{"id":"663958d5-6860-4080-a11a-57432ead11b4","arxiv_id":"2501.03955","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Epitaxial Bi2Se3/WSe2/Co heterostructures combine terahertz nonlinear optics, spin-to-charge conversion, and proximity magnetism, with WSe2 polymorph (1T' vs 2H/3R) governing the emission symmetry.","lead":"Researchers grew wafer-scale stacks of three different atomically thin materials and showed that changing just one atomic layer changes how the stack emits terahertz light, including its magnetic response. The work points toward building multifunctional 'Lego' materials at industrial scale rather than by hand-stacking tiny flakes.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 2-ML WSe2 polymorphism claim rests on faint RHEED lines plus the very two-fold THz symmetry it is used to explain; independent structural identification of the 1T' phase is required.","rationale":"The reader's conditional verdict is appropriate, and the weakest assumption identified by the reader matches the concern I consider most load-bearing: the 2-ML WSe2 sample is the key evidence for polymorph-controlled properties, but its 1T' assignment is supported only by a weak RHEED signature and by the same two-fold THz symmetry it is invoked to explain. The additional reliance on a cos(3φ) subtraction from Bi2Se3 introduces a possible artifact channel, because interface effects can modify the TI contribution in a way that produces a spurious two-fold residual. The magnetic proximity interpretation inherits the same uncertainty, since its anisotropic part is read off after subtracting the isotropic spin contribution and no non-magnetic control is provided. These issues do not invalidate the growth or the thickness-dependent THz observations, but they do mean the central claim should remain conditional until the 2-ML structural phase and the subtraction assumption are independently checked. I therefore see no reason to move the reader's verdict; the recommended action is to request the structural and control measurements before full acceptance.","tokens_in":16484,"tokens_out":8942,"duration_ms":96465,"concrete_test":"Perform atomic-resolution STEM and nano-beam electron diffraction on the same Bi2Se3/WSe2(2ML)/Co stack, quantifying the 1T' versus 2H volume fraction and the in-plane domain orientation distribution at several wafer positions. If the 1T' phase is not dominant in the 2-ML film, or if the domains are equally populated along three equivalent directions, the observed two-fold THz pattern cannot be assigned to single-domain 1T' WSe2 and the polymorph-control claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim is that inserting one monolayer of WSe2 in a different polymorph changes the stack's nonlinear and magnetic response. The lynchpin is the 2-ML sample: its two-fold S+(NL) pattern is attributed to an even-layer 1T' WSe2 polymorph. This attribution is not independently established. In Section I, the structural evidence for 1T' in the 2-ML film is only described as 'faint lines' in RHEED resembling a (2x1) reconstruction; the paper then uses the two-fold THz symmetry as confirmation of the 1T' interpretation (Section II, Fig. 7). If the RHEED signature is ambiguous or the 1T' domains are a minority, the two-fold pattern could instead arise from a symmetry change of the Bi2Se3 surface states induced by the WSe2 overlayer (strain, charge transfer, or interface hybridization), which would support a monolayer effect but not a polymorph effect. The Fig. 7 subtraction of a cos(3φ) Bi2Se3 template makes this concern sharper, because it assumes the TI contribution is unchanged in the heterostructure and adds linearly; any interface-modified TI term would produce a residual two-fold component misassigned to WSe2. The anisotropic magnetic S-(M) component for 1- and 2-ML samples is similarly interpreted as proximity-induced χc(2), but no non-magnetic control is shown, leaving the same artifact channel open. Thus the central polymorph-control claim is only as strong as the independent identification of 1T' in the 2-ML WSe2 layer.