Data-Driven Many-Body Models for Molecular Fluids: CO₂/H₂O Mixtures as a Case Study.

Area of Science:

• Computational chemistry
• Molecular modeling
• Physical chemistry

Background:

• The TTM-nrg and MB-nrg potential energy functions (PEFs) were developed for ion-water interactions.
• Accurate modeling of molecular fluids like carbon dioxide (CO2) and water (H2O) is crucial for understanding their properties.

Purpose of the Study:

• To extend the TTM-nrg and MB-nrg PEFs for modeling CO2-H2O mixtures.
• To evaluate the performance of these PEFs in capturing many-body effects in molecular fluids.

Main Methods:

• Derived PEFs from coupled cluster electronic structure data.
• Utilized TTM-nrg and MB-nrg functional forms for electrostatics, polarization, and dispersion.
• Employed multidimensional permutationally invariant polynomials for short-range interactions in MB-nrg.

Main Results:

• MB-nrg PEFs quantitatively represent potential energy surfaces for CO2-CO2 and CO2-H2O dimers.
• Accurately reproduced energetics of small clusters and properties in gas and liquid phases.
• Demonstrated compatibility with the MB-pol framework for water modeling.

Conclusions:

• The MB-nrg framework accurately and efficiently models molecular interactions in CO2-H2O fluid mixtures.
• The study provides evidence for the framework's applicability across various temperature and pressure conditions.
• This extension enhances the capability of many-body PEFs for complex fluid systems.