Improvement of Component Flux Estimating Model for Pervaporation Processes.

Area of Science:

• Chemical Engineering
• Separation Processes
• Membrane Technology

Background:

• Pervaporation (PV) offers an alternative to traditional separation methods like distillation for non-ideal mixtures.
• Accurate vapor-liquid equilibrium (VLE) data complicates traditional modeling, necessitating robust PV models.
• Existing empirical models primarily focus on estimating component flux (J).

Purpose of the Study:

• To improve and rigorously test the Mizsey and Valentinyi pervaporation (PV) model.
• To extend exponential modeling from alcohol-water to ethyl acetate-water mixtures.
• To present a flowchart for exponential PV model development.

Main Methods:

• Statistical testing and least squares approximation were used to validate the models.
• The Mizsey and Valentinyi exponential model was extended to ethyl acetate-water mixtures.
• Model performance was evaluated using laboratory-scale experimental data.

Main Results:

• Valentinyi's model (Model I) and an alternative model (Model III) showed strong performance for PV modeling, especially in organophilic cases.
• Model I is recommended for predicting permeation flux due to its adherence to exponential functions.
• The study confirmed the universal feasibility of exponential models for organic-water binary mixtures.

Conclusions:

• Exponential pervaporation models are effective for a wide range of organic-water binary mixtures.
• The validated models provide accurate predictions for component flux, aiding separation process design.
• The study demonstrated the applicability of semi-empirical models to PEBA-based membranes for water flux description.