{"id":"45a31bff-726c-4dad-b878-fb08656fb1e9","arxiv_id":"1908.07843","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Combined lateral and shear force microscopy reveals grain structure in monolayer WS2, and angle-dependent shear signals yield fitted estimates of the monolayer's elastic stiffness constants.","lead":"This paper shows that two scanning force microscopy modes, lateral force and transverse shear microscopy, can map grain boundaries, crystal orientation, and strain in single-layer WS2 grown by chemical vapor deposition. The same angle-dependent shear measurements are used to estimate elastic stiffness constants of the monolayer.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The fitted elastic constants are not identifiable: as printed, Eq. (1) has only two independent angular functions for the four elastic constants plus unknown scale G.","rationale":"The qualitative microstructure findings are reasonably supported by correlative LFM, TSM, PL, and transfer experiments, and the paper is valuable as a microscopy demonstration. The quantitative elastic-constant claim, however, is the load-bearing part of the abstract and is the place where the argument is least secure. The reader identified the unknown scale G and the missing derivation as the weak assumption. My read sharpens this: even granting the general form of Eq. (1), the printed angular basis functions for the first three constants are linearly dependent, so the four reported constants cannot be determined by the fit. The reported values must be inherited, at least in part, from the initial MoS2 DFT guess or from an implicit constraint that is never stated. This is an internal consistency problem rather than merely a calibration problem, and it directly affects the central quantitative claim. The proposed digitized refit with random starts would settle whether the parameter manifold is flat; if it is, the elastic-constant values should be withdrawn or explicitly labeled as non-identifiable model parameters. If the printed equation is only a transcription error, the corrected derivation and a fresh fit are still required. Thus the appropriate verdict remains CONDITIONAL, but the conditions should be strengthened to require an identifiability analysis and a corrected, complete derivation.","tokens_in":10603,"tokens_out":10640,"duration_ms":115925,"concrete_test":"Digitize the angle-dependent TSM data in Fig. 6d and fit Eq. (1) as printed with G, C1111, C3333, C1122, C1133 free, starting from many random initial guesses, including the MoS2 DFT starting values and the reported solution. Compute the Jacobian rank and the residual norm for each converged fit. If different parameter sets give statistically indistinguishable residuals, or if the numerical rank is below the number of fitted parameters, the reported constants are not identified. Separately, re-derive Eq. (1) from the cited Kalihari derivation to determine whether the printed coefficients are miscopied; if the correct angular basis functions break the collinearity, the paper must provide the corrected equation and rerun the fit with uncertainty quantification before the claim can be assessed.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central quantitative claim is that angle-dependent TSM signals determine fourth-order elastic constants of monolayer WS2. This requires that the model in Eq. (1) allow a unique fit of the reported parameters. Even before considering whether the scale factor G is rotation-independent, the printed equation is underdetermined. Writing c = cos(3θ+α) and s = sin(3θ+α), the first three elastic-constant terms are proportional: the second term, 2c^4s - 2cs^4, equals -2 times the first term, cs^4 - c^4s, and the third term, c^4s - cs^4, equals -1 times the first term. Thus those three prefactors can only be recovered as a single linear combination, not as independent constants. The fourth term, c^9 - s^9, adds one more independent angular function. A one-dimensional angle sweep plus the unknown multiplicative G can therefore determine at most two independent combinations, while the paper reports four elastic constants and G. The reported values C1111 ≈ 15000 GPa, C3333 ≈ 7000 GPa, C1122 ≈ 500 GPa, C1133 ≈ -800 GPa, and G = 3.4 mV/GPa are consequently one point on a flat valley of equally good fits, likely constrained