Quantum Monte Carlo Benchmark of Exchange-Correlation Functionals for Bulk Water

Area of Science:

• Computational physics and chemistry
• Materials science
• Quantum mechanics

Background:

• Accurately describing liquid water's thermodynamic and dynamical properties from first-principles is a significant challenge.
• Density Functional Theory (DFT) has historically struggled with water's unique liquid-state characteristics.
• Recent advancements in DFT, including exact exchange and improved dispersion interactions, offer new possibilities for accurate simulations.

Purpose of the Study:

• To benchmark various Density Functional Theory (DFT) exchange-correlation functionals for simulating bulk liquid water.
• To identify optimal DFT functionals for accurate first-principles simulations of water.
• To evaluate the impact of functional components like hybrid and van der Waals interactions on water's properties.

Main Methods:

• Utilized highly accurate quantum Monte Carlo (QMC) calculations as a benchmark.
• Assessed a selection of commonly used DFT exchange-correlation functionals.
• Employed a hybrid DFT and Coupled Cluster (CC) scheme with a Many-Body expansion for short-range interaction correction.

Main Results:

• Quantum Monte Carlo calculations provide a reliable method for benchmarking DFT functionals for liquid water.
• The study identifies specific DFT functional features (hybrid, vdW) crucial for accurately predicting water's properties.
• The combined DFT-CC approach shows promise in correcting short-range interaction inaccuracies.

Conclusions:

• This work provides a pathway to selecting superior DFT functionals for first-principles simulations of liquid water.
• Understanding the role of different functional components is key to improving water simulations.
• The benchmarked functionals and methods pave the way for more accurate theoretical studies of water.