Predicting miscibility of binary liquids from small cluster QCE calculations.

Area of Science:

• Physical Chemistry
• Computational Chemistry
• Thermodynamics

Background:

• Modeling molecular solvent mixtures is crucial for understanding chemical processes.
• Accurate prediction of thermodynamic properties like Gibbs energy of mixing is essential.
• Existing methods often require extensive experimental data or high computational cost.

Purpose of the Study:

• To apply the quantum cluster equilibrium method to model binary molecular solvent systems.
• To minimize computational effort and experimental input for solvent mixture modeling.
• To evaluate the method's accuracy for miscible and immiscible solvent pairs.

Main Methods:

• Utilized the quantum cluster equilibrium method with small cluster sizes (n=3).
• Employed the low-cost PBEh-3c functional for cluster optimization.
• Approximated empirical parameters using linear interpolation, reducing reliance on binary system data.

Main Results:

• Thermodynamic functions of pure liquids showed good agreement with experimental data.
• Achieved high accuracy (≈0.25 kJ/mol) for Gibbs energy of mixing in non-water systems.
• Correctly predicted mixing behavior for acetone/acetonitrile, acetone/benzene, and acetone/water systems.
• Accurately predicted the immiscibility of benzene/water, with a minor error at high benzene concentrations.

Conclusions:

• The quantum cluster equilibrium method, even with approximations and small cluster sets, effectively predicts mixing behavior for acetone-based binary solvent systems.
• The approach demonstrates potential for reducing experimental data requirements in solvent mixture modeling.
• Further refinement may be needed for accurate modeling of water-containing systems using small clusters.