by the initial MoS2 DFT guess rather than by the data. The missing Supporting Information derivation is important, but this identifiability failure can be checked directly from the printed equation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a combined lateral force microscopy (LFM) and transverse shear microscopy (TSM) study of CVD-grown WS2 monolayers. It demonstrates that LFM can reveal strain fields and grain boundaries, and that TSM can reveal relative crystallographic orientation, with support from PL mapping, transfer experiments, and SAED. The authors further claim that angle-dependent TSM measurements allow the experimental determination of the fourth-order elastic constants of monolayer WS2, reporting C1111 ≈ 15000 GPa, C3333 ≈ 7000 GPa, C1122 ≈ 500 GPa, and C1133 ≈ −800 GPa, with a scale factor G = 3.4 mV/GPa and the relation C1111 ≈ 2C3333.","tokens_in":10891,"tokens_out":8645,"duration_ms":78051,"significance":"The microstructure characterization part is significant: the combined LFM/TSM approach is non-destructive, rapid, and large-area, and the paper provides a useful table of expected contrasts for different microstructures. The evidence for grain boundary identification (sharp line contrast persisting after transfer, corresponding PL quenching, agreement with SAED) is compelling. If the elastic constant extraction were valid, it would be the first experimental determination of fourth-order elastic constants of monolayer TMDCs and thus of substantial interest. However, as detailed below, the elastic constant extraction from Eq. (1) is underdetermined, so the quantitative claim is not supported.","major_comments":[{"comment":"The displayed Eq. (1) is underdetermined for the claimed parameters. Let A = cos(3θ+α) sin^4(3θ+α) − cos^4(3θ+α) sin(3θ+α). The first three elastic-constant terms are C1111 A, C4444 (−2A), and C1122 (−A), so they collapse into a single coefficient G(C1111 − 2C4444 − C1122) multiplying A. Only the fourth term, C1133 [cos^9(3θ+α) − sin^9(3θ+α)], is an independent angular function. A rotation scan therefore constrains at most two combinations of the five unknowns (C1111, C4444, C1122, C1133, G), so the individual reported values are not identifiable from the data. The 'good agreement' in Fig. 6d is a back-fit, not an independent validation.","section":"Eq. (1), Section 'Elastic constants of WS2'"},{"comment":"There is an inconsistency between Eq. (1) and the reported constants. The equation contains C1111, C4444, C1122, and C1133, whereas the text reports fitted values for C1111, C3333, C1122, and C1133, with no value given for C4444 and no C3333 appearing in the model. Additionally, the derivation of Eq. (1) is placed in the Supporting Information, which is not included in the submitted manuscript, so the assumptions (linear-elastic response, rotation-independent G, hexagonal symmetry reduction) cannot be checked. The authors must provide the full derivation or clearly state any symmetry relations that reduce the number of independent constants.","section":"Eq. (1) and reported fit values"},{"comment":"The sensitivity analysis in Fig. 6d does not support the claimed constraints. Varying C1111 or C3333 while keeping the other constants fixed changes the curve in a way that can be compensated by adjusting the unidentifiable combination G(C1111 − 2C4444 − C1122) and G, so the dashed lines do not demonstrate that the data are sensitive to the individual constants. Furthermore, the paper's caveat about the unknown G only addresses absolute values; the relative relations C1111 ≈ 2C3333, C1111 >> C1122, and C1111 >> C1133 are not established because they rely on the same unidentifiable decomposition.","section":"Figure 6d, sensitivity analysis"}],"minor_comments":[{"comment":"The manuscript contains placeholder text for the DOI and 'Received' lines that should be completed.","section":"General"},{"comment":"Reference [46] in the text is cited as the source of the TSM relation, but the reference list assigns [46] to a different paper than [47]; please verify the citation intention.","section":"References"},{"comment":"The elastic-constant labels in Eq. (1) (C&&&&, C&&88, C&&44) are garbled and should be typeset unambiguously as C1111, C1122, etc., with a definition of the index convention for a two-dimensional crystal.","section":"Eq. (1) notation"},{"comment":"The