A systematic study on the fixed-node and localization error in quantum Monte Carlo calculations with pseudopotentials for group III elements

doi:10.1016/0009-2614(94)00355-6

Chemical Physics Letters

Volume 222, Issue 3, 13 May 1994, Pages 274-280

https://doi.org/10.1016/0009-2614(94)00355-6 Get rights and content

Abstract

Short-time fixed-node pure diffusion quantum Monte Carlo (PDMC) calculations with pseudopotentials for the group III elements B, Al, Ga and In have been carried out in order to study the capability of multireference trial wavefunctions to reduce the fixed-node and localization error. For this purpose PDMC energies for the ²P and ⁴P states of the neutral atoms and for the ¹S state of the positive ions have been compared with the results of MRCI calculations. For these simple systems, we demonstrate the possibility of a systematic reduction of these errors by adding supplementary configurations to the trial wavefunction.

References (21)

P. Fuentealba et al.
Chem. Phys. Letters
(1982)
S.J. Chakravorty et al.
C.J. Umrigar et al.
J. Chem. Phys.
(1993)
L.D. Landau et al.
(1958)
B.L. Hammond et al.
J. Chem. Phys.
(1987)
M.M. Hurley et al.
J. Chem. Phys.
(1987)
L. Mitas et al.
J. Chem. Phys.
(1991)
S. Fahy et al.
Phys. Rev. Letters
(1988)
S. Fahy et al.
Phys. Rev.
(1990)
H.J. Flad et al.
J. Chem. Phys.
(1992)
G. Igel-Mann et al.
Mol. Phys.
(1988)
P.A. Christiansen
J. Chem. Phys.
(1988)
P.A. Christiansen

There are more references available in the full text version of this article.

Cited by (12)

Quantum Monte Carlo methods for the solution of the Schrödinger equation for molecular systems
2003, Handbook of Numerical Analysis
This chapter discusses quantum Monte Carlo methods for the solution of the schrödinger equation for molecular systems. The solution of the time independent Schrödinger equation for molecular systems requires the use of modern computers, because analytic solutions are not available. This review deals with some of the methods known under the umbrella term quantum Monte Carlo (QMC), specifically those that have been most commonly used for electronic structure. QMC methods have several advantages: "Computer time scales with system size roughly as N³, where N is the number of particles of the system. Recent developments have made possible the approach to linear scaling in certain cases." “Computer memory requirements are small and grow modestly with system size.” “QMC computer codes are significantly smaller and more easily adapted to parallel computers than basis set molecular quantum mechanics codes.” “Basis set truncation errors are absent in the QMC formalism.” The chapter presents a description of the commonly used algorithms of QMC for electronic structure and to report some recent developments in the field.
Quantum Monte Carlo with reoptimised perturbatively selected configuration-interaction wave functions
2016, Molecular Physics
Stochastic simulations of clusters: Quantum methods in flat and curved spaces
2009, Stochastic Simulations of Clusters: Quantum Methods in Flat and Curved Spaces
Benchmark quantum Monte Carlo calculations
2002, Journal of Chemical Physics
Quantum Monte Carlo: Atoms, Molecules, Clusters, Liquids, and Solids
1999, Reviews in Computational Chemistry
Probing the accuracy of pseudopotentials for transition metals in quantum Monte Carlo calculations
1997, Journal of Chemical Physics

View all citing articles on Scopus

¹: Present address: Laboratoire Dynamique des Interactions Moléculaires (CNRS, UPR 271), Université Pierre et Marie Curie, Tour 22, 4 Place Jussieu, 75252 Paris, Cedex 05, France.

View full text

A systematic study on the fixed-node and localization error in quantum Monte Carlo calculations with pseudopotentials for group III elements

Abstract

Chem. Phys. Letters

J. Chem. Phys.

J. Chem. Phys.

J. Chem. Phys.

J. Chem. Phys.

Phys. Rev. Letters

Phys. Rev.

J. Chem. Phys.

Mol. Phys.

J. Chem. Phys.