As green solvents,ionic liquids (ILs) are quite suitable for the absorption of volatile organic compounds (VOCs) such as benzene and its homologues.However,solvent selection is the key to the VOC absorption process.In the present study,a rapid solvent screening tool,Conductor-like Screening Model for Real Solvents (COSMO-RS),was used to predict the solubility of toluene in 816 ILs.The effects of four structure characters,namely,the type and alkyl chain length of the cations and anions on the solubility of toluene were discussed.The following conclusions were drawn from the results:(1) ILs with pyrrolidinium-based cations showed better solubility than pyridinium-and imidazolium-based ones.(2) The solubility of toluene in PF6-based ILs increased with the increasing alkyl chain length,while its solubility in Ac-based ILs exhibited the opposite trend.(3) Toluene showed greater solubility in Cl-based ILs than those based on other anions.(4) The solubility of toluene increased with the anion alkyl chain length.Ac-based ILs were chosen as the most promising potential solvents,and further studied to determine the relationship between various interaction energy parameters and toluene solubility.The results showed that the misfit energy played a dominant role during the absorption process.Furthermore,several ILs were selected for experimental verification of the predicted solubility behavior using liquid and gaseous toluene.The results demonstrated that COSMO-RS could be used to semi-quantitatively and qualitatively predict the solubility of toluene,and this model had promising prospects in screening ILs for VOCs absorption.In summary,this study provided a fundamental basis and practical data for the control and treatment of VOCs.