A Quest To Find a New Reactant To Replace Ethylene Oxide in Reactive–Extractive Distillation for Ternary Azeotropic Separation

Reactive–extractive distillation (RED) is a hybrid process that combines reaction-driven and entrainer-based separations into a single unit. Over the years, it has been widely studied for separating water-containing ternary azeotropic mixtures, with all studies utilizing ethylene oxide (EO) as the reactant. However, EO is highly flammable, which poses a significant technical challenge for large-scale commercialization of the process. This study investigated the potential of glycidol as a new reactant in RED systems, offering an alternative to the commonly used EO. Glycidol is an organic compound having a molecule containing both epoxide and alcohol functional groups, and thus, it reacts with water to produce glycerol. Glycerol has been employed as an entrainer in various extractive distillation (ED) processes, particularly for the recovery of tetrahydrofuran (THF) and methanol (MeOH) from an azeotropic mixture of THF/MeOH/water, which served as the case study in this work. In this study, glycidol was utilized as the reactant in a double-column RED (DCRED) system to recover THF and MeOH from the wastewater. This configuration was evaluated and compared to triple-column ED (TCED) systems using glycerol and dimethyl sulfoxide (DMSO) as entrainers. To ensure a fair comparison, the design parameters of all configurations were optimized by using a genetic algorithm to maximize the total net revenue (TNR). The results demonstrated the superiority of DCRED with glycidol, achieving a total annual cost that was 25.3 and 17.3% lower than those of TCED with glycerol and TCED with DMSO, respectively. Additionally, DCRED with glycidol reached the highest TNR with the value of 9.371 × 108 $ a–1, which was 4.5 and 4.49% higher than those of TCED with glycerol and TCED with DMSO, respectively.