Laboratory for Energy and Sustainability (LENS)

As numerous disasters such as wildfires, floods, droughts and heatwaves are occurred, the world attempts to achieve carbon emission free society. Hence, chemical production industries also make effort to reduce CO2 emission. Electrochemical conversion of CO2 into value-added products such as carbon monoxide, ethylene and ethanol could reduce the carbon emission compared to fossil-fuel based chemical production. When electrochemical conversion of CO2 is powered by renewable energy, it could be negative carbon emissions. Therefore, the research about catalysts, systems and theoretical studies of electrochemical conversion for CO2 is continuously increasing. AI is redefining the paradigm of electrochemical research by extending the field into domains beyond the limits of conventional methodologies, while accelerating the discovery–validation cycle to substantially improve research efficiency. 1. Experimental design — Literature- and data-driven hypothesis generation, Bayesian optimization of experimental conditions, and AI-assisted catalyst screening enable the efficient identification of promising catalyst candidates and optimized reaction parameters. 2. Automated experimentation — Robotic-hardware-based automation of electrochemical experiments enables the rapid and reproducible generation of high-quality, large-scale datasets. 3. Automated data analysis — An AI-agent-based automated data processing pipeline enables rapid refinement and analysis of large experimental datasets. 4. Computational simulation — AI-assisted DFT-based mechanism elucidation and key-descriptor exploration identify the factors governing activity and selectivity.