{"id":"27d23663-8b2b-4e53-9064-5a191052c36a","arxiv_id":"2501.02337","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Inserting a single MoSe2 monolayer between PtSe2 and graphene flips the sign and roughly triples the THz spin-charge conversion signal, with interface hybridization rather than charge transfer identified as the cause.","lead":"This paper shows that inserting a single layer of MoSe2 between PtSe2 and graphene in an epitaxial stack reverses the polarity and multiplies the intensity of the terahertz signal produced by spin-charge conversion. A reader interested in atomic-scale engineering of spintronic devices would look at this to see whether one-monolayer insertions can be used as a control knob.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Fermi-level assumption in the DFT kappa_yx check is load-bearing; a modest shift could flip the predicted sign.","rationale":"I read the paper's central claim as the demonstration of monolayer-level control of spin-charge conversion, with DFT and Spin-ARPES providing the mechanistic interpretation. The strongest experimental observation, the sign reversal and intensity increase of THz emission upon MoSe2 insertion, is direct and does not depend on the theoretical model. The vulnerable link is the computed kappa_yx tensor: its sign and magnitude are obtained by integrating spin-polarized valence bands over an energy/momentum window whose boundaries depend on the assumed Fermi level and the chosen klim. The paper itself flags the Fermi-level uncertainty, and the positive kappa_yx for PtSe2/MoSe2/Gr rests on the sombrero band lying inside the integration window. A moderate Fermi-level shift could change which bands contribute and therefore flip the predicted sign. This matches the reader's weakest assumption. The concern does not invalidate the experimental observation but it weakens the claim that the DFT calculation is in quantitative agreement with the THz polarity, which supports a conditional rather than an unconditional verdict. Since the reader already recommended CONDITIONAL, my stress-test does not change that verdict.","tokens_in":10368,"tokens_out":6054,"duration_ms":62852,"concrete_test":"Recompute the cumulative kappa_yx in Fig. 4(c,f) with the Fermi level shifted by -0.3, -0.2, -0.1, +0.1, +0.2, and +0.3 eV around the assumed zero, keeping all other parameters fixed, and repeat for klim = 0.4, 0.5, 0.6, 0.7, and 0.8 |Gamma-K|. If PtSe2/MoSe2/Gr remains positive and PtSe2/Gr remains negative over the full range consistent with the ARPES-determined Fermi level (sombrero top ~0.5 eV below E_F), the assumption is not decisive; if either sign flips within that range, the theoretical consistency check is inconclusive and the mechanism claim should be qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The theoretical consistency check in the DFT section assumes the Fermi level sits at zero energy in the band diagrams, even though the paper explicitly states that the exact position is unknown and may differ between PtSe2/Gr and PtSe2/MoSe2/Gr. The reported values -150 <= kappa_yx <= 0 for PtSe2/Gr and kappa_yx ~ 600 for PtSe2/MoSe2/Gr are obtained by integrating spin-polarized bands over a narrow window set by the 1.55 eV photon energy and by choosing klim = 0.6 |Gamma-K|. For PtSe2/MoSe2/Gr, the large positive kappa_yx comes from the hybridized sombrero band whose top is about 0.5 eV below the assumed Fermi level. A Fermi-level shift of even ~0.2 eV, from doping, band-alignment error, or the admitted uncertainty, could depopulate or repopulate that band and change the sign of the integrated kappa_yx. Since the central claim uses this computed sign to explain the observed THz polarity reversal, the Fermi-level and integration-window assumptions are load-bearing for the mechanism attribution, not merely technical details. The raw THz polarity change and the epitaxial growth are not in question; the vulnerable link is the quantitative DFT-based consistency check.