text states that the measured TSM signal has a 60° periodicity, while the functional form in Eq. (1) depends on cos(3θ+α) and sin(3θ+α), which have 120° periodicity; the relationship between the model and the observed periodicity should be clarified.","section":"Periodicity, Section 'Elastic constants of WS2'"}],"recommendation":"major_revision","confidential_remarks":"The microstructure part of the paper is solid and could form the basis of a publishable manuscript, but the elastic-constants claim is not supported by the presented model and fitting procedure. I recommend that the editor ask the authors to either provide a correct and complete derivation with identified parameters, or to remove or severely qualify the quantitative elastic-constants claim. The current version overstates what the angle-dependent TSM data can determine."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: the qualitative part of this paper is solid, but the quantitative part is not. The authors use combined lateral force and transverse shear microscopy to identify grain boundaries, crystal orientation, and strain in CVD-grown monolayer WS2. That part is backed by correlative PL mapping, transfer experiments, and SAED, and it is genuinely new for TMDCs. The technique works and will be useful.\n\nThe elastic constants are another story. The claim is that angle-dependent TSM measurements yield the fourth-order elastic constants of WS2, with values like C1111 ≈ 15000 GPa and C3333 ≈ 7000 GPa. I checked Eq. (1) directly. The first three angular terms are linearly dependent: the second is -2 times the first, the third is -1 times the first. So the model only has two independent angular shapes, and with the unknown scale G, one angle sweep can determine at most two independent parameter combinations, not four elastic constants plus G. The reported values are one point on a flat valley of equally good fits, probably held up by the initial MoS2 DFT guess. That is a load-bearing flaw. The derivation is only in the absent SI, G is assumed rotation-independent without evidence, and there are no error bars. The elastic constant claim should not be accepted in its present form.\n\nThe microstructure claim holds up. The authors distinguish strain contrast from grain boundaries, and the transfer test is convincing. That part deserves to be published and cited.\n\nRecommendation: send this to peer review, but the reviewers should insist that the quantitative section be fixed. Either provide a correct derivation and a clear statement of what is actually identifiable, or drop the elastic constant claim and refocus on the imaging method. As written, the abstract overstates what the data support.","headline":"The LFM/TSM microstructure mapping is a useful, well-validated contribution, but the elastic constant extraction is underdetermined by the printed equation and should not be trusted as reported.","tokens_in":11472,"tokens_out":2639,"would_cite":true,"duration_ms":25470,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Angle-dependent shear-force microscopy on CVD-grown WS2 monolayers yields experimental fourth-order elastic constants and a nondestructive way to map grain boundaries, crystal orientation, and strain.","keywords":["transition metal dichalcogenides","WS2 monolayers","lateral force microscopy","transverse shear microscopy","grain boundaries","elastic constants","strain fields","chemical vapor deposition"],"falsifier":"Measure the angle-dependent TSM signal on the same WS2 monolayer at two different applied loads or scan speeds; if the fitted elastic constants change, the rotation-independent $G$ premise fails. Alternatively, compare the reported $C_{1111} \\approx 15000\\ \\mathrm{GPa}$ and $C_{3333} \\approx 7000\\ \\mathrm{GPa}$ with independent elastic measurements of a suspended monolayer, such as nanoindentation or Brillouin scattering.","tokens_in":10408,"feed_emoji":"🔬","tokens_out":14034,"duration_ms":188258,"temperature":0.7,"pith_summary":"The paper sets out to show that two contact-mode scanning probe techniques, lateral force microscopy (LFM) and transverse shear microscopy (TSM), can jointly read the microstructure of chemical-vapor-deposited tungsten disulfide (WS2) monolayers. In the images, single-crystal flakes are featureless, strained regions show diffuse friction