Strain-Induced Decoupling Drives Gold-Assisted Exfoliation of Large-Area Monolayer 2D Crystals

Abyay Ghosh; Fumin Huang; Jakob Ziewer; Luka Pirker; Mat\v{e}j Velick\'y; Michaela Hanu\v{s}ov\'a; Myrta Gr\"uning; Otakar Frank

REVIEW 3 major objections 6 minor 5 references

Strain-Induced Decoupling Drives Gold-Assisted Exfoliation of Large-Area Monolayer 2D Crystals

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Strain in the adhered layer severs the first MoS2-MoS2 interface, making it the weakest point, so gold-assisted exfoliation peels off large-area monolayers.

desk verdict Solid new evidence that gold weakens the first MoS2-MoS2 interface, but the strain-as-driver claim outruns the DFT, which supports it only at three times the measured strain. read the letter →

arxiv 2412.05898 v1 pith:A6RN6FF3 submitted 2024-12-08 cond-mat.mtrl-sci cond-mat.mes-hall

Jakob Ziewer , Abyay Ghosh , Michaela Hanušová , Luka Pirker , Otakar Frank , Matěj Velický , Myrta Grüning , Fumin Huang This is my paper · ORCID

classification cond-mat.mtrl-scicond-mat.mes-hall

keywords 2DmaterialsMoS2gold-assistedexfoliationstraindecouplingultralow-frequencyRamanspectroscopyshearmodesinterlayercoupling

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to explain why gold-assisted exfoliation so reliably produces centimeter-scale monolayers rather than random few-layer flakes. Using MoS2 on gold as the model system, the authors argue that biaxial strain from the gold substrate weakens the coupling at the first MoS2-MoS2 interface, i.e., between the adhered bottom layer and the layer above it. The weakening grows with crystal thickness: roughly 20% for bilayers, 50% for tetralayers, and nearly 100% for crystals thicker than five layers. This makes the first interface the weakest point, so the crystal preferentially cleaves there, leaving a large-area monolayer on the gold. If correct, the result replaces the old requirement that substrate adhesion must beat the full interlayer van der Waals force with a softer condition: adhesion only needs to beat the strain-weakened first-interface bond.

What carries the argument

The load-bearing tool is a modified linear chain model (mLCM) in which each MoS2 layer is a mass point and adjacent layers are connected by springs, with the first interface spring $K_\alpha$ treated as variable and all other interlayer springs fixed at the pristine value $K_0$. The S1 shear-mode frequencies are obtained as eigenvalues of the N x N force-constant matrix in Eq. (3), and matching those eigenvalues to the measured S1 frequencies on gold yields the thickness-dependent $K_\alpha$. The second piece of machinery is the DFT exfoliation-energy comparison $\Delta E_{\mathrm{exfo}}$, which contrasts a stack that separates into an unstrained (N-1)-layer crystal plus a strained monolayer against the whole relaxed strained stack; negative values mark the regime where monolayer exfoliation is preferred. Together they connect a spectroscopic observable (the S1 downshift) to a mechanical property (interfacial spring constant) and to the energetics of cleavage.

What would settle it

Apply controlled biaxial strain to a MoS2 bilayer or multilayer on an inert substrate while recording its S1 shear mode; the strain-decoupling claim requires the mode to soften continuously and the extracted $K_\alpha$ to approach zero at the ~1-1.5% strain level found on gold, so observing no such softening at those strains would refute the mechanism.

Watch

Extended reading notes

Core claim

The central discovery is that the first MoS2-MoS2 interface in a crystal adhered to gold is decoupled by strain, not by adhesion alone. The evidence comes from ultralow-frequency Raman spectroscopy: the lowest shear mode (S1) of MoS2 on gold is downshifted relative to the same crystals on SiO2, and the shift is thickness-dependent in a way that a single universal interlayer spring constant cannot explain. Fitting the S1 frequencies with a modified linear chain model in which only the first-interface spring constant $K_\alpha$ is allowed to vary while all other springs keep the pristine value $K_0$ yields $K_\alpha \approx 0.8K_0$ for bilayer, $\approx 0.5K_0$ for tetralayer, and $\approx 0$ for crystals thicker than five layers. Supporting observations include the splitting of the high-frequency E' mode into a strained bottom-layer peak and an unstrained top-layer peak, the suppression of breathing modes on gold, and the formation of bubbles upon annealing that burst to expose monolayers only. Density functional theory calculations of the exfoliation energy show that beyond a strain threshold and thickness, separating the top stack from the strained bottom layer becomes energetically favorable, and the paper identifies the strain-induced decoupling as the primary mechanism of gold-assisted exfoliation.