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports epitaxial growth of PtSe2/MoSe2/graphene heterostructures and shows, via THz time-domain spectroscopy, that inserting a single MoSe2 monolayer between PtSe2 and graphene reverses the polarity and increases the intensity of the THz spintronic emission by about 3.5 times relative to PtSe2/graphene. Spin-ARPES and DFT calculations are used to attribute the change to two different interfacial mechanisms: charge transfer at the PtSe2/graphene interface and electronic hybridization at the PtSe2/MoSe2 interface. The DFT-based calculation of the spin-charge conversion tensor κ_yx yields opposite signs for the two stacks, which the authors present as qualitative agreement with the measured THz polarity.","tokens_in":10569,"tokens_out":5014,"duration_ms":48481,"significance":"If confirmed, the result would demonstrate monolayer-level control of spin-charge conversion in fully epitaxial, large-area van der Waals heterostructures, which is a notable step toward atomic-scale design of spintronic and THz devices. The paper combines high-quality MBE growth, structural characterization, spin-resolved photoemission, THz emission, and first-principles calculations; the κ_yx calculation is not fitted to the THz amplitudes, so the comparison is an independent consistency check rather than a circular fit. The main caveat is that the computed sign depends sensitively on the assumed Fermi level and integration window, which the authors explicitly acknowledge but do not test for robustness.","major_comments":[{"comment":"The computed sign of κ_yx depends directly on the assumed Fermi level position, which the authors set to zero energy in the band diagrams while noting that the exact position is unknown and may differ between PtSe2/Gr and PtSe2/MoSe2/Gr. The large positive κ_yx for PtSe2/MoSe2/Gr arises from the hybridized sombrero band whose top sits about 0.5 eV below the assumed Fermi level; a shift of roughly 0.2 eV, well within plausible doping or band-alignment uncertainty, could depopulate that band and reverse the sign of the integrated κ_yx. Because the sign agreement is a central consistency check for the proposed mechanism, the authors should provide a sensitivity analysis of κ_yx as a function of Fermi energy (for example over at least ±0.3 eV) and of the integration cutoff k_lim, and demonstrate that the sign difference between the two stacks is robust within the physically relevant range.","section":"DFT calculations (Eq. (1) and Fig. 4)"},{"comment":"The claim that the PtSe2/MoSe2/Gr bilayer emits a THz electric field approximately 3.5 times stronger than PtSe2/Gr, with opposite polarity, is presented without error bars, repeated-sample statistics, or an uncertainty estimate. Since the intensity increase is one of the paper's central quantitative findings, the authors should report the mean and standard deviation over at least three independent samples or measurement runs, and specify how the 3.5x ratio was extracted from the time-domain traces.","section":"THz measurements (Fig. 3(c-d))"},{"comment":"The restriction of the κ_yx integration to valence bands within the 1.55 eV photon-energy window is motivated only by the pump photon energy. The authors state that conduction bands show almost no spin polarization, but they do not demonstrate that deeper valence bands contribute negligibly. A justification of the integration window, or an estimate of the contribution from bands outside the window, is needed to establish that the computed values (-150 to 0 for PtSe2/Gr and about 600 for PtSe2/MoSe2/Gr) are not artifacts of the chosen cutoff.","section":"DFT calculations (Fig. 4(b) and 4(e))"}],"minor_comments":[{"comment":"The sentence 'Only more recently have growth methods such as molecular beam epitaxy (MBE) and chemical vapor deposition (CVD) have produced high crystalline quality 2D materials' contains a duplicated auxiliary verb; it should read '...have growth methods ... produced...'.","section":"Introduction"},{"comment":"The symbol v_x^g is used for the group velocity but is not defined in the main text; please define it immediately before or after Eq. (1).","section":"Eq. (1)"},{"comment":"The phrase 'Supplemental Metarial' is a typo and should read 'Supplemental Material'.","section":"Reference 28"},{"comment":"The discussion of the non-magnetic contribution S_NM would benefit from a brief explanation of why the absence of six-fold symmetry is interpreted as evidence for hybridization; as written, the link between symmetry reduction and hybridization is not immediately evident.","section":"Fig. 3(e) and surrounding