Parametrização da equação de estado PC-SAFT para o cálculo de propriedades termofísicas e solubilidade de CO2 em líquidos iônicos

Abstract

In this work, modeling was carried out using PC-SAFT (Perturbed Chain Statistical Associating Fluid Theory) in its complete and simplified versions, and evaluating the influence of its terms on the solubility calculation. The equation of state was used to describe the thermophysical properties of ILs, predict the solubility of gases, and compare the results obtained with experimental data in the literature. This work extends the methodology proposed by Loreno (2017), in which the model parameters, including the associative parameters, were estimated for the “IL core” using specific mass and sound speed data at atmospheric pressure as input parameters, and applying the 2B association scheme. This strategy was compared with that using specific mass data at different temperatures and pressures. A sensitivity analysis of the model parameters was also carried out by evaluating the parametric confidence region. In addition, the influence of differences in experimental density, thermal expansion coefficient and isothermal compressibility on model performance was analyzed. For imidazolium IL families, strategy 1 (which uses density and sound speed data at atmospheric pressure) showed the best results. The study was extended to ILs from methylpyridinium, methylpyrrolidinium and pyridinium families using strategy 1. The proposed modeling was able to describe thermophysical properties and CO2 solubility data in these LIs with good accuracy, dispensing with the use of binary parameters, and being compatible with GC methods. Predictions of methane solubility in imidazolium family ILs were also analyzed. The model was able to describe the concavity of the solubility curves, as well as the order of magnitude of methane solubility. The deviations obtained for CH4 solubility are considerable. However, it is expected that the selectivity will not be so different from the ideal selectivity, and that these deviations will not significantly influence the selectivity. It can therefore be concluded that the model developed in this work can be used in the context of removing CO2 from gas streams.