Rational design and experimental evaluation of novel amino acid-based natural deep eutectic solvents for CO2 capture

The inherent merits of natural deep eutectic solvents (NADESs) have endowed them as a class of promising solvents in many applications. In this study, amino acid-based DESs are applied as absorbent for CO 2 capture. A systematic method combining solid–liquid equilibria (SLE) prediction, EHS-impacts (environmental, health, and safety) evaluation of the second components, and experimental validation is employed to rationally identify promising CO 2 absorbents. The melting point and fusion enthalpy of pure amino acid components are predicted by a recently proposed machine learning model, while COSMO-RS is used to predict the SLE of all possible combinations. Five key EHS-related properties of the second components are evaluated via the VEGA platform. From above, eight promising binary systems based on L-lysine and L-proline are screened from the initial pool of 2,500 combinations. The SLE, density, and viscosity are experimentally determined and the CO 2 absorption experiments are finally performed, identifying L-proline: nonanoic acid (1:4) as the most promising eutectic absorbent.