contrast, and sharp persistent lines mark grain boundaries; TSM adds sensitivity to relative crystal orientation because elastic anisotropy deflects the tip transversely. The quantitative step is angle-dependent TSM: rotating the sample against the scan direction gives a 60°-periodic signal that the authors fit with a linear-elastic formula to obtain fourth-order elastic constants of monolayer WS2, $C_{1111} \\approx 15000\\ \\mathrm{GPa}$, $C_{3333} \\approx 7000\\ \\mathrm{GPa}$, $C_{1122} \\approx 500\\ \\mathrm{GPa}$, $C_{1133} \\approx -800\\ \\mathrm{GPa}$. They argue that, even with an unknown experimental factor $G$, the relative relations $C_{1111} \\approx 2C_{3333}$ and $C_{1111} \\gg C_{1122}, C_{1133}$ stand, giving theorists the first experimental numbers for monolayer WS2 elasticity.","feed_headline":"Shear microscopy extracts WS2 elastic constants from rotation scans","feed_subtitle":"Combined friction and shear imaging also separates grain boundaries from strain in CVD-grown monolayers.","key_machinery":"The load-bearing identity is Eq. (1), a linear-elastic relation previously derived for TSM that writes the transverse shear signal $T$ as a constant $G$ times a combination of the fourth-order elastic constants $C_{1111}$, $C_{3333}$, $C_{1122}$, and $C_{1133}$ multiplied by trigonometric functions of $3\\theta+\\alpha$, where $\\theta$ is the scan angle and $\\alpha$ the grain tilt. The mechanism is rotation: as the sample turns against the scan direction, the threefold symmetric lattice changes the shear stress on the tip, producing the 60°-periodic signal whose amplitude and phase are used to fit the constants. The factor $G$ is meant to absorb applied stress, cantilever-tip geometry, instrument sensitivity, and tip-sample contact area.","core_discovery":"The central discovery is that the transverse shear microscopy signal of a monolayer WS2 flake changes periodically as the flake is rotated relative to the scan direction, with a period of 60°, and that this rotation dependence can be fit by a linear-elastic relation to yield estimates of the fourth-order elastic constants. On the paper's own terms, angle-dependent TSM measurements enable the authors to acquire these elastic constants experimentally, reporting $C_{1111} \\approx 15000\\ \\mathrm{GPa}$, $C_{3333} \\approx 7000\\ \\mathrm{GPa}$, $C_{1122} \\approx 500\\ \\mathrm{GPa}$, and $C_{1133} \\approx -800\\ \\mathrm{GPa}$, with $G = 3.4\\ \\mathrm{mV/GPa}$. The authors also establish a contrast taxonomy: single crystals show no LFM/TSM contrast, strained grains show diffuse LFM contrast that disappears after transfer, sharp LFM lines locate grain boundaries, and TSM reveals orientation differences except for high-symmetry 120° boundaries that are invisible because of the threefold lattice symmetry.","pith_inferences":["A natural extension, not stated in the paper, is to vary the applied load or scan speed during angle-dependent TSM; if the fitted constants change, the assumption that $G$ is rotation-independent would need revision.","The extracted constants concern in-plane shear stiffness, so connecting $C_{1111} \\approx 2C_{3333}$ to conventional in-plane Young's moduli would require a separate elasticity model relating monolayer stiffness to the measured transverse shear response.","The same rotational protocol could be applied to map local orientation fields grain-by-grain over large CVD films, producing statistics on grain-boundary misorientation distributions that growth models could be checked against."],"forward_implications":["Rotational TSM can assign the relative crystallographic orientation of irregularly shaped flakes from the angular shift of fitted $T(\\theta)$ curves, offering a nondestructive alternative to dark-field transmission electron microscopy.","The reported $C_{1111} \\approx 2C_{3333}$ and the dominance of $C_{1111}$ over $C_{1122}$ and $C_{1133}$ provide benchmark values for density-functional and continuum calculations of WS2 monolayer elasticity.","Combining LFM and TSM distinguishes single-grain from multi-grain flakes even when TSM alone is ambiguous, because 120° twin boundaries are invisible to TSM under the threefold symmetry.","Because the contrast mechanism is elastic anisotropy, the same LFM/TSM protocol should apply to other monolayer TMDCs, and the paper demonstrates