Thermodynamic modelling of carbon dioxide solubility in aqueous amino acid salt solutions and their blends with alkanolamines

Abstract

In this study, two thermodynamic approaches, namely explicit model and modified Kent-Eisenberg model are presented for the accurate determination of carbon dioxide solubility in aqueous solutions of amino acid salts with different levels of complexity. Both models were further extended to the correlation of carbon dioxide solubility in aqueous blends of amino acid salts with alkanolamines. Firstly, the explicit model consists of a single mathematical equation. Its structure is computationally simple and derived from equilibrium thermodynamics theory. Secondly, the modified Kent-Eisenberg model utilizes detailed reaction mechanism for development of a polynomial equation, that is solved by combining all non-idealities in two correction factors. Both models suitably determined the thermodynamics of carbon dioxide loaded aqueous amino acid salt solutions and their blends with alkanolamines. The results were in good agreement with experimental data for representative amino acid salt solutions (potassium and/or sodium salts of lysine, glycine, proline, sarcosine, serine, threonine, alanine, phenylalanine, amino-butyric acid, glutamine and asparagine) and their blends with alkanolamines (2-amino-2-methyl-1-propanol and piperazine) for range of process parameters. For carbon dioxide solubility in aqueous amino acid salt solutions, the AARE% for explicit and modified Kent-Eisenberg model was 13.85% and 13.84%, respectively. For carbon dioxide solubility in aqueous blends of amino acid salt and alkanolamine solutions, the AARE% for explicit and modified Kent-Eisenberg model was 18.25% and 17.25%, respectively. Both models use a small number of adjustable parameters. This indicates that generated parameters are able to accurately predict the carbon dioxide solubility at other process conditions, with minimum computational intricacy.