2019
DOI: 10.1021/acs.jctc.9b00317
|View full text |Cite
|
Sign up to set email alerts
|

Stochastic Vertex Corrections: Linear Scaling Methods for Accurate Quasiparticle Energies

Abstract: New stochastic approaches for the computation of electronic excitations are developed within the many-body perturbation theory. Three approximations to the electronic self-energy are considered:All three methods are formulated in the time domain and the latter two incorporate non-local vertex corrections. In case of G 0 W tc 0 Γ x , the vertex corrections are included both in the screened Coulomb interaction and in the expression for the self-energy. The implementation of the three approximations is verified b… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

3
42
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
5
3

Relationship

1
7

Authors

Journals

citations
Cited by 35 publications
(45 citation statements)
references
References 86 publications
(272 reference statements)
3
42
0
Order By: Relevance
“…For nanoscale systems with thousands of atoms, ∼ 100 samples suffice to represent the GF, with only ∼ 10 needed to represent δn(r, t). [57,58,60] The stochastic methodology capitalizes on the fact that the key quantities (G and W ) are determined by collective properties, which are inherently low-rank and captured by the dynamics of a few (random) states within the Hilbert space. This approach leads to a linear scaling algorithm that can treat thousands of atoms.…”
Section: A Stochastic Many-body Theorymentioning
confidence: 99%
See 1 more Smart Citation
“…For nanoscale systems with thousands of atoms, ∼ 100 samples suffice to represent the GF, with only ∼ 10 needed to represent δn(r, t). [57,58,60] The stochastic methodology capitalizes on the fact that the key quantities (G and W ) are determined by collective properties, which are inherently low-rank and captured by the dynamics of a few (random) states within the Hilbert space. This approach leads to a linear scaling algorithm that can treat thousands of atoms.…”
Section: A Stochastic Many-body Theorymentioning
confidence: 99%
“…In this work, we overcome these limitations: we pro-pose two new developments in the stochastic many-body perturbation theory (MBPT) techniques [57][58][59][60][61][62] which can readily elucidate how the electronic structure behaves in giant moiré systems. We investigate tBLG with a large twist angle of θ ≈ 6 • (with 2184 atoms, i.e., 8736 valence electrons), which is weakly correlated at ambient conditions but develops flat bands at high compressions.…”
Section: Introductionmentioning
confidence: 99%
“…43,75,76 Nevertheless, the GW correlation neglects the quantum fluctuations, which may become important for high energy excitations and/or for unoccupied states. 64,77 The method of subspace separations is, however, general and applicable to beyond-GW approaches; it will be explored in the future. Here, a single-shot perturbative correction is computed within the random phase approximation (RPA) on top of a density functional theory (DFT) starting point (see the SI for details).…”
Section: Stochastic Calculation Of the Self-energymentioning
confidence: 99%
“…[44][45][46][47][48][49][50][51][52][53][54][55][56][57][58][59] Further, the methodology is open for systematic improvements; embedding within the GF framework 21,43,47,48,54,60 is conceptually straightforward and "seamless." Finally, recent developments of stochastic GF methods [60][61][62][63][64] enabled calculations for giant systems with thousands of electrons, practically treating large and inhomogeneous systems on equal footing.…”
Section: Introductionmentioning
confidence: 99%
“…"Vertex correc- tions" to GW has a long history. Several different types of vertex corrections have been proposed and tested [26,36,37,41,[44][45][46][47][48][49][50][51][52][53][54][55]. Despite considerable efforts, none of these vertexcorrections have yet found wide-spread use in production calculations.…”
Section: Introductionmentioning
confidence: 99%