Abstract

A new type of electronic trial wavefunction suitable for quantum Monte Carlo calculations of molecular systems is presented. In contrast with the standard Jastrow-Slater form built with a unique global Jastrow term, it is proposed to introduce individual Jastrow factors attached to molecular orbitals. Such a form is expected to be more physical since it allows to describe differently the local electronic correlations associated with various molecular environments (1s-core orbitals, 3d-magnetic orbitals, localized two-center sigma-orbitals, delocalized pi-orbitals, atomic lone pairs, etc.). In contrast with the standard form, introducing different Jastrow terms allows us to change the nodal structure of the wavefunction, a point which is important in the context of building better nodes for more accurate fixed-node diffusion Monte Carlo (FN-DMC) calculations. Another important aspect resulting from the use of local Jastrow terms is the possibility of defining and preoptimizing local and transferable correlated units for building complex trial wavefunctions from simple parts. The practical aspects associated with the computation of the intricate derivatives of the multi-Jastrow trial function are presented in detail. Some first illustrative applications for atoms of increasing size (O, S, and Cu) and for the potential energy curve and spectroscopic constants of the FH molecule are presented. In the case of the copper atom, the use of the multi-Jastrow form at the variational Monte Carlo level has allowed us to improve significantly the value of the total ground-state energy (about 75% of the correlation energy with only one determinant and three atomic orbital Jastrow factors). In the case of the FH molecule (fluorine hydride), it has been found that the multi-Jastrow nodes lead to an almost exact FN-DMC value of the dissociation energy [D(0)=-140.7(4) kcal/mol instead of the estimated nonrelativistic Born-Oppenheimer exact value of -141.1], which is not the case with standard nodes, D(0)=-138.3(4) kcal/mol.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.