Modeling of membrane separation of liquid mixture in Aspen HYSYS

Objectives. To develop and subsequently verify the calculation block of the mass transfer process in the pervaporation membrane module based on a HybSi ® ceramic membrane using experimental data as a basis for the verification process. Methods. The task was implemented using a mathematical simulation within the Aspen HYSYS application package, which is designed for modeling chemical engineering processes. The differential equations of the mathematical model were represented as a system of difference equations, which were then solved numerically with an adaptive area step. The membrane pervaporation module of area S during its modeling is divided into n intervals, based on ensuring within the i th interval the condition that the temperature change D Т is less than 1°C. A model was constructed to simulate the performance of the membrane module under isothermal and adiabatic operating conditions. Results. The mathematical model of the pervaporation process employed in the developed computational membrane pervaporation module considers variations in the concentration and temperature of the feedstock flux along the surface of the HybSi ® membrane. The performance of the software module was evaluated by comparing the calculated results with the available experimental data for the dehydration of ethanol and isopropanol. The results demonstrated a high degree of agreement for three isotherms (60, 70, and 80°C) and two variations of pressure on the permeate side (5 and 20 mm Hg). Modeling of the operation of the membrane module with the area of 1 m 2 in adiabatic mode showed that the processes of alcohol dehydration on HybSi ® membranes are accompanied by significant thermal effects associated with heat consumption to provide evaporation through the membrane due to large transmembrane fluxes. Conclusions. The comparative analysis of the results of modeling the HybSi ® membrane module in isothermal and adiabatic modes of operation demonstrated that the calculation of the membrane module without consideration of thermal effects results in significant errors. These include an overestimation of the permeate flow rate by up to 50% and an underestimation of the water concentration in the retentate by up to 1.3–1.8 times. It can be reasonably deduced that the omission of thermal effects in design calculations will result in a considerable underestimation of the requisite membrane module surface area.