H4: A model system for assessing the performance of diffusion Monte Carlo calculations using a single Slater determinant trial function.
Kevin GasperichMichael DeibleKenneth D JordanPublished in: The Journal of chemical physics (2018)
A model H4 system is used to investigate the accuracy of diffusion Monte Carlo (DMC) calculations employing a single Slater determinant to fix the nodal surface. The lowest energy singlet state of square H4 is a diradical which is poorly described by DMC calculations using a single determinant (SD) trial function. Here we consider distortions to rectangular structures, which decrease the amount of diradical character. The falloff of the error in the SD-DMC energy with increasing separation between the two H2 molecules is found to be much more rapid for small distortions away from square than for large distortions. This behavior is shown to be correlated with the extent of mixing between the two configurations needed to properly describe the diradical character. The error in the SD-DMC energy is found to be sizeable (∼0.1 eV) even for separations at which the coefficient of the dominant configuration in a four-electron, four-orbital complete active space self-consistent-field wave function is as large as 0.9.
Keyphrases
- monte carlo
- study protocol
- clinical trial
- density functional theory
- phase iii
- phase ii
- molecular dynamics simulations
- molecular dynamics
- lymph node
- high resolution
- randomized controlled trial
- radiation therapy
- neoadjuvant chemotherapy
- computed tomography
- diffusion weighted imaging
- capillary electrophoresis
- contrast enhanced