A comprehensive back-extraction methodology for the regeneration of ionic liquids in oil extraction desulfurization

Abstract

A comprehensive framework for screening back-extractants was established to regenerate the ionic liquid (IL) used in the ultra-deep desulfurization of oils. The extraction indices of 196 back-extractant candidates were predicted using the COSMO-RS method, aiming at removing the typical aromatic sulfides of thiophene (TS), benzothiophene (BT), dibenzothiophene (DBT), and 4,6-dimethyldibenzothiophene (4,6-DMDBT) from the IL of 1-butyl-3-methylpyridinium dicyanamide ([BMPy][DCA]). The environmentally safety-related phy-chemical properties of these candidates were estimated based on QSAR models. These indices and properties were then statistically integrated into a representative index using the FCE combined with the EWM to identify a high-performance, eco-friendly back-extractant. According to this index, indane was chosen as the optimal back-extractant. Four ternary LLE experiments of the [BMPy][DCA] (1) + TS/BT/DBT/4,6-DMDBT (2) + indane (3) systems were conducted, and the results confirmed the satisfactory selectivities and distribution coefficients of indane. The NRTL model was employed to correlate the experimental data, yielding RMSDs of 0.0007, 0.0011, 4.0147 × 10^–5, and 3.6053 × 10^–5, respectively. The consistency of the obtained model parameters was validated through the Gibbs energy of mixing function analysis. Finally, the IL-regeneration process using indane as the back-extractant was simulated and compared with a CCl₄-based process regarding industrial and economic performance. The results indicated that the indane performed better than CCl₄ in both respects. This study presents a universal framework for screening back-extractants, offering a referable research idea to other separation fields involving IL-regeneration.