Fast and accurate quantum Monte Carlo for molecular crystals

Area of Science:

• Materials Science
• Computational Chemistry
• Condensed Matter Physics

Background:

• Computer simulations are crucial in materials science, balancing accuracy and computational expense.
• Molecular crystals, vital for technology, are difficult to model due to weak intermolecular forces and large unit cells.
• Existing ab initio methods struggle with the accuracy and cost for complex molecular crystals.

Purpose of the Study:

• To evaluate the accuracy and computational cost of diffusion quantum Monte Carlo (DMC) for molecular crystals.
• To demonstrate DMC's capability in describing materials with weak intermolecular interactions.
• To explore DMC's potential for predicting properties of large-molecule molecular crystals.

Main Methods:

• Diffusion Quantum Monte Carlo (DMC) simulations.
• Application to a diverse set of molecular crystal systems.
• Analysis of computational cost versus achieved accuracy.

Main Results:

• DMC achieved subchemical accuracy for various molecular crystals.
• The computational cost of DMC was found to be surprisingly moderate.
• DMC's performance was validated across different molecular crystal structures.

Conclusions:

• DMC offers a highly accurate and computationally feasible approach for studying molecular crystals.
• This method can significantly aid in understanding and predicting properties of challenging molecular materials.
• DMC is a promising tool for materials discovery, especially for systems beyond the reach of other high-accuracy techniques.