Physical Chemistry

Excited States of Crystalline Point Defects with Multireference Density Matrix Embedding Theory

Authors

Abstract

Accurate and affordable methods to characterize the electronic structure of solids are important for targeted materials design. Embedding-based methods provide an appealing balance in the trade-off between cost and accuracy - particularly when studying localized phenomena. Here, we use the density matrix embedding theory (DMET) algorithm to study the electronic excitations in solid-state defects with a restricted open-shell Hartree--Fock (ROHF) bath and multireference impurity solvers, specifically, complete active space self-consistent field (CASSCF) and n-electron valence state second-order perturbation theory (NEVPT2). We apply the method to investigate an oxygen vacancy (OV) on a MgO(100) surface and find absolute deviations within 0.05 eV between DMET using the CASSCF/NEVPT2 solver, denoted as CAS-DMET/NEVPT2-DMET, and the non-embedded CASSCF/NEVPT2 approach. Next, we establish the practicality of DMET by extending it to larger supercells for the OV defect and a neutral silicon-vacancy in diamond where the use of non-embedded CASSCF/NEVPT2 is extremely expensive.

Content

Thumbnail image of pDMET2_manuscript_10_01_2021.pdf

Supplementary material

Thumbnail image of SI_pDMET_10_01_2021.pdf
Supplementary Information for "Excited States of Crystalline Point Defects with Multireference Density Matrix Embedding Theory"
This is the supplementary information for the original manuscript. Here are the contents: Computational details; A discussion on the better performance of the ROHF bath; Excitation energies for OV on a MgO(100) surface using a 2,2 active space; Total energies for supercells considered for OV in Mg(100) surface; Convergence of excitation energies in-plane; Excitation energies using the SiC6 impurity cluster in the silicon-vacancy defect; Total energies for the cluster calculations for the silicon-vacancy defect; Total Energies for supercells considered in the silicon-vacancy defect; Active spaces explored within this work.