A predictive model for the diffusion of a highly non-ideal ternary system

Area of Science:

• Chemical Engineering
• Physical Chemistry

Background:

• Diffusion is crucial in chemical unit operations.
• The Maxwell-Stefan model, derived from gas kinetic theory, is widely used but its applicability to non-ideal liquids is questioned.
• The dynamic fluctuation model offers an alternative, accounting for concentration fluctuations.

Purpose of the Study:

• To extend the dynamic fluctuation model to a highly non-ideal ternary liquid system.
• To compare its predictive performance against the Maxwell-Stefan model.
• To investigate the role of local mole fractions in diffusion modeling.

Main Methods:

• Extension of the dynamic molecular fluctuation model to ternary systems via matrix manipulation.
• Utilizing self-diffusion data from Nuclear Magnetic Resonance (NMR).
• Comparison of model predictions with experimental mutual diffusion data for ethanol/toluene/n-decane.

Main Results:

• The dynamic fluctuation model demonstrated improved prediction accuracy over the Maxwell-Stefan model for the ternary system.
• Incorporating local mole fractions enhanced predictions for both models, particularly for molecular association.
• An optimal scaling factor (α) of approximately 0.45 was found, consistent with previous binary system studies.

Conclusions:

• The dynamic fluctuation model provides a more accurate description of mutual diffusion in non-ideal liquid mixtures.
• The model suggests a non-linear correction to the thermodynamic correction factor.
• The Maxwell-Stefan model's assumptions may not fully apply to highly non-ideal liquid systems.