text"}],"recommendation":"major_revision","confidential_remarks":"The manuscript reports an interesting and potentially impactful observation, and the experimental growth and THz characterization appear solid. The main weakness is the sensitivity of the DFT-based sign check to the assumed Fermi level; this is load-bearing for the mechanism attribution and should be addressed with a robustness analysis. I would also require error bars for the 3.5x intensity ratio before publication. The paper is within the scope of the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Facts first: the paper grows epitaxial PtSe2/MoSe2/Gr and PtSe2/Gr on graphene/SiC, measures THz emission from CoFeB/TMD stacks, and finds that inserting one MoSe2 monolayer flips the THz polarity and increases the amplitude roughly 3.5x. That is a clean, interesting experimental result, and it is new. The comparison with spin-ARPES and DFT is mostly honest: the theory is not fitted to the THz data, the band structures match the measured dispersion, and the two-interface-mechanism story (charge transfer versus hybridization) is plausible.\n\nThe soft spots are real, though not fatal. The THz ratio and sign change come with no quoted error bars or repeated-sample statistics; from the data shown, the signals are large and the polarity reversal is convincing, but quantitative intensity claims need more support. The bigger issue is the kappa_yx calculation. As the authors admit, the Fermi level is unknown and set to zero energy. The stress-test concern is on target: a shift of about 0.2 eV could depopulate the hybridized sombrero band and change the sign of the integrated kappa_yx. The chosen integration window (1.55 eV photon energy) and k_lim = 0.6|Gamma-K| are also somewhat arbitrary. So the \"good agreement\" in sign and magnitude is not yet robust. This does not undo the experimental observation, but it does mean the mechanism attribution is supported rather than proven.\n\nOne more minor point: the spin-ARPES does not directly resolve the spin splitting; the authors use the second derivative to set an upper bound of 0.15 eV. That is a reasonable interpretation, but it is less strong than a direct spin-resolved separation.\n\nWho should read this: anyone working on 2D spintronics, THz emission from vdW heterostructures, or interface engineering of Rashba states. The paper would benefit from a serious referee asking for error bars, sample-to-sample reproducibility, and a sensitivity analysis of kappa_yx to Fermi-level position and k_lim. I would take it for peer review, with a request for major revision on the quantitative comparison. The core claim is strong enough to deserve scrutiny.","headline":"Monolayer MoSe2 insertion flips sign and boosts THz spin-charge conversion; the experimental core is solid, but the quantitative kappa_yx agreement rests on an unverified Fermi-level assumption.","tokens_in":11305,"tokens_out":2114,"would_cite":true,"duration_ms":19978,"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":"Inserting a single monolayer of MoSe2 between PtSe2 and graphene reverses the sign of spin-charge conversion and roughly triples the emitted THz signal.","keywords":["van der Waals heterostructures","spin-charge conversion","THz spintronic emission","Rashba spin splitting","PtSe2","MoSe2","spin-resolved ARPES","density functional theory"],"falsifier":"Measure the actual Fermi-level position or doping of the PtSe2/MoSe2/Gr and PtSe2/Gr stacks, for example by determining the Fermi-surface radius of the sombrero band with high-resolution ARPES or by core-level spectroscopy, and recompute the conversion coefficient over the corresponding energy window; if the sign then disagrees with the observed THz polarity, the proposed mechanism as stated would be falsified.","tokens_in":10150,"feed_emoji":"⚡","tokens_out":9995,"duration_ms":87131,"temperature":0.7,"pith_summary":"This paper reports that spin-charge conversion—the interconversion of spin and charge currents—in an epitaxial van der Waals stack can be switched at the level of a single inserted monolayer. Growing one monolayer of MoSe2 between PtSe2 and graphene changes the THz spintronic emission from a weak, negative-polarity signal to a strong, positive-polarity signal about 3.5 times larger. The authors attribute the switch to a change in how the interfacial