Objectives: сomparison of extractive agents for the separation of ternary azeotropic mixtures using the Gibbs excess energy diagrams on the example of methanol – tetrahydrofuran – acetonitrile mixtures; prediction of extractive distillation results.
Methods. Computational experiment for calculating phase equilibria and the distillation process using the Non Random Two Liquid (NRTL) model; construction and visualization of data arrays for excess Gibbs energy for four-component systems using the marching cubes algorithm, implemented in the Python programming language have been developed.
Results. A computational experiment was conducted on simulation of vapor–liquid equilibria in four–component derivative systems, including a separable substances methanol, tetrahydrofuran, acetonitrile and an extractive agent. The Gibbs excess energy diagrams of derivative systems are constructed, the selective properties of extractive agents are analyzed, and the results of extractive distillation of methanol – tetrahydrofuran – acetonitrile mixtures with dimethyl sulfoxide, valeronitrile, and glycerin are predicted. The forecast is confirmed by calculations of extractive distillation for different compositions of methanol – tetrahydrofuran – acetonitrile mixtures with agents under discussion. The extractive distillation flowsheet with dimethyl sulfoxide is energy efficient at separating ternary mixtures of different compositions.
Conclusions. Using the marching cubes algorithm, it is possible to visualize the composition-excess Gibbs energy data array for four-component systems. The possibility of evaluating the selective action of agents and predicting the product flows of an extractive distillation column for the methanol – tetrahydrofuran – acetonitrile mixtures has been demonstrated, which plays a crucial role in the development of principal separation flowsheets.