Fang Liu uses supercomputers to simulate the structures of molecules and the vast array of interactions that can occur during a reaction. The goal is to make predictions about how a molecule will behave under certain conditions.
Emory chemist Fang Liu received $875,000 from the U.S. Department of Energy (DOE) for her work aimed at optimizing the use of light to spark the transfer of an electron. Known as photoredox catalysis, this powerful chemical process is one of the fastest growing areas of organic synthesis, with applications spanning everything from health care to renewable energy.
"Photoredox catalysis has broad applications in organic synthesis and it's also eco-conscious," says Liu, assistant professor of chemistry. "By using light to initiate reactions you can save energy, avoid the use of harsh chemical reagents and the need to dispose of toxic waste."
The five-year award from the DOE's Early Career Research Program will fund Liu's quest to understand how the chemical structures of electron donors and acceptors impact light-driven electron transfer dynamics. That work is a crucial step toward improving the efficiency of photoredox catalysis — and to further expand its capabilities.
A DOE news release describes its early career awards as "critical to longstanding efforts to develop the next generation of STEM leaders to solidify America's role as the driver of science and innovation around the world."
Automating analyses
A theoretical chemist, Liu specializes in computational quantum chemistry, including modeling and deciphering molecular properties and reactions in the solution phase. She uses supercomputers to simulate the structures of molecules and the vast array of interactions that can occur during a reaction. The goal is to make predictions about how a molecule will behave under certain conditions.
In 2022, Liu and her Emory group created an open-source toolkit called AutoSolvate to automate the process of computing molecular properties in the solution phase. AutoSolvate eliminates the painstaking process of requiring a researcher to determine the geometry of a molecule and the location and orientation of the surrounding solvent molecules before running a quantum-chemistry computer program. It also reduces the risk of technical issues that can arise at each step in the process, leading to errors in the results.
The development of AutoSolvate opened the door for Liu's current exploration of photoredox catalysis.
Redox reactions
Molecules suitable for oxidation-reduction, or redox, reactions are those that can easily gain or lose elections during chemical processes. Redox-active molecules are important to everything from the development of anticancer drugs to chemical batteries for renewable-energy storage.
In traditional redox catalysis, a chemical reagent drives a molecule to grab or give out an extra electron, turning it into another oxidation state with the desired reactivity, such as a radical.
Redox reactions can be used to synthesize small molecules or large molecules like polymers — a process known as polymerization. The extra charge can spark a monomer, or a single, short molecule, to connect with another molecule. A chain reaction builds, linking more molecules together into a string or a net configuration, forming a polymer.
Seeing chemistry in a new light
Traditional redox reactions have been a mainstay of chemistry labs for centuries. Around 30 years ago a way to initiate these reactions through light was developed. Only in more recent years, however, has photoredox catalysis become refined enough to make it practical for a wide range of applications.
Dentistry is one field benefitting from photoredox catalysis. The technique allows the use of resin instead of a metal alloy to fill a cavity. Shining an ultraviolet light on the resin after it is injected into the cavity initiates a redox reaction, causing the resin to harden, or polymerize.
The technique is also driving advances in 3D printing for applications in renewable energy engineering and in health care. For example, health care technicians can now input data from a CT scan directly into a computer to generate a 3D model of a patient's joint and generate a custom implant through a photopolymerization process.
Expanding the possibilities
The DOE grant will fund Liu's work to generate the data needed to further optimize photoredox catalysis for polymerization and to keep expanding its capabilities.
Running simulations on a supercomputer will allow Liu and her colleagues to rapidly explore ways to modify the structure of a molecule to more efficiently initiate an electron transfer. They can also run simulations using light with shorter or longer wave lengths. They can adjust the light wavelength to determine the proper amount of energy to not damage the molecule or initiate unwanted side reactions, while optimizing the efficiency of light utilization for the electron transfer.
After accumulating large amounts of simulation data, the researchers will apply machine learning to analyze it, while also verifying the results through experimental work.
"Although for this project we are focusing on a specific polymer reaction, we hope the results we get will yield universal rules for how to design better photoredox catalysts," Liu says.