Rashba states form: charge transfer from graphene to PtSe2 in the bilayer, versus strong electronic hybridization between PtSe2 and MoSe2 in the trilayer. They support this assignment with spin-resolved photoemission and density-functional calculations of the spin-charge conversion tensor. If correct, the result makes monolayer-level van der Waals stacking a practical design knob for spintronic THz emitters and related spin-orbitronic devices.","feed_headline":"One MoSe2 layer flips THz spin-emission sign and triples intensity","feed_subtitle":"A single MoSe2 monolayer between PtSe2 and graphene reverses spin-charge conversion polarity and boosts THz output.","key_machinery":"The central object is the spin-charge conversion tensor $\\kappa_{yx}$, a momentum-space integral of the spin expectation $\\langle \\sigma_y \\rangle$ times the group velocity $v_x^g$ that quantifies how efficiently a spin current is converted into a charge current. The argument also rests on the 'sombrero' valence band, a Mexican-hat-shaped dispersion at the top of the PtSe2/MoSe2 valence bands whose large Rashba splitting (a momentum-dependent spin splitting caused by broken inversion symmetry at the interface) provides the dominant positive contribution to $\\kappa_{yx}$. Spin-resolved photoemission supplies the direct observation of opposite spin polarization at $\\pm k$, and THz time-domain spectroscopy supplies the experimental response that the tensor calculation is asked to match.","core_discovery":"The central claim is that the sign and amplitude of spin-charge conversion in PtSe2 can be set by choosing which single van der Waals layer sits next to it. In PtSe2/graphene, the Rashba-split spin texture of the highest valence band is weak and the computed conversion coefficient $\\kappa_{yx}$ is small and negative, matching the observed THz polarity; in PtSe2/MoSe2/graphene, the same band (the 'sombrero' band) undergoes strong hybridization with MoSe2, develops a large Rashba splitting, and drives $\\kappa_{yx}$ to large positive values, matching the observed sign reversal and roughly 3.5 times stronger emission. Spin-ARPES confirms spin-momentum locking with opposite spins at $\\pm k$, and DFT shows that the sombrero band sits closer to the Fermi level only when MoSe2 is present. The paper presents this as evidence that the controlling mechanism is electronic hybridization at the PtSe2/MoSe2 interface, distinct from the charge-transfer mechanism at the PtSe2/graphene interface.","pith_inferences":["An independent measurement of the Fermi level in both stacks would test the mechanism more sharply than the assumed zero-energy alignment used in the calculations.","The hybridized sombrero band hints at a TMD analogue of an interfacial two-dimensional electron gas; a testable extension is to choose TMD pairs whose hybridized band crosses the Fermi level and look for metallic interface conductivity.","Because hybridization strength should depend on interlayer registry, a twisted PtSe2/MoSe2 stack could continuously tune the conversion coefficient between the charge-transfer and hybridization regimes, a knob the zero-twist epitaxial samples do not explore."],"forward_implications":["A single-monolayer insert acts as a binary switch: replacing graphene with MoSe2 next to PtSe2 reverses the THz pulse polarity and increases its amplitude roughly threefold.","The same density-functional plus $\\kappa_{yx}$ workflow can be used to screen other TMD pairs for large, sign-controlled spin-charge conversion before growing them.","The similar sombrero hybridization found for PtSe2/WSe2 indicates the mechanism is likely general to PtSe2 on semiconducting TMDs, not limited to MoSe2.","Large-area epitaxial van der Waals stacks, rather than exfoliated flakes, are sufficient to observe monolayer-level control of spin-charge conversion, which eases device integration.","The loss of the expected sixfold nonlinear THz symmetry in the trilayer shows hybridization also changes the electronic symmetry of the stack, with possible consequences for other nonlinear or valleytronic responses."],"supporting_citations":[{"why":"It supplies the MBE growth recipe and the baseline THz spintronic-emission signature of PtSe2 on graphene that this work modifies.","marker":"[5]"},{"why":"It establishes that MoSe2 grows epitaxially