The classical barrier height for the H + H 2 exchange reaction, as well as the energies at two other points along the reaction path, are calculated using fixed-node quantum Monte Carlo (FNQMC).Several single-determinant importance functions are used at the saddle-point in order to relate the quality of the importance function to the accuracy and precision of the final result.The computed barrier is an upper bound since the energy of H and of H 2 is obtained exactly by FNQMC.Our best upper bound, 9.70 ~ 0.13 kcal/mole, has a mean within 0.1 kcal/mole of the presumed exact value.This best bound is obtained with a single determinant, double-zeta basis importance function.Contrary to experience with expansion methods, it is found that an importance function with a basis set of near Hartree-Fock quality, as well as one derived from a spin-unrestricted SCF calculation, are among the least efficient and least accurate of the importance functions used.Specifically, a nodal surface appearing in the lowest energy molecular orbital in these functions apparently increases the FNQMC energy.The FNQMC energy at the two other points along the reaction path is found to agree with the most accurate CI results of Liu to within statistical error.
The classical barrier height for the H + H 2 exchange reaction, as well as the energies at two other points along the reaction path, are calculated using fixed-node quantum Monte Carlo (FNQMC).Several single-determinant importance functions are used at the saddle-point in order to relate the quality of the importance function to the accuracy and precision of the final result.The computed barrier is an upper bound since the energy of H and of H 2 is obtained exactly by FNQMC.Our best upper bound, 9.70 ~ 0.13 kcal/mole, has a mean within 0.1 kcal/mole of the presumed exact value.This best bound is obtained with a single determinant, double-zeta basis importance function.Contrary to experience with expansion methods, it is found that an importance function with a basis set of near Hartree-Fock quality, as well as one derived from a spin-unrestricted SCF calculation, are among the least efficient and least accurate of the importance functions used.Specifically, a nodal surface appearing in the lowest energy molecular orbital in these functions apparently increases the FNQMC energy.The FNQMC energy at the two other points along the reaction path is found to agree with the most accurate CI results of Liu to within statistical error.
The classical barrier height for the H+H2 exchange reaction, as well as the energies at two other points along the reaction path, are calculated using fixed-node quantum Monte Carlo (FNQMC). Several single-determinant importance functions are used at the saddle point in order to relate the quality of the importance function to the accuracy and precision of the final result. The computed barrier is an upper bound since the energy of H and of H2 is obtained exactly by FNQMC. Our best upper bound (9.70±0.13 kcal/mol) has a mean within 0.1 kcal/mol of the presumed exact value. This best bound is obtained with a single determinant, double-zeta basis importance function. Contrary to experience with expansion methods, it is found that an importance function with a basis set of near Hartree–Fock quality, as well as one derived from a spin-unrestricted SCF calculation, are among the least efficient and least accurate of the importance functions used. Specifically, a nodal surface appearing in the lowest energy molecular orbital in these functions apparently increases the FNQMC energy. The FNQMC energy at the two other points along the reaction path is found to agree with the most accurate CI results of Liu to within statistical error.
The fixed-node quantum Monte Carlo (QMC) method is used to obtain the classical barrier height for the H + H2 exchange reaction. Using a spin-restricted, single-determinant trial function ΨT, the authors find that the reaction barrier Eb is less than 9.69 ± 0.25 kcal/mol. This mean value is below the calculated energy barrier obtained by Liu in the most extensive CI calculations on this system. Furthermore, the QMC saddle-point energy of – 1.65903 ± 0.00040 hartrees at the CI-determined geometry lies 0.00027 a.u. (0.17 kcal/mol) below Liu's best CI value. Finally, spinrestricted and spin-unrestricted single-determinant trial functions are contrasted. Although the variational energy 〈ΨT|H|ΨT〉 for an SCF ΨT must be lower for the unrestricted case, this is not true generally for QMC. In fact, we show that if the unrestricted SCF ΨT has the lower QMC energy, then there exists another spin-restricted, single-determinant ΨT whose QMC energy is lower than the QMC energy of the spin-restricted SCF ΨT.
