Recommended by

Indexed by

Efficient and accurate defect level modelling in monolayer MoS₂ via GW+DFT with open boundary conditions

Guido Gandus1*, Youseung Lee1*, Leonard Deuschle1*, Daniele Passerone2*, Mathieu Luisier1*

1 Integrated Systems Laboratory, ETH Zürich, Switzerland

2 nanotech@surfaces Laboratory, EMPA Zürich, Switzerland

* Corresponding authors emails: ggandus@ethz.ch, youslee@iis.ee.ethz.ch, dleonard@ethz.ch, daniele.passerone@empa.ch, mluisier@iis.ee.ethz.ch
DOI10.24435/materialscloud:h4-c0 [version v1]

Publication date: Nov 22, 2022

How to cite this record

Guido Gandus, Youseung Lee, Leonard Deuschle, Daniele Passerone, Mathieu Luisier, Efficient and accurate defect level modelling in monolayer MoS₂ via GW+DFT with open boundary conditions, Materials Cloud Archive 2022.157 (2022), doi: 10.24435/materialscloud:h4-c0.


Within the framework of many-body perturbation theory integrated with density functional theory (DFT), a novel defect-subspace projection GW method, the so-called p-GW, is proposed. By avoiding the periodic defect interference through open boundary self-energies, we show that the p-GW can efficiently and accurately describe quasi-particle correlated defect levels in two-dimensional (2D) monolayer MoS₂. By comparing two different defect states originating from sulfur vacancy and adatom to existing theoretical and experimental works, we show that our GW correction to the DFT defect levels is precisely modelled. Based on these findings, we expect that our method can provide genuine trap states for various 2D transition-metal dichalcogenide (TMD) monolayers, thus enabling the study of defect-induced effects on the device characteristics of these materials via realistic simulations.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.


File name Size Description
1.2 MiB Files used to obtain the results shown in the paper.


Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.



Version history:

2022.157 (version v1) [This version] Nov 22, 2022 DOI10.24435/materialscloud:h4-c0