Fixed-Node, Importance-Sampling Diffusion Monte Carlo for Vibrational Structure with Accurate and Compact Trial...

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Molecular Simulation

Background:

• Diffusion Monte Carlo (DMC) is a powerful method for simulating quantum systems.
• Accurately describing vibrational excited states remains a challenge in molecular simulations.
• Importance sampling is crucial for enhancing the efficiency of DMC calculations.

Purpose of the Study:

• To develop a new importance sampling technique for DMC simulations of vibrational excited states.
• To improve the accuracy and efficiency of calculating low-energy vibrational states.
• To avoid the need for a priori coordinate choices in DMC simulations.

Main Methods:

• Incorporating trial wave functions for low-energy states into diffusion equations.
• Enforcing orthogonality of trial wave functions within the diffusion process.
• Utilizing vibrational self-consistent field (VSCF) and vibrational configuration interaction (VCI) wave functions.
• Variationally optimizing internal coordinates for one-particle functions.

Main Results:

• The new approach effectively performs importance sampling for vibrational excited states.
• Simple VSCF and VCI wave functions are sufficient for effective importance sampling.
• Results are comparable in accuracy to unguided DMC calculations.
• The method eliminates the need for pre-selecting coordinates to define nodal hyperplanes.

Conclusions:

• The proposed method offers an accurate and efficient way to perform DMC simulations of vibrational excited states.
• This approach simplifies the simulation setup by removing the requirement for a priori coordinate selection.
• The technique holds promise for advancing quantum mechanical simulations in computational chemistry.