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports molecular beam epitaxy growth of wafer-scale Bi2Se3/WSe2/Co heterostructures with WSe2 thicknesses of 1-4 monolayers and uses phase-resolved terahertz emission spectroscopy to separate the emitted signal into nonlinear S+(NL) and magnetic S^{-}(M) components. The authors claim that the WSe2 polymorph—specifically a 1T' phase in the 2 ML sample—controls the symmetry of the nonlinear THz emission (six-fold for 2H/3R, two-fold for 1T'), that the WSe2 layer acts as a spin-transport barrier reducing spin-to-charge conversion, and that a proximity-induced magnetic nonlinearity appears in the 1 and 2 ML samples. Structural characterization includes RHEED, STEM, EDX, and Raman measurements.","tokens_in":16844,"tokens_out":7458,"duration_ms":67915,"significance":"If the polymorph-control claim is established, this is a valuable step toward scalable van der Waals heterostructures combining TI surface-state nonlinearities, TMD second-order nonlinearities, spin-to-charge conversion, and magnetic proximity effects in one wafer-scale stack. The strengths of the paper include the multi-technique structural characterization, the layer-thickness series from 0 to 4 ML, the phase-resolved S+/S^{-} decomposition, and the use of coherent THz emission to access the sign of nonlinear currents. However, the central claim that one monolayer of 1T' WSe2 determines the functional response currently rests on a single 2 ML sample whose 1T' assignment is not fully independent, and the magnetic-proximity interpretation lacks a non-magnetic control. These issues are load-bearing for the abstract's strongest claim.","major_comments":[{"comment":"The polymorph attribution for the 2 ML sample is not independently established. In Section I, the only structural evidence for 1T' is described as \"faint lines\" in RHEED resembling a (2x1) reconstruction (Fig. 2(e,f)), while the two-fold S+(NL) pattern in Figs. 6(c) and 7 is then used to confirm the 1T' interpretation. Because the RHEED signature is weak and because no atomic-resolution STEM of the 2 ML stacking or quantitative domain-fraction analysis is provided, the two-fold pattern could equally arise from a WSe2-induced symmetry change of the Bi2Se3 surface-state response (strain, charge transfer, hybridization) rather than from an even-layer 1T' WSe2 polymorph. This is load-bearing for the paper's central claim; independent identification (quantitative RHEED line profiles, STEM of the 2 ML film, or phase-sensitive SHG/Raman on the WSe2 film before Co capping) is required.","section":"Section I and Section II (Figs. 2, 6, 7)"},{"comment":"The extraction of the \"WSe2-only\" two-fold nonlinearity subtracts from the raw 2 ML data a cos(3φ) term attributed to Bi2Se3 (grey dashed curve in Fig. 7(a)). This assumes both that the Bi2Se3 contribution is unchanged when WSe2 is inserted and that the two nonlinearities add linearly. If the Bi2Se3/WSe2 interface modifies the TI surface states (for example via strain or charge transfer), the residual two-fold component is an artifact of that subtraction. The manuscript should provide a control experiment (for example, a sample in which WSe2 is replaced by a non-nonlinear spacer, or a 2 ML WSe2 film on a non-TI template) or a full angular fit with adjustable Bi2Se3 and WSe2 tensors and reported residuals.","section":"Section II, Fig. 7"},{"comment":"The anisotropic magnetic component obtained after subtracting the isotropic mean from S^{-}(M) for the 1 ML and 2 ML samples is attributed to a proximity-induced magnetic nonlinearity χc(2) of WSe2. No non-magnetic control is shown, and no demonstration is provided that this anisotropic component scales with the Co magnetization, reverses with applied field, or disappears when the ferromagnet is replaced by a non-magnetic metal. Without such a control, the same interface-modified TI contribution that could explain the nonlinear two-fold pattern could also leak into the S^{-}(M) channel through imperfect field reversal or magneto-optical effects, undermining the magnetic-proximity claim.","section":"Section II, Fig. 8"},{"comment":"The quantitative trends that support the central claims (increase of S+(NL) with WSe2 thickness, decrease of S^{-}(M), and the two-fold symmetry of the 2 ML sample) are presented without error bars, repeated measurements, or statistical fits. In particular, the two-fold \"fit\" in Fig. 7 is not quantitatively compared with alternative angular dependencies (e.g., cos(2φ), cos(3φ), or a sum with an isotropic