We gauge the importance of selfinteraction errors in density functional approximations (DFAs) for the case of water clusters. To this end, we used the Fermi–Löwdin orbital selfinteraction correction method (FLOSIC) to calculate the binding energy of clusters of up to eight water molecules. Three representative DFAs of the local, generalized gradient, and metageneralized gradient families [i.e., local density approximation (LDA), Perdew–Burke–Ernzerhof (PBE), and strongly constrained and appropriately normed (SCAN)] were used. We find that the overbinding of the water clusters in these approximations is not a densitydriven error. We show that, while removing selfinteraction error does not alter the energetic ordering of the different water isomers with respect to the uncorrected DFAs, the resulting binding energies are corrected toward accurate reference values from higherlevel calculations. In particular, selfinteraction–corrected SCAN not only retains the correct energetic ordering for water hexamers but also reduces the mean error in the hexamer binding energies to less than 14 meV/
 Award ID(s):
 1900338
 Publication Date:
 NSFPAR ID:
 10258454
 Journal Name:
 Faraday Discussions
 Volume:
 224
 Page Range or eLocationID:
 9 to 26
 ISSN:
 13596640
 Sponsoring Org:
 National Science Foundation
More Like this

${\mathrm{H}}_{2}\mathrm{O}$ from about 42 meV/${\mathrm{H}}_{2}\mathrm{O}$ for SCAN. By decomposing the total binding energy into manybody components, we find that large errors in the twobody interaction in SCAN are significantly reduced by selfinteraction corrections. Higherorder manybody errors are small in both SCAN andmore » 
Abstract We present a graph‐theoretic approach to adaptively compute many‐body approximations in an efficient manner to perform (a) accurate post‐Hartree–Fock (HF) ab initio molecular dynamics (AIMD) at density functional theory (DFT) cost for medium‐ to large‐sized molecular clusters, (b) hybrid DFT electronic structure calculations for condensed‐phase simulations at the cost of pure density functionals, (c) reduced‐cost on‐the‐fly basis extrapolation for gas‐phase AIMD and condensed phase studies, and (d) accurate post‐HF‐level potential energy surfaces at DFT cost for quantum nuclear effects. The salient features of our approach are ONIOM‐like in that (a) the full system (cluster or condensed phase) calculation is performed at a lower level of theory (pure DFT for condensed phase or hybrid DFT for molecular systems), and (b) this approximation is improved through a correction term that captures all many‐body interactions up to any given order within a higher level of theory (hybrid DFT for condensed phase; CCSD or MP2 for cluster), combined through graph‐theoretic methods. Specifically, a region of chemical interest is coarse‐grained into a set of nodes and these nodes are then connected to form edges based on a given definition of local envelope (or threshold) of interactions. The nodes and edges together define a graph,more »

Abstract The HohenbergKohn theorem of densityfunctional theory establishes the existence of a bijection between the groundstate electron density and the external potential of a manybody system. This guarantees a onetoone map from the electron density to all observables of interest including electronic excitedstate energies. TimeDependent DensityFunctional Theory (TDDFT) provides one framework to resolve this map; however, the approximations inherent in practical TDDFT calculations, together with their computational expense, motivate finding a cheaper, more direct map for electronic excitations. Here, we show that determining density and energy functionals via machine learning allows the equations of TDDFT to be bypassed. The framework we introduce is used to perform the first excitedstate molecular dynamics simulations with a machinelearned functional on malonaldehyde and correctly capture the kinetics of its excitedstate intramolecular proton transfer, allowing insight into how mechanical constraints can be used to control the proton transfer reaction in this molecule. This development opens the door to using machinelearned functionals for highly efficient excitedstate dynamics simulations.

M. Lewin, Rupert L. (Ed.)Abstract: Lieb and Oxford (1981) derived rigorous lower bounds, in the form of local functionals of the electron density, on the indirect part of the Coulomb repulsion energy. The greatest lower bound for a given electron number N depends monotonically upon N, and the N→∞ limit is a bound for all N. These bounds have been shown to apply to the exact density functionals for the exchange and exchangecorrelation energies that must be approximated for an accurate and computationally efficient description of atoms, molecules, and solids. A tight bound on the exact exchange energy has been derived therefrom for twoelectron ground states, and is conjectured to apply to all spinunpolarized electronic ground states. Some of these and other exact constraints have been used to construct two generations of nonempirical density functionals beyond the local density approximation: the PerdewBurkeErnzerhof (PBE) generalized gradient approximation (GGA), and the strongly constrained and appropriately normed (SCAN) metaGGA.

Abstract The formation of clusters at subsaturation densities, as a result of manybody correlations, constitutes an essential feature for a reliable modelization of the nuclear matter equation of state (EoS). Phenomenological models that make use of energy density functionals (EDFs) offer a convenient approach to account for the presence of these bound states of nucleons when introduced as additional degrees of freedom. However, in these models clusters dissolve, by construction, when the nuclear saturation density is approached from below, revealing inconsistencies with recent findings that evidence the existence of shortrange correlations (SRCs) even at larger densities. The idea of this work is to incorporate SRCs in established models for the EoS, in light of the importance of these features for the description of heavyion collisions, nuclear structure and in the astrophysical context. Our aim is to describe SRCs at suprasaturation densities by using effective quasiclusters immersed in dense matter as a surrogate for correlations, in a regime where cluster dissolution is usually predicted in phenomenological models. Within the EDF framework, we explore a novel approach to embed SRCs within a relativistic meanfield model with density dependent couplings through the introduction of suitable inmedium modifications of the cluster properties, in particularmore »