Load-bearing premise

The argument's load-bearing premise is that the measured softening of the lowest shear mode comes entirely from lowering the spring constant $K_\alpha$ at the first MoS2-MoS2 interface while all other interlayer springs stay at their pristine value $K_0$; if doping, stacking registry, or substrate stiffening also moves the mode, the near-zero-coupling conclusion does not follow.

Editorial extensions

If this is right

For MoS2 crystals thicker than five layers on gold, the system behaves as a monolayer pinned to the substrate plus a decoupled (N-1)-layer stack, so the S1 and B1 frequencies should match those of a pristine (N-1)-layer crystal shifted by one layer; the paper reports this match.
Because the first interface is the weakest point, bubbles that form and burst during annealing expose monolayer surfaces, directly confirming where cleavage occurs.
The adhesion requirement for high-yield exfoliation is relaxed: the substrate need only overcome the weakened first-interface bond, not the full interlayer van der Waals force, so strain-inducing substrates other than gold may work.
Electrostatic repulsion from gold-induced doping contributes only about 0.1% of the interlayer attractive pressure, so strain, not charge transfer, is the operative driver.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the decoupling scales with strain, then choosing or patterning substrates to maximize biaxial strain in the adhered layer could extend monolayer-yield exfoliation to metals and dielectrics that adhere more weakly than gold; this is testable by measuring the E' Raman shift and monolayer yield across candidate substrates.
The measured average strain (~1.1-1.5%) is below the DFT threshold (~3.5%) for negative exfoliation energy, which suggests the cleavage-relevant strain may be local rather than average; a spatially resolved strain map at the exfoliation edge could resolve this discrepancy.
A natural next test is to apply the same ULF Raman and mLCM analysis to other chalcogenides (WS2, MoSe2, WSe2) and to alternative metals, since the paper's mechanism predicts the same thickness-dependent $K_\alpha$ collapse wherever the adhered layer is strained.
The near-zero $K_\alpha$ for thick crystals implies the top stack is nearly freestanding, which could affect electronic or thermal transport measurements performed on gold-supported thick flakes; interpreting such measurements may need to account for the decoupled interface.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: the decoupling observation is solid and new; the strain-causation claim is plausible but not quantitatively supported, and their own DFT is the weakest link.

What is genuinely new: ULF Raman of 2-10L MoS2 on Au with a modified linear chain model yields the first quantitative estimate of the first-interface force constant in gold-assisted exfoliation. Two independent checks do not depend on the fitted K_alpha at all: the N-layer-on-Au data matching (N-1)-layer-on-SiO2 curves, and the annealed-bubble bursts that expose monolayers everywhere. The SiO2 reference fits reproduce literature interlayer constants, a good sanity check. The electrostatic-repulsion estimate is a tidy order-of-magnitude argument. I believe the core observation: the first interface is substantially weakened, nearly decoupled for crystals thicker than five layers.

Soft spots, in proportion. K_alpha is fitted per thickness without error bars, and every substrate effect is squeezed into that one spring; a modified effective mass, strain-induced stacking registry change, or a stiffening not describable as a spring reduction would also shift S1. The shifted-curve check softens this but does not eliminate it. Bigger issue: the strain mechanism. The DFT computes a cleavage energy for normal separation, whereas the S1 mode probes the interlayer shear constant - a layer can have near-zero shear restoring force and still a positive cleavage energy. Their own threshold is ~3.5% strain versus the measured 1.1-1.5%, a factor-of-three gap the paper calls "qualitative agreement". The DFT cell contains no gold, so it cannot separate strain from doping or the quasi-covalent Au interaction; the admitted unexplained suppression of the B1 modes is a sign the substrate coupling is richer than the model. Finally, the strained-layer E' peak is unobservable above five layers, so strain in exactly the thickness range of complete decoupling is inferred, not measured.

