Experimental and data-driven investigation of hydrophilic and hydrophobic ionic liquids for supercapacitors

Ionic liquids (ILs) are promising electrolytes of supercapacitors for high voltage tolerance, zero vapor pressure, excellent thermal stability and environmental friendliness. However, the high viscosity and low ion mobility of ILs limit the capacitance and high-rate performance of the devices. Rather than relying on black-box predictions and screening of advanced ILs for supercapacitors, machine learning models informed by experimentally derived physicochemical parameters can achieve significantly higher accuracy and relevance. According to the comprehensive experimental and data-driven investigation based on electrochemical characterization, nuclear magnetic resonance (NMR) dynamics, quantum and molecular dynamics simulations, we reveal an effective boosting of the specific capacity based on the water solvation mechanism in hydrophilic ILs. We then apply chemistry-informed machine learning to inverse screening and design ILs for supercapacitors based on the critical experimental parameters and morgan fingerprints. These findings elucidate the efficiency and mutual reinforcement of experimental and data-driven investigations in discovering promising materials for energy storage and conversion devices.