A quantum Monte Carlo study of the ground state chromium dimer

TL;DR

This study uses quantum Monte Carlo methods to investigate the ground-state binding curve of the chromium dimer, exploring various wavefunctions, computational techniques, and their effects on the results.

Contribution

It provides a comprehensive quantum Monte Carlo analysis of the chromium dimer, comparing different wavefunctions and computational parameters for the first time.

Findings

DMC predicts a bound state for chromium dimer.

Binding energies are negative across all methods.

Wavefunction choice significantly affects binding curves.

Abstract

We report variational and diffusion quantum Monte Carlo (VMC and DMC) studies of the binding curve of the ground-state chromium dimer. We employed various single determinant (SD) or multi-determinant (MD) wavefunctions multiplied by a Jastrow fuctor as a trial/guiding wavefunction. The molecular orbitals (MOs) in the SD were calculated using restricted or unrestricted Hartree-Fock or density functional theory (DFT) calculations where five commonly-used local (SVWN5), semi-local (PW91PW91 and BLYP), and hybrid (B1LYP and B3LYP) functionals were examined. The MD expansions were obtained from the complete-active space SCF, generalized valence bond, and unrestricted configuration interaction methods. We also adopted the UB3LYP-MOs to construct the MD expansion (UB3LYP-MD) and optimized their coefficients at the VMC level. In addition to the wavefunction dependence, we investigated the time-step bias in the DMC calculation and the effects of pseudopotentials and backflow transformation for the UB3LYP-SD case. Some of the VMC binding curves show a flat or quite shallow well bottom, which gets recovered deeper by DMC. All the DMC binding curves have a minimum indicating a bound state, but the comparison of atomic and molecular energies gives rise to a negative binding energy for all the DMC as well as VMC calculations.