Multiphase Equilibria Modeling of CO2 the Homologous Series of Carboxylic Acids

JACQUELINE MANSANO ORTEGA BACICHETI; VLADIMIR FERREIRA CABRAL; FRANCISCO A. SÁNCHEZ

Los Cocos

Conferencia; VI Iberoamerican Conference on Supercritical Fluids (PROSCIBA 2023); 2023

IPQA - UNC/CONICET

Supercritical fluid extraction (SFE) is an attractive substitute for conventional solvent extraction methods for obtaining fine chemicals, such as fatty and dicarboxylic acids. Carbon dioxide (CO2) is the most commonly used supercritical solvent due to its favorable properties including non-flammability, non-toxicity, non-polluting nature and easy recoverability. Designing efficient separation units for SFE necessitates careful thermodynamic modeling. The knowledge of the critical properties and phase equilibria of systems involving CO2 and carboxylic acids is crucial in the context of SFE, such as oil deacidification processes and extraction of acids from aqueous substrates. Nevertheless, modeling phase equilibria in mixtures containing CO2 and carboxylic acids is a challenge because of their highly asymmetrical behavior. Moreover, while experimental data of CO2 + monocarboxylic acids are available, data of CO2 + dicarboxylic acids are scarce. Consequently, the development of predictive thermodynamic models becomes essential for process design and optimization. In this study, we employ the Group Contribution with Association Equation of State (GCA-EOS) to comprehensively model the phase equilibria of CO2 + carboxylic acid homo-logue series. The GCA-EOS has been successfully applied to represent the global phase be-havior of carbon dioxide mixtures with alkanes and alcohols. In this work, we extend this model to represent vapor−liquid, liquid−liquid, and vapor−liquid−liquid equilibria of CO2 mixtures with short and long-chain carboxylic acids, as well as dicarboxylic acids. Through a meticulous parametrization strategy coupled with a robust thermodynamic model, we develop a tool for assessing the multiphase behavior of CO2 + carboxylic acids. The model is  validated against an experimental database covering C2−C20 carboxylic acids, temperatures from 308 to 473 K, and pressures up to 400 bar. Remarkably, utilizing a single set of parameters, the model successfully predicts phase equilibria of binary mixtures not included in the parametrization procedure.