Abstract
Defects are ubiquitous in 2D materials and can affect the structure and properties of the materials and also introduce new functionalities. Methods to adjust the structure and density of defects during bottom-up synthesis are required to control the growth of 2D materials with tailored optical and electronic properties. Here, the authors present an Au-assisted chemical vapor deposition approach to selectively form SW and S2W antisite defects, whereby one or two sulfur atoms substitute for a tungsten atom in WS2 monolayers. Guided by first-principles calculations, they describe a new method that can maintain tungsten-poor growth conditions relative to sulfur via the low solubility of W atoms in a gold/W alloy, thereby significantly reducing the formation energy of the antisite defects during the growth of WS2. The atomic structure and composition of the antisite defects are unambiguously identified by Z-contrast scanning transmission electron microscopy and electron energy-loss spectroscopy, and their total concentration is statistically determined, with levels up to ≈5.0%. Scanning tunneling microscopy/spectroscopy measurements and first-principles calculations further verified these antisite defects and revealed the localized defect states in the bandgap of WS2 monolayers. This bottom-up synthesis method to form antisite defects should apply in the synthesis of other 2D materials.
Original language | English |
---|---|
Article number | 2106674 |
Journal | Advanced Materials |
Volume | 34 |
Issue number | 3 |
DOIs | |
State | Published - Jan 20 2022 |
Funding
This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences (BES), Materials Sciences and Engineering Division (synthesis science and first‐principles modeling). This research was conducted at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility. The modeling defect thermodynamics was supported by the Creative Materials Discovery Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (NRF‐2016M3D1A1919181). This research used resources of the Oak Ridge Leadership Computing Facility and the National Energy Research Scientific Computing Center, DOE Office of Science User Facilities.