term), so its statistical significance is unclear. Adding measurement repeats and fit residuals with confidence intervals is necessary to support the monolayer-level control claim.","section":"Section II, Figs. 3(c), 5-8"},{"comment":"The abstract's statement that the simple change of one atomic monolayer \"entirely chang[es] its optical, electrical and magnetic properties\" overstates the demonstrated scope. The reported experiments probe THz emission from optical nonlinearities and spin-to-charge conversion; they do not include direct electrical transport or magnetization measurements. The wording should be narrowed to the properties actually measured (THz nonlinear emission symmetry, spin-to-charge conversion efficiency, and inferred magnetic nonlinearity) or supported by additional transport and magnetometry data.","section":"Abstract and Conclusions"}],"minor_comments":[{"comment":"There are several typos: \"epitaxal\" should be \"epitaxial\", \"struture\" should be \"structure\", \"the the\" appears in the Section II text, and \"cos2(ϕ)\" should be typeset as cos²φ.","section":"Introduction and Section I"},{"comment":"The table's nonlinear and magnetic azimuthal curves are not defined in enough detail; the red/blue sign convention appears only in the Supp. Info. and should be stated in the caption.","section":"Table I"},{"comment":"Panels (b) and (c) are described as a comparison of 1 and 2 ML, but the panel labels are not visible in the caption; please label each panel with the corresponding WSe2 thickness.","section":"Figure 8"},{"comment":"The statement that the miscut axis was found at φ ≈ 150° by x-ray diffraction is not supported by any data or citation; provide the XRD measurement or a reference.","section":"Section I"},{"comment":"The phrase \"1T′ WSe2 likely grows in three domains, oriented every 120°\" is an important assumption for the domain-orientation argument but is presented without a citation or quantitative evidence; it should be supported.","section":"Section II, Fig. 7"},{"comment":"The tensor coefficients d11, d12, and d26 are not defined; please specify the coordinate frame and the components of d_ij they refer to, and check the index convention against the point group C_s.","section":"Eq. (2)"}],"recommendation":"major_revision","confidential_remarks":"The main risk is that the paper's most striking claim (monolayer polymorph control) rests on a single 2 ML sample and a subtraction whose assumptions are untested. The authors should be required to provide an independent structural probe and a non-magnetic control before publication. This is a major-revision issue, not a rejection issue; the growth methodology and THz spectroscopy are otherwise of high quality and would make a strong contribution if the lynchpin attribution is secured."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper is worth a serious look. The genuinely new thing: wafer-scale epitaxial Bi2Se3/WSe2/Co stacks, grown by MBE, with phase-resolved THz emission used to separate nonlinear and spin-current contributions. The monolayer-level control of the WSe2 spacer is real, and the observation of a two-fold nonlinear pattern in the 2-ML sample—right where 2H stacking would be centrosymmetric and silent—is a striking result. The authors also extract an anisotropic magnetic contribution from the S± decomposition that they attribute to proximity-induced χc(2). If correct, that is a nice advance: THz emission as a phase-sensitive probe of polymorph symmetry on macroscopic areas, beyond exfoliated flakes and beyond SHG intensity-only measurements. The growth and structural work (RHEED, STEM, EDX, Raman) is plausibly careful, and the comparison of different WSe2 thicknesses is the right kind of control series.\n\nThe soft spots are all clustered around the interpretation of the 2-ML sample. The structural evidence for 1T' WSe2 is thin: the paper says 'faint lines' in RHEED resemble a reconstruction, and then uses the two-fold THz pattern as confirmation. That is a circularity risk. The layer-parity argument helps—even-layer 2H should have no second-order response, so something non-centrosymmetric is present—but it does not uniquely prove 1T' over, say, an interface-modified Bi2Se3 contribution. The Fig. 7 subtraction assumes the Bi2Se3 cos(3φ) component is unchanged in the heterostructure and adds linearly; if the WSe2 overlayer strains or dopes the TI surface, the residual two-fold pattern could be an artifact. I would want either atomically resolved STEM with clear 1T' contrast, or SHG data from the same 2-ML sample, or a thicker-series control showing the 1T' signature disappears where the polymorph mix changes.