Researchers at Stony Brook University have developed a computational method that uses clean electricity to identify the best liquid solvents for dissolving carbon dioxide. The goal is to make CO2 electroreduction more efficient, turning captured emissions into fuels and products like carbon monoxide, ethylene, and ethanol. The team screened 1.3 million candidate molecules using physics simulations and machine learning, narrowing the list to six promising solvents that had never been tested for this purpose. They also published their data in an open access database called COSMIC so other researchers can build on the work. This approach avoids years of trial and error in the lab. If the solvents perform well in real devices, it could speed up development of carbon utilization technology that runs on renewable electricity.