The paper is honest about the DFT gap and about the B1 puzzle, and the decoupling result should survive even if the strain story is wrong. But the title says strain "drives" the effect, and that half needs either a strain-variation study measuring K_alpha, a DFT shear-constant calculation under strain, or a softened claim.

Send it to peer review rather than desk-reject; a serious referee should push on the strain link. The paper is for the 2D materials and Raman communities, and the central measurement deserves to be published.

Referee Report

3 major / 6 minor

Summary. The manuscript reports an ultralow-frequency Raman study of MoS2 flakes exfoliated on Au and SiO2 substrates. The authors observe that the first-order shear mode S1 is downshifted on Au relative to SiO2, and they interpret this with a modified linear chain model in which only the first MoS2-MoS2 interface spring constant K_alpha is reduced. The fit yields K_alpha approximately 0.8 K0 for bilayer, about 0.5 K0 for tetralayer, and nearly zero for thicknesses above five layers. Two independent checks are provided: the S1/B1 frequencies of an N-layer flake on Au match those of an (N-1)-layer flake on SiO2 for N >= 5, and annealing-induced bubbles that burst expose monolayers. High-frequency E' Raman indicates 1.1-1.5% biaxial strain in the bottom layer. DFT exfoliation energy calculations show a strain threshold around 3.5-4% for exfoliation to become energetically favorable. The authors conclude that strain-induced decoupling at the first interface is the primary mechanism of gold-assisted exfoliation.

Significance. If the mechanism is correct, it transforms the understanding of gold-assisted exfoliation from a simple adhesion-over-van-der-Waals picture to a strain-controlled interfacial decoupling picture, with practical implications for substrate engineering. The experimental core is strong: the S1 downshifts are systematic, the shifted-curve comparison in Figure 4b provides a parameter-free check of effective decoupling, and the bubble-burst images in Figure 5 are direct evidence that the first interface is the preferential cleavage plane. The strain estimate from the E' mode is consistent across substrate thicknesses. However, the causal link from strain to near-zero K_alpha is supported only by a DFT cleavage-energy calculation that does not directly address the shear force constant and whose strain threshold is several times larger than the measured strain. The manuscript would be strengthened by a direct calculation of the shear constant or phonon frequency as a function of biaxial strain.

major comments (3)

[Mechanism of the Decoupling (DFT, Figure 6)] The DFT calculation defines Delta E_exfo as the energy to separate the top stack normally from the strained bottom layer; this is a cleavage energy, not the interlayer shear force constant probed by the S1 mode. A vanishing shear restoring force does not require a negative cleavage energy, because the two quantities probe different directions of the interlayer potential. Moreover, the calculated threshold strain for negative Delta E_exfo is above 3.5%, whereas the measured biaxial strain in the adhered layer is 1.1-1.5% (Section 'High-Frequency Raman Spectra'); the factor-of-two-to-three gap is acknowledged in the text but not bridged. Therefore the statement in the abstract and conclusions that biaxial strain is 'identified as the driving factor' and 'the primary cause' is an extrapolation, not a quantitative result. Please provide a calculation that directly links the measured strain range to the shear force constant K_alpha (e.g., phonon calculations for strained bilayers), or explicitly reframe the strain attribution as a hypothesis.
[Linear Chain Model and Decoupling Effect, Eq. (3) and Figure 4d] The modified linear chain model attributes the entire S1 shift on Au to a single parameter K_alpha at the first interface while fixing all other force constants at the pristine value K0. If the S1 shift also reflects a modified effective mass of the bottom layer, a changed stacking registry, or substrate-induced stiffening that is not a simple spring-constant reduction, the fitted values (0.8 K0, 0.5 K0, approximately 0) would be biased. The independent shifted-curve check in Figure 4b supports the phenomenological picture of decoupling and does not depend on the K_alpha fit, but the quantitative claim of 'binding force reduced to nearly zero' rests on this single-parameter assumption. I recommend adding a sensitivity analysis or an ab initio phonon calculation to confirm that the first-interface shear constant is indeed the dominant parameter.
[Mechanism of the Decoupling (DFT setup)] The DFT compares a fully strained N-layer stack with detached unstrained layers, whereas the Raman data (Figure 2c-d) show that the strain is localized almost entirely in the bottom adhered layer, with the upper layers nearly unstrained. The energy balance for exfoliation may be quite different when only the bottom layer is strained. Please either justify the full-stack approximation or repeat the calculation with strain applied only to the bottom layer, as this is the configuration probed experimentally.

