Electronic-structure calculations by first-principles density-based embedding of explicitly correlated systems

Abstract

A first-principles embedding theory that combines the salient features of density functional theory ͑DFT͒ and traditional quantum chemical methods is presented. The method involves constructing a DFT-based embedding potential and then using it as a one-electron operator within a very accurate ab initio calculation. We demonstrate how DFT calculations can be systematically improved via this procedure. The scheme is tested using two closed shell systems, a toy model Li 2 Mg 2 , and the experimentally well characterized CO/Cu͑111͒ system. Our results are in good agreement with near full configuration interaction calculations in the former case and experimental adsorbate binding energies in the latter. This method provides the means to systematically include electron correlation in a local region of a condensed phase.

Communication: Dynamical embedding: Correct quantum response from coupling TDDFT for a small cluster with classical near-field electrodynamics for an extended region

“…14 (For other approaches see Refs. [15][16][17][18][19][20] Our approach describes nanostructures as quantum systems embedded within a frequency-dependent dielectric medium; the systems are propagated simultaneously, coupled through an overall time-dependent potential. The method was demonstrated on a one-dimensional jellium system.…”

PyADF — A scripting framework for multiscale quantum chemistry

“…This new molecule object is then used to initialize and run an adfsinglepointjob, for which a higher integration accuracy and a larger basis set are used (lines [19][20][21][22]. The new results object is then used to initialize and run an adfnmrjob (lines [24][25][26]. Finally, the get_all_shieldings method of the NMR job's results object is used to extract the calculated shieldings, which can then be printed (lines [28][29][30][31].…”

“…[25][26][27][28] It can be shown that using such an embedding potentials is formally exact (i.e., it is exact if the exact exchangecorrelation and kinetic-energy functionals are used and the limit of an exact WFT description is reached). 65,66 In ref.…”

“…These are based on the frozendensity embedding (FDE) scheme initially proposed by Wesolowski and Warshel 22 (following earlier work by Senatore and Subbaswamy 23 and by Cortona 24 ) and its extension to WFT-in-DFT embedding, first proposed by Carter and coworkers. [25][26][27][28] Particularly for such quantum-chemical multiscale simulations one encounters very complex workflows that are beyond the capabilities of standard tools. Typically, they involve hundreds of individual calculations, and the results of a subset of these calculations are needed as input for following steps.…”

eQE: An open‐source density functional embedding theory code for the condensed phase

“…[20][21][22][23][24][25] Another notable implementation [26,27] resides in the CP2K code. [28] CASTEP [29] also has an implementation of FDE [30] which was employed in simulations involving two subsystems, one of which was treated at the correlated wavefunction level. [31][32][33] In addition, Turbomole [34] has its own implementation by the Della Sala group.…”