In this study, novel molecular structure encoding descriptors composed of feature encoding and one‐hot encoding was developed and then convolutional autoencoder was used to denoise based on the structure of ionic liquids (ILs). It could be used to predict the CO2 solubility in ILs at different temperatures and pressures, when combined with three different machine learning algorithms (multilayer perceptron [MLP], random forest [RF], and support vector machine [SVM]). Statistics of the prediction results show that the newly proposed molecular structure‐based coding has better regression prediction performance than the conventional molecular cheminformatics descriptors. SE‐MLP model with R2 of 0.9873 and mean square error of 0.0007 has the best performance in predicting the CO2 solubility in ILs. In addition, the relationship between features and dissolved CO2 capacity was analyzed through model interpretation to retrieve physical insights for the underlying system. This work provided a new predictive tool for enriching and refining data on CO2 solubility in ILs and for solving phase equilibrium problems.

No abstract

No abstract

Ionic liquids (ILs) have been proposed as promising green solvents for the separation of azeotropes. In this work, the efficient separation of methyl tert-butyl ether (MTBE) from gasoline model oil using ILs was investigated from the molecular level (i.e., extraction mechanism) to process system integration (i.e., process simulation). The quantum chemistry (energy analysis and weak interaction) and molecular dynamics were implemented to explore the separation mechanism. Compared to benchmark organic solvents, 1-butyl-3-methylimidazolium nitrate ([BMIM]-[NO 3 ]) and 1-ethanyl-3-methylimidazole hydrogen sulfate ([EMIM][HSO 4 ]) had the highest extraction rate, and [NO 3 ] − played a major role in the separation process. The extraction performance of MTBE from gasoline model oil using ILs was evaluated. [BMIM][NO 3 ] had the highest extraction effect at 298.15 K. The efficient process of liquid−liquid extraction and extractive distillation with [BMIM][NO 3 ] as an extractant were designed, which confirmed that ILs are promising extractants. This work provides a theoretical guideline to search and develop task-specific ILs for efficient recovery of MTBE.

No abstract