Developing Efficient and Accurate Polarizable Molecular Simulation Models for Studying Refrigerants and Their Binary...

Area of Science:

• Thermodynamics
• Materials Science
• Computational Chemistry

Background:

• Hydrofluoroolefins (HFOs) are crucial low-global-warming-potential (GWP) refrigerants.
• Limited mixture data hinders accurate prediction of HFO phase behavior and thermodynamic properties.
• Developing reliable predictive models is essential for practical applications.

Purpose of the Study:

• To develop a data-efficient polarizable molecular modeling strategy for HFO-1234yf, HFO-1234ze(E), and their binary mixtures.
• To accurately predict thermodynamic properties and phase behavior using minimal experimental data.
• To provide a framework for designing and screening new low-GWP refrigerants.

Main Methods:

• Combined high-level quantum-mechanical potential-energy data with limited experimental properties (liquid density, heat capacity, enthalpy of vaporization).
• Developed polarizable force fields for pure HFOs and their binary mixtures.
• Validated models against experimental data for thermodynamic properties and vapor-liquid equilibrium.

Main Results:

• Force fields accurately reproduced pure fluid properties (heat capacities, enthalpies of vaporization) with <5% deviation.
• Accurate prediction of vapor-liquid equilibrium across wide temperature and composition ranges (saturation pressure MAE: 0.43 bar).
• Predicted surface tensions with a mean absolute error of 1.36 mN/m.
• Identified fluorinated-group interactions as key drivers of volatility and nonideal mixing.

Conclusions:

• Polarizable force fields, refined with high-level quantum data and minimal experimental input, offer accurate predictions for refrigerant blends.
• This data-efficient strategy is effective in data-sparse regimes for molecular design and screening.
• The approach provides a practical framework for developing next-generation low-GWP refrigerants.