Prediction of Phase Behavior of CO₂ Absorbents Using Conductor-like Screening Model for Real Solvents (COSMO-RS): An Approach to Identify Phase Separation Solvents of Amine/Ether/Water Systems upon CO₂ Absorption

Abstract

Developing energy-saving absorbents for carbon dioxide (CO 2 ) is essential for improving carbon capture and storage (CCS) technologies. Recently, we have designed phase separation solvents, which can significantly reduce the regeneration energy for CO 2 capture and separation down to 1.6 GJ/ton-CO 2 et al. Int. J. Greenhouse Gas Control2018, 75, 1−7]. For further developing better solvents, this paper studied a theoretical approach with conductor-like screening model for real solvents (COSMO-RS) to screen the amine/ether/water systems to identify phase separation solvents upon CO 2 absorption. In this work, liquid−liquid equilibria of 21 amine/ether/water systems were determined both before and after CO 2 absorption. Experimentally, it has already been demonstrated that the phase behavior of these systems is classified into three categories: phase separation type, miscible type, and immiscible type. This study demonstrates that the octanol/water partition coefficient is empirically able to search for a combination of amine and ether compounds, in which the system exhibits the phase separation behavior by absorbing CO 2 . The COSMO-RS calculations successfully reproduced experimental phase behavior with a rate of agreement of more than 80% by accounting the following two factors: (i) ion pairing for the description of the association/dissociation state for ionic species and (ii) relevant low-lying conformations of ether and amine, which are rationalized by experiments such as conductivity and excess enthalpy measurements. Moreover, we also validated the ability of the COSMO-RS calculation to qualitatively describe the compositions of 2-(ethylamino)ethanol, diethylene glycol diethyl ether, and water in CO 2 -rich and CO 2 -lean phases at varying CO 2 loading conditions.

An Alternative CO2 Capture with a Pressure Swing Amine Process Driven by Cryogenic Pumping with the Unused Cold Energy of Liquefied Natural Gas

“…We set one reaction for CO 2 capture. Since the developed solvent did not contain water, the ionic species were assumed to occur as a neutral ion pair and could be counted as one molecule. Our previous work, such as Walden plots and COSMO calculations, supports this assumption. Vapor pressure or vapor–liquid equilibrium was estimated from the molecular structure (UNIFAC model).…”

“…The solution was adjusted with a 30 wt % MEA aqueous solution and EAE/DEGMME = 30/70 (wt %). Our laboratory has been developing phase-separated absorbers that consist of EAE/diethylene glycol diethyl ether (DEGDEE)/water = 30/60/10 (wt %) − and show a large solubility gap for the temperature swing process. We also proposed a hydrogen stripping process that can reduce the CO 2 partial pressure at the desorber and can reduce the regeneration temperature.…”

“…Since the developed solvent did not contain water, the ionic species were assumed to occur as a neutral ion pair and could be counted as one molecule. Our previous work, such as Walden plots 22 and COSMO calculations, 23 Figure 4 shows a comparison between the developed model and experimental data. The systematic deviation was observed in the high loading region, which may be due to the formation of alkyl carbonates.…”

Prediction of the Aqueous Solubility of Compounds Based on Light Gradient Boosting Machines with Molecular Fingerprints and the Cuckoo Search Algorithm

“…To replace traditional mechanistic models, many researchers turned to the quantitative structure–property relationship (QSPR) model. − In QSPR, the quantitative relationship among the physicochemical properties, biological properties, and molecular structures of compounds is explored with various statistical methods and mathematical models. , Usually, the molecular descriptors were selected as inputs of the models . In previous studies, the main methods used in the QSPR model include multivariable linear regression (MLR), artificial neural network (ANN), Gaussian process (GP), and support vector machine (SVM). , However, the complex correlation between molecular descriptors and high-dimensional nonlinear data required for dissolution prediction poses great difficulties in traditional machine learning methods …”

New Screening Protocol for Effective Green Solvents Selection of Benzamide, Salicylamide and Ethenzamide

“…Hence, the first principle approaches seem to be attractive, even if prediction accuracies are only semi-quantitative or qualitative [ 55 , 65 , 66 ] Among many of them the COSMO-RS methodology [ 67 ], is a very powerful tool applied for predicting various physicochemical properties using exclusively information of the chemical formula. In addition to many molecular affinity-related properties such as activity coefficients [ 68 , 69 ], equilibrium constants [ 70 , 71 , 72 ], cocrystals and solvates screening [ 10 , 11 , 67 , 73 , 74 , 75 , 76 , 77 ], phase diagrams [ 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 ], solubility in neat and multicomponent solvents [ 18 , 33 , 40 , 45 , 55 , 61 , 74 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 ], solubility parameters estimation [ 96 , 97 , 98 , 99 ] and partition coefficients [ 92 , 100 ,…”