2022
DOI: 10.1021/acs.nanolett.2c00550
|View full text |Cite
|
Sign up to set email alerts
|

STEM Image Analysis Based on Deep Learning: Identification of Vacancy Defects and Polymorphs of MoS2

Abstract: Scanning transmission electron microscopy (STEM) is an indispensable tool for atomicresolution structural analysis for a wide range of materials. The conventional analysis of STEM images is an extensive hands-on process, which limits efficient handling of highthroughput data. Here we apply a fully convolutional network (FCN) for identification of important structural features of two-dimensional crystals. ResUNet, a type of FCN, is utilized in identifying sulfur vacancies and polymorph types of MoS2 from atomic… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
11
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 19 publications
(14 citation statements)
references
References 47 publications
(93 reference statements)
0
11
0
Order By: Relevance
“…To explore the possibility of the presence of multiple MoS 2 polymorphs , (i.e., 1T, 1H (2H-mono), 2H, and 3R), several dozens of NFs were thoroughly examined in each sample in terms of their lattice structure. In all monolayer NFs, only the thermodynamically favored 1H phase was identified, with no evidence of transversal displacement of one layer of sulfur atoms, characteristic of the 1T phase.…”
Section: Resultsmentioning
confidence: 99%
“…To explore the possibility of the presence of multiple MoS 2 polymorphs , (i.e., 1T, 1H (2H-mono), 2H, and 3R), several dozens of NFs were thoroughly examined in each sample in terms of their lattice structure. In all monolayer NFs, only the thermodynamically favored 1H phase was identified, with no evidence of transversal displacement of one layer of sulfur atoms, characteristic of the 1T phase.…”
Section: Resultsmentioning
confidence: 99%
“…The local atomic registries in our VHs were analyzed via atomic-resolution scanning transmission electron microscopy (STEM). Conventionally, high-angle annular dark field (HAADF) STEM is utilized to distinguish various stacking orders based on a Z -contrast mechanism Figure c,d shows HAADF-STEM images of R-type and H-type VHs near the edge of each top WS 2 layer, respectively.…”
Section: Resultsmentioning
confidence: 99%
“…Conventionally, high-angle annular dark field (HAADF) STEM is utilized to distinguish various stacking orders based on a Z-contrast mechanism. 28 Figure 2c,d shows HAADF-STEM images of R-type and H-type VHs near the edge of each top WS 2 layer, respectively. Again, there is no observable moirésuperlattice at the VH regions, at least within our experimental resolution, confirming the commensurate VHs.…”
Section: Resultsmentioning
confidence: 99%
“…Therefore, it can be difficult to identify the structural defects created by Ar‐plasma by STEM and TEM. [ 46 ] Alternatively, it has been demonstrated that Raman and PL signals are competent in characterizing plasma‐generated surface defects of ultrathin TMDs. [ 47–50 ] Figure 4b,c displays the Raman and PL spectra of MoS 2 under Ar‐plasma bombardment with different duration.…”
Section: Resultsmentioning
confidence: 99%