Extractive Mechanism and Process Design of Hydroxyl-Functionalized Pyridinium-Based Ionic Liquids for Separating m-Cresol and Cumene

Abstract

In this work, we adopt hydroxyl-functionalized pyridinium-based ionic liquids (ILs) as extractants for separating m-cresol and cumene with higher product purity and lower energy consumption. First, a systematic approach combining thermodynamic property calculation and physical property estimation was employed to screen suitable ILs. Second, the extraction conditions, including stirring time, temperature, and IL dosage were optimized to obtain good separation performance of the identified ILs. The recycling performance of identified ILs were further explored. Then, the combination of quantum chemical calculation, wave function analysis, and FT-IR characterization was applied to explore the separation mechanism. Finally, Aspen Plus software was applied to simulate the extraction process of the optimal IL. The results indicate that two pyridinium-based ILs ([HO-C 3 py][SCN] and [HO-C 3 py][N(CN) 2 ]) are identified because of their high separation performance and suitable physical property. The m-cresol extraction efficiencies of [HO-C 3 py][SCN] and [HO-C 3 py][N(CN) 2 ] at optimized extraction conditions are 98.02 and 98.03%, and their cumene entrainments are 11.33 and 14.37%, respectively. Hydrogen bonding interaction with covalent character between pyridinium-based IL and m-cresol is the key driving force in the separation process. The process simulation results show that [HO-C 3 py][SCN] is a promising IL in industrial applications.

A Deep Learning Hybrid Framework Combining an Efficient Evolutionary Algorithm for Complex Many-Objective Optimization of Sustainable Triple CO2 Feed Methanol Production

“…To bridge the aforementioned research gaps, this study proposes a deep learning and evolutionary algorithm-based integrated deep learning hybrid framework, which encompasses various components such as process simulation (simulated by Aspen Plus V12), − data-driven modeling, many-objective optimization, and many-criteria decision-making. Further elucidation of the innovation is provided in the following:…”