Nuclear Quantum Effect and Its Temperature Dependence in Liquid Water from Random Phase Approximation via Artificial...

Area of Science:

• Computational chemistry
• Materials science
• Physical chemistry

Background:

• Accurate description of liquid water is crucial for understanding various chemical and physical processes.
• Traditional methods often struggle to balance accuracy and computational cost for complex systems like water.

Purpose of the Study:

• To investigate the structural and dynamical properties of liquid water.
• To incorporate nuclear quantum effects and their temperature dependence.
• To develop a computationally feasible approach for high-level electronic structure calculations.

Main Methods:

• Utilized density functional theory with exact exchange (EXX) and random phase approximation (RPA) correlation.
• Employed thermostated ring polymer molecular dynamics to capture nuclear quantum effects.
• Adapted an artificial neural network model to overcome computational limitations of first-principles simulations.

Main Results:

• The EXX+RPA level of theory accurately reproduces both structural and dynamical properties of liquid water.
• Nuclear quantum effects were successfully incorporated and their temperature dependence analyzed.
• The artificial neural network model provided a computationally efficient alternative for high-level simulations.

Conclusions:

• The combination of EXX+RPA and ring polymer molecular dynamics, accelerated by neural networks, offers a powerful approach for studying liquid water.
• This methodology provides accurate insights into water's behavior, including the significant role of nuclear quantum effects.
• The findings pave the way for more precise simulations of water in various chemical and biological contexts.