Infinite dilution activity coefficient from SMD calculations: accuracy and performance for predicting liquid-liquid...

Area of Science:

• Physical Chemistry
• Computational Chemistry
• Thermodynamics

Background:

• Predicting liquid-liquid phase equilibria (LLPE) is vital for understanding solution behavior.
• Quantum chemistry methods offer a first-principles approach to modeling LLPE.
• Continuum solvation models are increasingly used to approximate solvent effects.

Purpose of the Study:

• To evaluate the performance of the continuum solvation model based on density (SMD) for predicting activity coefficients at infinite dilution (γ∞).
• To assess the accuracy of SMD in calculating LLPE for binary liquid mixtures.
• To determine the potential of this computational approach for semi-quantitative phase behavior predictions.

Main Methods:

• Utilized 46 experimental values of γ∞ to assess the SMD model's predictive capability.
• Calculated LLPE using Redlich-Kister expansion parameters derived from computed γ∞.
• Validated predictions against 26 experimental solubility data points.

Main Results:

• The SMD model demonstrated a mean uncertainty of 0.92 kcal mol⁻¹ for RTlnγ∞.
• Calculated LLPE showed an uncertainty of 0.66 in the logarithm of the molar fraction for the minor component.
• The approach provides fast and semi-quantitative predictions of phase behavior.

Conclusions:

• The continuum solvation model based on density (SMD) shows promise for predicting LLPE in binary mixtures.
• Further refinements to the SMD model could enhance prediction accuracy.
• This method offers a valuable tool for rapid, semi-quantitative assessment of solution phase behavior.