\n\nTwo more moderate issues. First, there are no error bars or statistics anywhere; peak-to-peak values are shown as single points. For a claim about a ten-times-smaller magnetic nonlinearity, that matters. Second, the proximity magnetism interpretation would be much stronger with a non-magnetic control (e.g., the same stack with a non-magnetic cap or with the Co replaced by Au) to show the anisotropic S− component is not an artifact of the field-reversal decomposition. Also, the abstract says 'electrical' properties change, but no transport data appears; that overclaim should be trimmed.\n\nWho is this for? People working in THz spintronics, vdW heterostructure growth, and nonlinear optics in 2D materials. The paper is not flawless, but the core observation—monolayer-controlled THz symmetry from wafer-scale epitaxial stacks—is novel and plausible enough to deserve referee time. I would accept it for peer review and ask for the 1T' confirmation, error bars, and a control. My own verdict is conditional: the polymorphism interpretation is the load-bearing part, and right now it is supported but not proven.","headline":"A solid wafer-scale vdW heterostructure paper with a genuinely new THz nonlinearity probe, but the central 1T' polymorphism claim in the 2-ML sample needs stronger independent structural confirmation before it fully lands.","tokens_in":17481,"tokens_out":2092,"would_cite":true,"duration_ms":23678,"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":"This paper reports that replacing one atomic monolayer of WSe2 in a wafer-scale Bi2Se3/WSe2/Co stack changes the stack's THz nonlinear, spin, and magnetic response, with the 1T′ polymorph producing a two-fold emission symmetry and a…","keywords":["van der Waals heterostructures","topological insulator","WSe2 polymorph","1T′ phase","terahertz emission spectroscopy","spin-to-charge conversion","magnetic proximity effect","second-order nonlinearity"],"falsifier":"Grow the same WSe2(2 ML) film on Bi2Se3 but stop before depositing Co, and measure its polarization-resolved second-harmonic generation: if the bare film does not show a dominant two-fold nonlinear pattern matching the 1T′ formula, or if a film that lacks the faint RHEED (x2) reconstruction still shows the two-fold THz pattern in the full stack, the polymorph assignment and the subtraction that reveals it are wrong.","tokens_in":16283,"feed_emoji":"⚡","tokens_out":9349,"duration_ms":82312,"temperature":0.7,"pith_summary":"This paper shows that wafer-scale, epitaxially grown stacks of a topological insulator (Bi2Se3), a transition-metal dichalcogenide (WSe2), and a ferromagnet (Co) can host several functions at once: tunable terahertz optical nonlinearities, spin-to-charge conversion, and magnetic proximity effects. Its central claim is that the crystal polymorph of the WSe2 spacer, which changes with spacer thickness at the monolayer level, controls the symmetry of the emitted terahertz current. A two-monolayer WSe2 film takes the 1T′ polymorph, whose reduced symmetry gives a two-fold nonlinear emission pattern rather than the six-fold pattern of the 2H and 3R phases, and the same reduced symmetry produces a proximity-induced magnetic nonlinearity about ten times smaller than the intrinsic one. If correct, this makes single-monolayer polymorph control a practical way to engineer multifunctional two-dimensional devices on technologically relevant scales, moving beyond exfoliated flakes.","feed_headline":"One WSe2 monolayer controls THz, spin and magnetic response","feed_subtitle":"In wafer-scale 2D stacks, 1T' WSe2 turns THz emission two-fold and adds a proximity magnetic term.","key_machinery":"The load-bearing mechanism is phase-resolved coherent THz emission spectroscopy combined with magnetic-field reversal. Recording the THz pulse at +B and −B and taking the sum $S^+(\\mathrm{NL}) = [S(+B)+S(-B)]/2$ and difference $S^-(\\mathrm{M}) = [S(+B)-S(-B)]/2$ separates the nonlinear current, which does not flip