We investigate here the performance of our recently developed linear-scaling Jastrow-generalized-valence-bond (J-LGVB) wave functions based on localized orbitals, for the quantum Monte Carlo (QMC) calculation of the barrier heights and reaction energies of five prototypical chemical reactions. Using the geometrical parameters from the Minnesota database collection, we consider three hydrogen-exchanges, one heavy-atom exchange, and one association reaction and compare our results with the best available experimental and theoretical data. For the three hydrogen-exchange reactions, we find that the J-LGVB wave functions yield excellent QMC results, with average deviations from the reference values below 0.5 kcal/mol. For the heavy-atom exchange and association reactions, additional resonance structures are important, and we therefore extend our original formulation to include multiple coupling schemes characterized by different sets of localized orbitals. We denote these wave functions as J-MC-LGVB, where MC indicates the multiconfiguration generalization, and show that such a form leads to very accurate barrier heights and reaction energies also for the last two reactions. We can therefore conclude that the J-LGVB theory for constructing QMC wave functions, with its multiconfiguration generalization, is valid for the study of large portions of ground-state potential energy surfaces including, in particular, the region of transition states.
Abstract In this work Manten and Liichow assessed the accuracy of allelectron fixed-node diffusion QMC calculations with nodes taken from Hartee-Fock wavefunctions with high-quality basis sets. The electronic energies associated with 17 different reactions involving a total of 20 small molecules were determined and compared with experimental values and with those calculated by Klopper et al.a using the coupled cluster method CCSD(T) with correlation consistent basis sets cc-pVDZ and cc-pVTZ. The molecules involved were limited to those containing the atoms H, C, 0, N, and F. Typical reactions were H2CO + 2H2 --+ CH4 + H2O and H2O + F --+ HOF + H. The geometries for the QMC calculations were optimized with MP2/cc-pVTZ calculations. The QMC total energies for the 20 molecules were lower than CCSD(T) values, but only valence electrons were correlated in the CCSD(T) calculations. Node location error was approximately 0.0015 hartrees for the C, N, 0, and F atoms and their hydrides, but substantial cancellation of node error is to be expected in determining enthalpy changes for reactions.
The Diffusion Monte Carlo (DMC) method is a powerful strategy to estimate the ground state energy E0 of an N-body Schrödinger Hamiltonian H = -½Δ + V with high accuracy. It consists of writing E0 as the long-time limit of an expectation value of a drift-diffusion process with a source term, and numerically simulating this process by means of a collection of random walkers. As for a number of stochastic methods, a DMC calculation makes use of an importance sampling function ψI which hopefully approximates some ground state ψ0 of H. In the fermionic case, it has been observed that the DMC method is biased, except in the special case when the nodal surfaces of ψI coincide with those of a ground state of H. The approximation due to the fact that, in practice, the nodal surfaces of ψI differ from those of the ground states of H, is referred to as the Fixed Node Approximation (FNA). Our purpose in this paper is to provide a mathematical analysis of the FNA. We prove that, under convenient hypotheses, a DMC calculation performed with the importance sampling function ψI, provides an estimation of the infimum of the energy 〈ψ, Hψ〉 on the set of the fermionic test functions ψ that exactly vanish on the nodal surfaces of ψI.
In this paper we discuss the application of quantum Monte Carlo (QMC) techniques to the electronic many-body problem as encountered in computational chemistry. The Fixed-Node Diffusion Monte Carlo (FN-DMC) algorithm —the most common QMC scheme for treating molecules—is presented. The impact of the fixed-node error is illustrated through numerical applications including the calculation of the electronic affinity of the chlorine atom, the dissociation barrier of the O 4 molecule, and the binding energy of the dichromium molecule, Cr 2 . Although total energies calculated with FN-DMC are very accurate (more accurate than the best alternative methods available), it is emphasized that the error associated with approximate nodes can lead to important errors in the small differences of total energies, quantities which are particularly important in chemistry.