find and predict structure property/(re)activity relationships in the studied photochemical systems.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.