Abstract: The high-effect wastewater purification has become a serious problem for acetate ester production. The traditional technology for the separation of acetic acid and water including extractive distillation, azeotropic distillation and other methods, which is low efficiency in the selection of entrainer. Conductor-like Screening Model -Real Solvents (COSMO-RS) model can predict the thermodynamic properties of the fluid through quantitative calculation, which is helpful to select the high-effect entrainer for the acetic acid+water system, then simplifying operation and improving efficiency in the research. The latest research shows that organic solvents containing nitrogen and phosphorus have a good extraction effect on the water+acetic acid system. In this study, COSMO-RS model was applied to screen the high-effect entrainer for the separation of water+acetic acid system in extractive distillation, and the high-effect entrainer were specified in the ionic liquids composed of 5 phosphorus anions and 12 imidazole cations. The geometric structures of ions which were not included in COSMO software's original database were optimized by TURBOMOLE quantitative calculation module, and the solvent capacities of water in different ionic liquids, the selectivity of acetic acid to water in different ionic liquids, and the excess enthalpies when ionic liquids were mixed with water or acetic acid were analyzed. The results showed that the extraction effects of [MIM][DBP], [ODMIM][DEP], and [ODMIM][Me2PO4] were better than others. When water or acetic acid was mixed with ionic liquids, hydrogen bonding dominated the intermolecular interaction absolutely. Imidazole cations had strong ability to provide hydrogen bonding, and was easy to combine with acetic acid, so as to improve the relative volatility of water and acetic acid and achieve the purpose of separation. Comparing the excess enthalpy value, the interaction between [ODMIM]+ cation and acetic acid was the strongest intermolecular force. Finally, [ODMIM][DEP] was selected as the high-effective entrainer for the separation of water+acetic acid system. The results provide basic data for the separation of components in water+acetic acid system.

Key words: COSMO-RS model, glacial acetic acid, ionic liquid, extractive distillation

摘要： 乙酸酯类产品工业生产中产生的含酸废水的高效分离是化工生产中的难题，传统分离乙酸与水的技术如萃取精馏、共沸精馏等在溶剂选择时效率低。类导体屏蔽-真实溶剂模型(Conductor-like Screening Model-Real Solvents, COSMO-RS)可以通过量化计算预测流体热力学性质，从大量溶剂中筛选出适用于某一体系的夹带剂，可减少实验量，提高研究效率。含氮类与含磷类的有机溶剂对水+乙酸体系的萃取效果较好，因此，本工作使用COSMO-RS模型从5种含磷阴离子和12种咪唑类阳离子组成的离子液体中筛选出适用于萃取精馏分离水+乙酸体系的夹带剂。优化了未收录于COSMO软件原始数据库的离子的几何结构，通过σ-Profile图、水在不同离子液体中的溶解度、不同离子液体中乙酸对水的选择性、离子液体与水或乙酸混合时的超额焓四个方面进行分析。结果表明，[MIM][DBP], [ODMIM][DEP]和[ODMIM][Me2PO4]的萃取效果较好。水或乙酸与离子液体混合时氢键作用占据绝对主导，咪唑类阳离子提供氢键的能力强，易与乙酸结合，提高水与乙酸的相对挥发度，达到分离的目的。对比超额焓数值，[ODMIM]+与乙酸之间的相互作用最强，最终确定[ODMIM][DEP]离子液体为分离水+乙酸体系首选的夹带剂。研究结果为萃取精馏实验验证及水+乙酸体系的工业化分离提供了基础数据。

关键词: COSMO-RS模型, 乙酸, 离子液体, 萃取精馏