with the field, from the spintronic current, which does. On top of this, the paper uses polymorph-dependent second-order susceptibility formulas: $\\chi_i^{(2)} = d_{26}\\cos 3\\phi$ for the 2H/3R phases and $\\chi_i^{(2)} = d_{11}\\cos^3\\phi + (d_{12}+2d_{26})\\cos\\phi\\sin^2\\phi$ for 1T′ WSe2, with the total nonlinearity written $\\chi_{\\mathrm{total}}^{(2)} = \\chi_i^{(2)} \\pm \\chi_c^{(2)}$, where $\\chi_c^{(2)}$ is a magnetic, field-reversing component arising from proximity with Co. These symmetry formulas are what convert a measured azimuthal THz pattern into an identification of the WSe2 polymorph and of the magnetic proximity effect.","core_discovery":"The core discovery is that the simple exchange of one atomic monolayer of WSe2 inside a Bi2Se3/WSe2/Co stack changes the stack's optical, electrical, and magnetic response, and that the change can be read directly from phase-resolved terahertz emission. Decomposing the emitted THz field into a magnetic component (from spin-to-charge conversion) and a nonlinear component (from $\\chi^{(2)}$ optical rectification) shows that 1, 3, and 4 monolayers of WSe2 give the six-fold azimuthal symmetry expected for 2H/3R stacking, while the 2-monolayer sample shows a clear two-fold pattern. The authors attribute this two-fold pattern to the 1T′ polymorph, whose even-layer form breaks inversion symmetry and is described by $\\chi_i^{(2)} = d_{11}\\cos^3\\phi + (d_{12}+2d_{26})\\cos\\phi\\sin^2\\phi$. The magnetic component is isotropic for the pure TI and for 3–4 monolayer barriers, but for 1 and 2 monolayers it acquires a two-fold anisotropy with the opposite phase of the intrinsic nonlinearity; since WSe2 is not itself magnetic, the authors conclude this is a c-type magnetic nonlinearity induced by proximity to Co. This is presented as the first demonstration of these combined effects in large-area epitaxial heterostructures rather than in exfoliated stacks.","pith_inferences":["Inference: if the two-fold magnetic nonlinearity is genuinely proximity-induced, the same c-type signal should appear in other 1T′-TMD/ferromagnet pairs, and its phase should track the magnetization direction relative to the crystal axis; this can be checked in remanent magnetic states.","Inference: the monolayer-level symmetry switch should also show up in transport-based spin-orbit torque or spin-pumping measurements on the same stacks, where 1T′ WSe2 would yield a low-symmetry angular dependence distinct from the isotropic TI response.","Inference: the data imply a growth strategy for maximizing both nonlinear and spintronic output: multiple periods of Bi2Se3/WSe2/Co, or inversion of the stacking order to put the TMD above the ferromagnet, could combine large nonlinear volume with efficient spin injection."],"forward_implications":["Wafer-scale epitaxial van der Waals stacks can support several functionalities simultaneously, so multifunctional devices do not have to rely on exfoliation and manual stacking.","THz emission spectroscopy can act as a non-contact probe of monolayer parity and polymorph over macroscopic areas, distinguishing the two-fold 1T′ response from the six-fold 2H/3R response.","Adding or removing a single WSe2 monolayer changes the symmetry of the emitted THz waveform, allowing layer thickness to program the nonlinear response of the stack.","The two-fold magnetic nonlinearity of the 1T′/Co interface introduces a field-dependent THz contribution that is absent in higher-symmetry stacks, offering a magnetic handle on nonlinear emission.","Because WSe2 acts as a spin barrier, increasing its thickness trades away spin-to-charge conversion amplitude while increasing the nonlinear contribution, a design constraint for future stacks."],"supporting_citations":[{"why":"Supplies the WSe2 second-order susceptibility value and the layer-parity behavior used to interpret the nonlinear THz emission.","marker":"[23]"},{"why":"Provides the reduced-symmetry two-fold second-harmonic form used to identify the 1T′ polymorph's nonlinear response.","marker":"[44]"},{"why":"Supports assignment of the RHEED (x2) reconstruction and the electronic structure to single-layer 1T′ WSe2.","marker":"[50]"},{"why":"Provides the sum-and-difference method that separates nonlinear S+(NL) from magnetic S−(M) THz signals.","marker":"[19]"},{"why":"Establishes