Liquid-Liquid Equilibrium of Deep Eutectic Solvent-Aromatic-Aliphatic Ternary Systems: Experimental Study with COSMO Model Predictions

Abstract

Common solvents used for aromatic extraction from aliphatics typically degrade into toxic compounds, while green alternatives perform poorly compared to the state-of-the-art solvents. Deep eutectic solvents (DES) are a novel solvent type made of hydrogen bond donors (HBD) and hydrogen bond acceptors (HBA). DES have been applied in various applications, including advanced separations. In this study, DES were studied experimentally and using the Conductor-like Screening Model (COSMO) to separate benzene from cyclohexane as model compounds for an aromatic:aliphatic system. Both equilibrium and kinetic studies were performed to determine the liquid liquid equilibrium (LLE) and mass transfer rate for the DES-based separation. Selected HBAs including tetrabutylammonium bromide (N4444Br), tetrahexylammonium bromide (N6666Br), choline chloride (ChCl), and methyltriphenylphosphonium bromide (METPB) were paired with HBDs including ethylene glycol (EG) and glycerol (Gly). COSMO was used, with adjustments to reflect DES specific interactions, to predict the liquid-liquid equilibrium (LLE). COSMO results showed that ChCl and N6666Br-based DES extracted too little benzene or too much cyclohexane, respectively, to be considered for experimental evaluation. Overall, the COSMO model predictions for LLE of EG-based DES were very accurate, with root-mean-square deviations (RMSD) below 1% for both N4444Br:EG and METPB:EG. The glycerol systems were less accurately modeled, with RMSD’s of 4% for N4444Br:Gly and 6% for METPB:Gly. The lower accuracy of glycerol system predictions fmay be due to limitations in COSMO for handling glycerol’s influence on polarizability in the DES that is not seen in EG-based DES. Mass transfer kinetics were determined experimentally for DES and the results were fit to a first order kinetics model. METPB:Gly had the highest mass transfer coefficient at 0.180 min−1, followed by N4444Br:EG at 0.143 min−1. N4444Br:Gly and METPB:EG had the lowest mass transfer coefficients at 0.096 min−1 and 0.084 min−1, respectively. It was found that mass transfer rate was not directly related to maximum benzene solubility, as N4444Br:EG and METPB:Gly had the highest and lowest benzene removal, respectively, but had similar mass transfer coefficients.