minor comments (6)

[Equations (1)-(2)] Equations (1) and (2) in the current manuscript contain corrupted symbols such as 'omega*#' and '9+'; please ensure the equations are typeset correctly.
[Figure 4b caption] The caption refers to 'dashed blue curves' and 'solid blue curves' in a way that is difficult to follow; please label the curves directly in the figure or use a more explicit caption.
[Figure 4d] The legend describing K_alpha values (K0, 0.8 K0, 0.5 K0, 0) and the experimental data points is not fully legible; please enlarge and distinguish the symbols.
[Methods and References] In the Methods section, 'Broyden-Fletcher-Goldfrab-Shanno' should be 'Broyden-Fletcher-Goldfarb-Shanno', and reference 46 has 'Brillonin' instead of 'Brillouin'.
[Raman Methods] The claimed spectral resolution of 2.1 cm-1 and the stated peak-position fitting errors should be reconciled with reported averages such as (377.8 +/- 0.1) cm-1; please state how many spectra and which fitting errors contribute to these uncertainties.
[Introduction] The word 'groundbreaking' in the first sentence of the Introduction is promotional; consider replacing it with a neutral description.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central decoupling claim is supported by new Raman measurements, a fitted model that is explicitly labeled as fitting, and independent checks (B1-mode comparison and bubble-burst imaging).

full rationale

The paper's derivation chain is experimental and self-contained. The central quantity K_alpha is not presented as a prediction; it is explicitly deduced by matching calculated S1 frequencies to measured Raman data via the mLCM (Eq. 3, Fig. 4d), so this is ordinary parameter inference rather than 'fitted input called prediction'. The near-zero K_alpha conclusion for N>=5 is cross-checked by two observations that do not reduce to the S1 fit: first, the B1 modes of N-layer MoS2 on Au match the shifted (N-1)-layer SiO2 curves even though the B1 frequencies were not used to determine the shear K_alpha; second, annealed samples show bubbles forming and bursting at the first interface, exposing monolayers, which directly confirms that the first MoS2-MoS2 interface is the weakest point. The DFT exfoliation-energy calculation is an independent simulation of cleavage energetics; the paper openly acknowledges that its strain threshold is larger than the measured strain, and the DFT quantity is a cleavage energy rather than a shear force constant, but that is a quantitative-support gap, not circular reasoning. Self-citations (refs 1, 20, 27) are used for sample fabrication and strain/doping calibration, not as load-bearing justification for the decoupling mechanism. No step in the derivation is equivalent by construction to its own inputs.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central claim rests on a small number of free parameters: the fitted first-interface force constant K_alpha (the quantitative evidence for decoupling) and the reference spring constants fitted on SiO2. The main assumptions are the linear chain model, strain localization, and the interpretation of Raman shifts. No new physical entities are introduced. The DFT calculation is parameter-free apart from standard functional choices, but its quantitative mismatch with the measured strain limits its support.

free parameters (3)

K_alpha (first-interface spring constant, per thickness) = 0.8 K0 (bilayer), 0.5 K0 (tetralayer), near 0 K0 (N>=5)
K_alpha is varied in the modified linear chain model so that the calculated S1 mode matches the measured Raman frequency on Au for each thickness (Figure 4d). The near-zero value for thick crystals is the quantitative basis for the decoupling claim.
K_s (interlayer shear spring constant on SiO2) = (2.8 +/- 0.1) x 10^19 N m^-3
Obtained by fitting the LCM to the S1 mode frequencies of MoS2 on SiO2 (Figure 4b). The value is consistent with literature, but it is a fit parameter used as the reference K0 in the modified model.
K_b (interlayer breathing spring constant on SiO2) = (8.3 +/- 0.1) x 10^19 N m^-3
Obtained by fitting the LCM to the B1 mode frequencies of MoS2 on SiO2 (Figure 4b). Used only as a reference and consistent with literature estimates.

