This study is aimed at evaluating the potential of the conductor-like screening model for real solvents (COSMO-RS) in designing biphasic systems for the separation of natural compounds. This model was used to predict binodal curves and phases composition of organic biphasic systems containing deep eutectic solvents. The results obtained demonstrate that COSMO-RS in most cases predict the binodal curves of the biphasic systems with a good precision. Moreover, the partition coefficients of six natural compounds, namely quercetin, apigenin, coumarin, β-ionone, retinol, and α-tocopherol, were evaluated, with COSMO-RS predictions showing excellent agreement with experimental data when the compound partition significantly in both phases. Furthermore, COSMO-RS was able to effectively capture the influence of changes in the system composition on partition coefficients. Thus, these results indicate that COSMO-RS can be a powerful tool in the design of separation and purification systems.
Read full abstract