the approach on CVD-grown MoS2 as well.","Angle-dependent TSM can also reveal minimal grain misorientation, giving quantitative statistics on how neighboring domains are tilted within a flake."],"supporting_citations":[{"why":"Supplies the linear-elastic Eq. (1) connecting the TSM signal to fourth-order elastic constants and rotation angle; the elastic-constant extraction stands on this relation.","marker":"[46]"},{"why":"The paper cites it as the source of the DFT MoS2 elastic constants used as the initial guess for fitting the TSM curves.","marker":"[47]"},{"why":"Establishes that TSM contrast arises from elastic anisotropy in oriented molecular films, the mechanism transferred to WS2.","marker":"[29]"},{"why":"Demonstrates friction anisotropy in graphene with LFM/TSM, providing the imaging background for the microstructure maps.","marker":"[30]"},{"why":"Documents that TMDC grain boundaries are narrow line defects and that dark-field TEM reveals grain orientation, the baseline the optical and force methods are compared with.","marker":"[20]"},{"why":"Reports star-shaped intra-flake grain boundaries in CVD MoS2 with nanoscale Auger mapping, used to assign the sharp LFM lines to grain boundaries.","marker":"[37]"}],"fun_headline_variants":["Shear force rotation scans extract WS2 elastic constants","Angle-dependent TSM measures WS2 fourth-order elastic constants","Combined friction and shear imaging exposes WS2 microstructure","Rotation scans give WS2 elastic constants and strain maps via TSM"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The quantitative claim presupposes that the measured TSM signal is a linear-elastic response described by Eq. (1) with a single factor $G$ that stays constant as the sample rotates; the derivation is placed in the Supporting Information and the paper does not verify that tip geometry, contact area, or instrument sensitivity are actually rotation-independent.","fun_headline_variants_meta":{"raw":{"variants":["Shear force rotation scans extract WS2 elastic constants","Angle-dependent TSM measures WS2 fourth-order elastic constants","Combined friction and shear imaging exposes WS2 microstructure","Rotation scans give WS2 elastic constants and strain maps via TSM"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000893,"raw_usage":{"total_tokens":3841,"prompt_tokens":928,"completion_tokens":2913,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":544,"completion_tokens_details":{"reasoning_tokens":2846}},"tokens_in":544,"tokens_out":2913,"duration_ms":526108,"temperature":1.0,"reasoning_tokens":2846,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:55:26.686362+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the angle-dependent TSM signal on the same WS2 monolayer at two different applied loads or scan speeds; if the fitted elastic constants change, the rotation-independent $G$ premise fails. Alternatively, compare the reported $C_{1111} \\approx 15000\\ \\mathrm{GPa}$ and $C_{3333} \\approx 7000\\ \\mathrm{GPa}$ with independent elastic measurements of a suspended monolayer, such as nanoindentation or Brillouin scattering.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the linear-elastic Eq. (1) connecting the TSM signal to fourth-order elastic constants and rotation angle; the elastic-constant extraction stands on this relation."},{"cited_title":"Kalihari, G","cited_arxiv_id":null,"evidence_quote":"The paper cites it as the source of the DFT MoS2 elastic constants used as the initial guess for fitting the TSM curves."},{"cited_title":"Kasas, G","cited_arxiv_id":null,"evidence_quote":"Establishes that TSM contrast arises from elastic anisotropy in oriented molecular films, the mechanism transferred to WS2."},{"cited_title":"Kalihari, E","cited_arxiv_id":null,"evidence_quote":"Demonstrates friction anisotropy in graphene with LFM/TSM, providing the imaging background for the microstructure maps."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents that TMDC grain boundaries are narrow line defects and that dark-field TEM reveals grain orientation, the baseline the optical and force methods are compared with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports star-shaped intra-flake grain boundaries in CVD MoS2 with nanoscale Auger mapping, used to assign the sharp LFM lines to grain boundaries."}],"review_version":1}