United polarizable multipole water model for molecular mechanics simulation

Area of Science:

• Computational chemistry
• Molecular modeling
• Physical chemistry

Background:

• Accurate molecular dynamics simulations require efficient and reliable water models.
• Existing polarizable water models like AMOEBA03 offer high accuracy but can be computationally expensive.

Purpose of the Study:

• To develop a computationally efficient coarse-grained polarizable water model (uAMOEBA).
• To achieve comparable accuracy to all-atom models for both gas-phase and liquid properties.

Main Methods:

• Developed a united-site polarizable water model (uAMOEBA) by reducing electrostatic and van der Waals representations to a single oxygen site.
• Retained hydrogen atoms solely for defining local frames and intramolecular modes.
• Parameterized the model using ab initio quantum mechanics and experimental data for gas-phase and liquid properties.

Main Results:

• The uAMOEBA model is 3-5 times more computationally efficient than AMOEBA03.
• It demonstrates comparable accuracy for gas-phase and liquid properties, including dimer energies, liquid structures, self-diffusion, and shear viscosity.
• The model shows good transferability across temperatures and environments due to molecular polarizability.

Conclusions:

• The uAMOEBA model provides an accurate and efficient alternative for simulating water.
• It successfully captures water coordination, hydrogen-bonding structure, and dynamic properties.
• This coarse-grained polarizable model is suitable for large-scale molecular dynamics simulations.