CO2 and H2S absorption in aqueous MDEA with ethylene glycol: Electrolyte NRTL, rate-based process model and pilot plant experimental validation

• Published in: Chemical engineering journal (Lausanne, Switzerland : 1996) Switzerland : 1996), 2023-01, Vol.451, p.138948, Article 138948
• Main Authors: Gonzalez, Karen, Boyer, Louis, Almoucachar, David, Poulain, Bénédicte, Cloarec, Eric, Magnon, Christophe, de Meyer, Frédérick
• Format: Article
• Language: English
• Subjects: absorption; acid gas; carbon dioxide; ethylene glycol; gas treatment; hydrogen sulfide; methyldiethanolamine; process simulation; regeneration
• ISSN: 1385-8947; 1873-3212
• DOI: 10.1016/j.cej.2022.138948

[Display omitted] •A process model of H2S and CO2 absorption with a blended solvent is presented.•It is based on new thermodynamic, kinetic, and physical experimental data.•It is validated with new high-pressure absorption-regeneration pilot plant tests.•Replacing part of the water by ethylene glycol lowers regeneration energy by ∼25%•However, a strong increase in solvent flow rate is required for iso-performance. Acid gas absorption with aqueous amines is an important industrial technology for natural or flue gas treatment. To lower the high energy required for the regeneration process it has previously been proposed to replace a significant part of the water by a physical co-solvent with a lower specific heat. To investigate this idea as well as the overall impact of adding a physical co-solvent, a detailed process model for CO2 and H2S absorption using an aqueous mixture of N-methyl diethanolamine (MDEA) - ethylene glycol (EG) is developed based on lab-scale experimental measurements of physical, thermodynamic, and kinetic properties. Those experiments indicate that the addition of EG reduces the solubility of the acid gases and that, despite the higher viscosity of the solvent, the impact on the rate of acid gas absorption is minor. Pilot plant tests further confirmed this. The Vapor-liquid equilibrium (VLE) experiments were used to regress the thermodynamic electrolyte e-NRTL (Non-Random Two-Liquid) models’ parameters. In parallel, the kinetic measurements allowed to develop a rate-based model. All models were implemented in a commercial process simulator that was validated by high-pressure absorption-regeneration pilot tests with the solvent composition of 34 wt% MDEA, 44.5 wt% EG and 21.5 wt% H2O and with the raw gas composition of 2.5 mol% H2S and 2.5 % mol CO2. Comparison with aqueous MDEA (50 wt% MDEA, and 50 wt% H2O) shows that a reduction of 25% in reboiler duty can be obtained when adding EG. Nevertheless, an 80% increase in solvent flow rate is needed to get a similar acid gas absorption performance, and a higher co-absorption of hydrocarbons is observed.