the six-fold surface-state nonlinear symmetry of Bi2Se3 that the subtraction in the two-monolayer analysis assumes.","marker":"[60]"},{"why":"Shows that magnetic (c-type) second-harmonic nonlinearities can be non-negligible in two-dimensional materials, supporting the proximity-nonlinearity interpretation.","marker":"[45]"},{"why":"Documents field-reversing nonreciprocal second-harmonic generation in magnetoelectric materials, underpinning the c-type $\\chi_c^{(2)}$ description.","marker":"[46]"},{"why":"Demonstrates vicinal-substrate epitaxy producing a dominant domain orientation, explaining the single two-fold domain observed for 1T′ WSe2.","marker":"[30]"}],"fun_headline_variants":["One WSe2 monolayer flips THz, spin, and magnetic responses","Wafer-scale 2D stacks: monolayer thickness dictates multifunctionality","Polymorph switch in epitaxial 2D stacks alters optical, spin, magnetic effects","Single atomic layer steers THz emission, spin conversion, and magnetism","Epitaxial heterostructures: one monolayer change alters optical, spin, magnetic response"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central attribution rests on two premises: that the Bi2Se3 nonlinear contribution in the two-monolayer stack is identical to its standalone six-fold form and simply subtracts, and that the faint RHEED (x2) lines plus the two-fold THz pattern uniquely identify the 1T′ polymorph rather than some other symmetry-breaking mechanism.","fun_headline_variants_meta":{"raw":{"variants":["One WSe2 monolayer flips THz, spin, and magnetic responses","Wafer-scale 2D stacks: monolayer thickness dictates multifunctionality","Polymorph switch in epitaxial 2D stacks alters optical, spin, magnetic effects","Single atomic layer steers THz emission, spin conversion, and magnetism","Epitaxial heterostructures: one monolayer change alters optical, spin, magnetic response"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001176,"raw_usage":{"total_tokens":4934,"prompt_tokens":1095,"completion_tokens":3839,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":711,"completion_tokens_details":{"reasoning_tokens":3735}},"tokens_in":711,"tokens_out":3839,"duration_ms":24303,"temperature":1.0,"reasoning_tokens":3735,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:43:41.523372+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Grow the same WSe2(2 ML) film on Bi2Se3 but stop before depositing Co, and measure its polarization-resolved second-harmonic generation: if the bare film does not show a dominant two-fold nonlinear pattern matching the 1T′ formula, or if a film that lacks the faint RHEED (x2) reconstruction still shows the two-fold THz pattern in the full stack, the polymorph assignment and the subtraction that reveals it are wrong.","supporting_citations":[{"cited_title":"Ribeiro-Soares, C","cited_arxiv_id":null,"evidence_quote":"Supplies the WSe2 second-order susceptibility value and the layer-parity behavior used to interpret the nonlinear THz emission."},{"cited_title":"Beams, L","cited_arxiv_id":null,"evidence_quote":"Provides the reduced-symmetry two-fold second-harmonic form used to identify the 1T′ polymorph's nonlinear response."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supports assignment of the RHEED (x2) reconstruction and the electronic structure to single-layer 1T′ WSe2."},{"cited_title":"Mičica, J","cited_arxiv_id":null,"evidence_quote":"Provides the sum-and-difference method that separates nonlinear S+(NL) from magnetic S−(M) THz signals."},{"cited_title":"Hsieh, J","cited_arxiv_id":null,"evidence_quote":"Establishes the six-fold surface-state nonlinear symmetry of Bi2Se3 that the subtraction in the two-monolayer analysis assumes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that magnetic (c-type) second-harmonic nonlinearities can be non-negligible in two-dimensional materials, supporting the proximity-nonlinearity interpretation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents field-reversing nonreciprocal second-harmonic generation in magnetoelectric materials, underpinning the c-type $\\chi_c^{(2)}$ description."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates vicinal-substrate epitaxy producing a dominant domain orientation, explaining the single two-fold domain observed for 1T′ WSe2."}],"review_version":1}