assumptions (6)

domain assumption The linear chain model with nearest-neighbor interlayer interactions only (Eqs. 1-3) describes the shear and breathing modes.
The model treats each layer as a mass point connected by springs. It is a standard approach for interlayer phonons, but it ignores intralayer degrees of freedom and longer-range interactions.
ad hoc to paper All interlayer force constants except the first interface remain at their pristine value K0 in the modified model.
This is the core assumption of the mLCM in Figure 4c. It is motivated by the observation that strain is localized to the bottom layer, but it is not independently verified for the upper interfaces.
domain assumption The E' Raman shift is caused solely by biaxial strain and the A1g shift solely by doping, with negligible covariance.
The authors state this in the text and cite a previous study (ref 34). If there is notable cross-sensitivity, the quoted strain of 1.1% could be inaccurate.
domain assumption The strain induced by the gold substrate is mostly localized in the bottom adhered layer.
The assignment of the low-frequency E(L) peak to the strained bottom layer and the conclusion that only the first interface is weakened rely on strain localization. This is supported by refs 22 and 33, but not directly measured in this paper.
domain assumption The DFT-D3/PBE functional captures the interlayer van der Waals interactions and exfoliation energetics well enough for the qualitative threshold.
Standard DFT with a dispersion correction is used. The calculated threshold strain of 3.5% may be inaccurate, and the authors acknowledge the simulation is simplified.
domain assumption Bubbles observed after annealing form at the first MoS2-MoS2 interface when they burst and expose a monolayer.
The reasoning is that exposed surfaces are always monolayers, so the bubble must have formed at the first interface. This assumes the top layers remain intact and that bubbles do not also form and burst at higher interfaces without exposing monolayers.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Strain-Induced Decoupling Drives Gold-Assisted Exfoliation of Large-Area Monolayer 2D Crystals." pith.science (2026). https://pith.science/paper/A6RN6FF3

@misc{pith2026241205898,
  author       = {Pith},
  title        = {Pith review of: Strain-Induced Decoupling Drives Gold-Assisted Exfoliation of Large-Area Monolayer 2D Crystals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A6RN6FF3}},
  note         = {Machine review of arXiv:2412.05898}
}

read the original abstract

Gold assisted exfoliation (GAE) is a groundbreaking mechanical exfoliation technique, producing centimeter scale single crystal monolayers of 2D materials. Such large, high quality films offer unparalleled advantages over the micron sized flakes typically produced by conventional exfoliation techniques, significantly accelerating the research and technological advancements in the field of 2D materials. Despite its wide applications, the fundamental mechanism of GAE remains poorly understood. In this study, using MoS2 on Au as a model system, we employ ultralow frequency Raman spectroscopy to elucidate how the interlayer interactions within MoS2 crystals are impacted by the gold substrate. The results reveal that the coupling at the first interface between the adhered layer on the gold substrate and the adjacent layer, is substantially weakened, with the binding force being reduced to nearly zero. This renders the first interface the weakest point in the system, thereby the crystal preferentially cleaves at this junction, generating large area monolayers with sizes comparable to the parent crystal. Biaxial strain in the adhered layer, induced by the gold substrate, is identified as the driving factor for the decoupling effect. We establish the strain-induced decoupling effect as the primary mechanism of GAE, which could also play a significant role in general mechanical exfoliations.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

[6]

Calculated exfoliation energy and its variation with strain and number of layers

DFT simulation of the exfoliation energy of MoS2 crystal. Calculated exfoliation energy and its variation with strain and number of layers. Negative/positive values of ∆𝐸CDE indicate that monolayer exfoliation is energetically preferable/unfavorable. As a final note, it is worthy to discuss the potential contribution of electrostatic force to the decoupli...

work page 2018
[13]

R.; Turner, A

(47) Billeter, S. R.; Turner, A. J.; Thiel, W. Linear Scaling Geometry Optimisation and Transition State Search in Hybrid Delocalised Internal Coordinates. Physical Chemistry Chemical Physics 2000, 2, 2177–2186. (48) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correcti...

