Benchmark all-electron ab initio quantum Monte Carlo calculations for small molecules.

Area of Science:

• Computational chemistry
• Quantum mechanics
• Electronic structure theory

Background:

• Quantum Monte Carlo (QMC) methods are powerful tools for electronic structure calculations.
• Slater basis sets are commonly used in quantum chemistry.
• Hartree-Fock and density functional theory provide starting wave functions for QMC.

Purpose of the Study:

• To evaluate the efficiency, precision, and accuracy of all-electron variational and diffusion QMC calculations.
• To develop and apply an algorithm for enforcing the electron-nucleus cusp condition.
• To assess the performance of QMC for molecular energies and reaction energies.

Main Methods:

• All-electron variational and diffusion quantum Monte Carlo calculations.
• Use of Slater basis sets.
• Hartree-Fock and density functional theory for initial wave functions.
• Linear projection algorithm to enforce the electron-nucleus cusp condition.

Main Results:

• Diffusion QMC recovered an average of 95% of the correlation energy for the G2 set of 55 molecules.
• Calculated bond energies showed a mean absolute deviation of 3.2 kcal/mol from experimental values.
• Analysis of error cancellation in atomization and reaction energies provided insights into nodal surface errors.

Conclusions:

• QMC methods, particularly diffusion QMC with an enforced electron-nucleus cusp condition, offer high accuracy for molecular energies.
• The approach demonstrates significant potential for reliable predictions of chemical properties.
• Further investigation into nodal surface errors can lead to even more accurate QMC calculations.