Using CIPSI Nodes in Diffusion Monte Carlo

The Journal of Chemical Physics

Giner

et al. 2018

Self Cite

134

163

Selected configuration interaction (sCI) methods including second-order perturbative corrections provide near full CI (FCI) quality energies with only a small fraction of the determinants of the FCI space. Here, we introduce both a state-specific and a multi-state sCI method based on the configuration interaction using a perturbative selection made iteratively (CIPSI) algorithm. The present method revises the reference (internal) space under the effect of its interaction with the outer space via the construction of an effective Hamiltonian, following the shifted-Bk philosophy of Davidson and co-workers. In particular, the multi-state algorithm removes the storage bottleneck of the effective Hamiltonian via a low-rank factorization of the dressing matrix. Illustrative examples are reported for the state-specific and multi-state versions.

Section: A State-specific Examplementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Selected configuration interaction dressed by perturbation

Garniron

The Journal of Chemical Physics

Giner

et al. 2018

Self Cite

134

163

“…47,48 Moreover, the mechanism and degree of error compensation of the fixed-node error [49][50][51][52][53] in the ground and excited states are mostly unknown, expect in a few cases. [54][55][56][57][58][59][60][61] Here, within our Jastrow-free QMC protocol relying on a deterministic construction of nodal surfaces, 40,[62][63][64][65] we report all-electron FN-DMC calculations for the ground and excited states of water (H 2 O) and formaldehyde (CH 2 O) using large Dunning's basis sets including diffuse functions. Our results for these two molecules evidence that one is able to obtain accurate excitation energies with relatively compact trial wave functions built with relatively small one-electron basis sets.…”

Section: Introductionmentioning

confidence: 99%

Excitation energies from diffusion Monte Carlo using selected configuration interaction nodes

The Journal of Chemical Physics

Benali

Jacquemin

et al. 2018

Self Cite

105

Quantum Monte Carlo (QMC) is a stochastic method that has been particularly successful for ground-state electronic structure calculations but mostly unexplored for the computation of excited-state energies. Here, we show that within a Jastrow-free QMC protocol relying on a deterministic and systematic construction of nodal surfaces using selected configuration interaction (sCI) expansions, one is able to obtain accurate excitation energies at the fixed-node diffusion Monte Carlo (FN-DMC) level. This evidences that the fixed-node errors in the ground and excited states obtained with sCI wave functions cancel out to a large extent. Our procedure is tested on two small organic molecules (water and formaldehyde) for which we report all-electron FN-DMC calculations. For both the singlet and triplet manifolds, accurate vertical excitation energies are obtained with relatively compact multideterminant expansions built with small (typically double-ζ) basis sets.

“…For more details about our implementation of pseudopotentials within QMC, we refer the interested readers to Ref. 127.…”

Section: Basismentioning

confidence: 99%

Influence of pseudopotentials on excitation energies from selected configuration interaction and diffusion Monte Carlo

Caffarel

Benali

et al. 2019

Results in Chemistry

Self Cite

Due to their diverse nature, the faithful description of excited states within electronic structure theory methods remains one of the grand challenges of modern theoretical chemistry.antum Monte Carlo (QMC) methods have been applied very successfully to ground state properties but still remain generally less e ective than other non-stochastic methods for electronically excited states. Nonetheless, we have recently reported accurate excitation energies for small organic molecules at the xed-node di usion Monte Carlo (FN-DMC) within a Jastrow-free QMC protocol relying on a deterministic and systematic construction of nodal surfaces using the selected con guration interaction (sCI) algorithm known as CIPSI (Con guration Interaction using a Perturbative Selection made Iteratively). Albeit highly accurate, these all-electron calculations are computationally expensive due to the presence of core electrons. One very popular approach to remove these chemically-inert electrons from the QMC simulation is to introduce pseudopotentials (also known as e ective core potentials). Taking the water molecule as an example, we investigate the in uence of Burkatzki-Filippi-Dolg (BFD) pseudopotentials and their associated basis sets on vertical excitation energies obtained with sCI and FN-DMC methods. Although these pseudopotentials are known to be relatively safe for ground state properties, we evidence that special care may be required if one strives for highly accurate vertical transition energies. Indeed, comparing all-electron and valence-only calculations, we show that using pseudopotentials with the associated basis sets can induce di erences of the order of 0.05 eV on the excitation energies. Fortunately, a reasonable estimate of this shi can be estimated at the sCI level.