work page 2000
[58]

Macroscopic uniform 2D moir\'e superlattices with controllable angles

27 (11) Liu, F.; Wu, W.; Bai, Y.; Chae, S. H.; Li, Q.; Wang, J.; Hone, J.; Zhu, X.-Y. Disassembling 2D van der Waals crystals into macroscopic monolayers and reassembling into artificial lattices. Science 2020, 367, 903–906. (12) Wu, K.; Wang, H.; Yang, M.; Liu, L.; Sun, Z.; Hu, G.; Song, Y.; Han, X.; Guo, J.; Wu, K.; Feng, B.; Shen, C.; Huang, Y.; Shi, Y...

work page Pith review arXiv 2020
[1996]

J.; Pack, J

(46) Monkhorst, H. J.; Pack, J. D. Special Points for Brillonin-Zone Integrations, 1976; Vol

work page 1976
[5410]

Epitaxy of Wafer-Scale Single-Crystal MoS2 Monolayer via Buffer Layer Control

(6) Li, L.; Wang, Q.; Wu, F.; Xu, Q.; Tian, J.; Huang, Z.; Wang, Q.; Zhao, X.; Zhang, Q.; Fan, Q.; Li, X.; Peng, Y.; Zhang, Y.; Ji, K.; Zhi, A.; Sun, H.; Zhu, M.; Zhu, J.; Lu, N.; Lu, Y.; Wang, S.; Bai, X.; Xu, Y.; Yang, W.; Li, N.; Shi, D.; Xian, L.; Liu, K.; Du, L.; Zhang, G. Epitaxy of Wafer-Scale Single-Crystal MoS2 Monolayer via Buffer Layer Control....

work page 2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [6]

Calculated exfoliation energy and its variation with strain and number of layers

DFT simulation of the exfoliation energy of MoS2 crystal. Calculated exfoliation energy and its variation with strain and number of layers. Negative/positive values of ∆𝐸CDE indicate that monolayer exfoliation is energetically preferable/unfavorable. As a final note, it is worthy to discuss the potential contribution of electrostatic force to the decoupli...

work page 2018

[2] [13]

R.; Turner, A

(47) Billeter, S. R.; Turner, A. J.; Thiel, W. Linear Scaling Geometry Optimisation and Transition State Search in Hybrid Delocalised Internal Coordinates. Physical Chemistry Chemical Physics 2000, 2, 2177–2186. (48) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correcti...

work page 2000

[3] [58]

Macroscopic uniform 2D moir\'e superlattices with controllable angles

27 (11) Liu, F.; Wu, W.; Bai, Y.; Chae, S. H.; Li, Q.; Wang, J.; Hone, J.; Zhu, X.-Y. Disassembling 2D van der Waals crystals into macroscopic monolayers and reassembling into artificial lattices. Science 2020, 367, 903–906. (12) Wu, K.; Wang, H.; Yang, M.; Liu, L.; Sun, Z.; Hu, G.; Song, Y.; Han, X.; Guo, J.; Wu, K.; Feng, B.; Shen, C.; Huang, Y.; Shi, Y...

work page Pith review arXiv 2020

[4] [1996]

J.; Pack, J

(46) Monkhorst, H. J.; Pack, J. D. Special Points for Brillonin-Zone Integrations, 1976; Vol

work page 1976

[5] [5410]

Epitaxy of Wafer-Scale Single-Crystal MoS2 Monolayer via Buffer Layer Control

(6) Li, L.; Wang, Q.; Wu, F.; Xu, Q.; Tian, J.; Huang, Z.; Wang, Q.; Zhao, X.; Zhang, Q.; Fan, Q.; Li, X.; Peng, Y.; Zhang, Y.; Ji, K.; Zhi, A.; Sun, H.; Zhu, M.; Zhu, J.; Lu, N.; Lu, Y.; Wang, S.; Bai, X.; Xu, Y.; Yang, W.; Li, N.; Shi, D.; Xian, L.; Liu, K.; Du, L.; Zhang, G. Epitaxy of Wafer-Scale Single-Crystal MoS2 Monolayer via Buffer Layer Control....

work page 2024