Growth Mechanism of Alternating Defect Domains in Hexagonal WS₂ via Inhomogeneous W‐Precursor Accumulation

Abstract

While a hexagonal WS2 monolayer, grown by chemical vapor deposition, exhibits distinctive patterns in photoluminescence mapping, segmented with alternating S‐vacancy (SV) and W‐vacancy (WV) domains in a single crystal, the formation mechanism for native alternating defect domains remains unresolved to date. Here, the formation mechanism of alternating defect domains in hexagonal WS2 via the precursor accumulation model is experimentally elucidated. A triangular WS2 seed is initially formed, followed by a hexagonal flake. Alternating W‐rich (SV) and W‐deficient (WV) domains are constructed in hexagonal WS2 flake, which is confirmed by confocal photoluminescence mapping and secondary ion mass spectroscopy. This is explained by the accumulation or scarcity of W‐precursors at the edge of the WS2 flake. The W‐precursors accumulate near the edges of the initial triangular WS2 seed over time, while they are deficient near the corners of the triangular WS2, eventually forming WV domains in the truncated hexagonal domains. The heterogeneous accumulation becomes more prominent in the presence of H2 gas through desorption of the W‐precursors.

Strain-Induced Alternating Photoluminescence Segmentation in Hexagonal Monolayer Tungsten Disulfide Grown by Physical Vapor Deposition

“…The dark regions further merge and eventually disappear, leading to a uniform bright triangle WS 2 flake (Figure i). The regions with S-terminated edges are dark and those with W-terminated edges are bright, which are the same as previous reports. , However, the integrated PL intensity from 1.80 to 2.10 eV in the dark regions is about 80% of that in the bright regions which is far less than the values of previous reports. ,, Two typical PL spectra from the bright and dark regions in the regular hexagonal WS 2 are shown in Figure j. The PL spectra are deconvoluted to separated peaks by fitting two Lorentzian curves, corresponding to the exciton emission ( A ) and the negative trion emission ( A – ), respectively .…”

“…The spatial nonuniformities of PL intensity in CVD-synthesized TMDCs are mainly from atomic defects and strain between deposited 2D films and substrates. ,− Several groups synthesized monolayer hexagonal-shaped WS 2 (h-WS 2 ) with alternating areas of bright and dark PL intensities by the CVD method under hydrogen-rich growth conditions. ,− They proposed the existence of tungsten vacancies as the origin of significantly quenched PL intensity (10 times weaker) in the dark domains of h-WS 2 . Alternatively, the stored strain from the distinct coefficient of thermal expansion (CTE) between the substrate and 2D TMDCs is also directional (tensile strain or compressive strain) and easier to form a regular distribution in a 2D thin film. − In a hexagonal TMDC flake, the adjacent domains with different terminated edges show different reactions to the strain. , Density functional theory results indicate that the S-terminated region shows more Young’s modulus under compression strain, meaning that the S-terminated region is difficult to deform compared with the W-terminated region.…”

The Hidden Flower in WS2 Flakes: A Combined Nanomechanical and Tip-Enhanced Raman Exploration

“…A sketch of the change in sulfur concentration with time/temperature can be found in Figure S2 of the Supporting Information. The initial flake is of hexagonal shape with different terminated edges and grown around a nucleation spot in the center of the flake. , Further growth is defect dominated. Based on the interpretation of the drop in the PL and Raman intensities, we conclude that the high sulfur concentration at the beginning of the APCVD experiment leads to a rapid growth of the flake which generates tungsten-vacancy domains − arranged in a three-fold symmetry.…”

Unveiling the Critical Intermediate Stages During Chemical Vapor Deposition of Two-Dimensional Rhenium Diselenide

“…Although the precursors of CVD of TMDs, and the structures of 2D TMDs are well known, the reaction kinetics in vapor or on solid surfaces in the CVD process are not well understood yet. Contradictory growth mechanisms have been proposed previously, mostly through simulations, such as reactions on oxide surfaces, intermediate MO x C y (M = transition metal, O = oxygen, C = chalcogenide) clusters, MC 2 clusters (M = transition metal, C = chalcogenide), , etc.…”

“…12 Undoubtedly, the clarification of kinetic reaction mechanisms of CVD growth of 2D TMDs will greatly benefit the synthesis, in terms of product quality and reproducibility, and will further influence widely the relevant application fields. 13 Although the precursors of CVD of TMDs, and the structures of 2D TMDs are well known, 14 the reaction kinetics in vapor or on solid surfaces 15 in the CVD process are not well understood yet. Contradictory growth mechanisms have been proposed previously, mostly through simulations, such as reactions on oxide surfaces, 16 intermediate MO x C y (M = transition metal, O = oxygen, C = chalcogenide) clusters, 17 MC 2 clusters (M = transition metal, C = chalcogenide), 18,19 etc.…”