on few-layer graphene, underpinning the crystalline quality claim for the insert layer.","marker":"[13]"},{"why":"It directly observes spin-layer locking and the local Rashba splitting in monolayer PtSe2, the spin texture the authors manipulate.","marker":"[27]"},{"why":"It predicts tunable giant Rashba splitting in the PtSe2/MoSe2 heterostructure, the specific interface realized here.","marker":"[31]"},{"why":"It provides the spin-resolved photoemission setup and analysis used to measure the spin texture.","marker":"[30]"},{"why":"It supplies the THz time-domain method and the magnetic/non-magnetic signal decomposition used to isolate spin-charge conversion.","marker":"[33]"},{"why":"It gives the free-electron model and the $\\kappa_{yx}$ band-structure mapping used to quantify the conversion sign and strength.","marker":"[42]"},{"why":"It provides the exchange-correlation functional used in the density-functional calculations.","marker":"[39]"},{"why":"It supplies the van der Waals dispersion corrections used for the layered heterostructures.","marker":"[40]"}],"fun_headline_variants":["Single MoSe2 layer flips spin-charge sign and boosts THz emission","Monolayer MoSe2 controls spin-charge conversion direction and strength","One atom-thick layer determines spin-to-charge conversion in vdW stack","Inserting one MoSe2 layer reverses THz spin emission polarity","Atomic-scale control of spin-charge conversion via single MoSe2 interlayer"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculations that convert the band structure into a sign for spin-charge conversion assume the Fermi level sits exactly at the zero of the computed bands in both stacks; if the real Fermi level lies elsewhere, the predicted sign of the conversion coefficient could change.","fun_headline_variants_meta":{"raw":{"variants":["Single MoSe2 layer flips spin-charge sign and boosts THz emission","Monolayer MoSe2 controls spin-charge conversion direction and strength","One atom-thick layer determines spin-to-charge conversion in vdW stack","Inserting one MoSe2 layer reverses THz spin emission polarity","Atomic-scale control of spin-charge conversion via single MoSe2 interlayer"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000216,"raw_usage":{"total_tokens":1462,"prompt_tokens":1005,"completion_tokens":457,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":621,"completion_tokens_details":{"reasoning_tokens":357}},"tokens_in":621,"tokens_out":457,"duration_ms":4909,"temperature":1.0,"reasoning_tokens":357,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:13:45.764024+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual Fermi-level position or doping of the PtSe2/MoSe2/Gr and PtSe2/Gr stacks, for example by determining the Fermi-surface radius of the sombrero band with high-resolution ARPES or by core-level spectroscopy, and recompute the conversion coefficient over the corresponding energy window; if the sign then disagrees with the observed THz polarity, the proposed mechanism as stated would be falsified.","supporting_citations":[{"cited_title":"Abdukayumov, M","cited_arxiv_id":null,"evidence_quote":"It supplies the MBE growth recipe and the baseline THz spintronic-emission signature of PtSe2 on graphene that this work modifies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It establishes that MoSe2 grows epitaxially on few-layer graphene, underpinning the crystalline quality claim for the insert layer."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It directly observes spin-layer locking and the local Rashba splitting in monolayer PtSe2, the spin texture the authors manipulate."},{"cited_title":"Xiang, Y","cited_arxiv_id":null,"evidence_quote":"It predicts tunable giant Rashba splitting in the PtSe2/MoSe2 heterostructure, the specific interface realized here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It provides the spin-resolved photoemission setup and analysis used to measure the spin texture."},{"cited_title":"Rongione, L","cited_arxiv_id":null,"evidence_quote":"It supplies the THz time-domain method and the magnetic/non-magnetic signal decomposition used to isolate spin-charge conversion."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It gives the free-electron model and the $\\kappa_{yx}$ band-structure mapping used to quantify the conversion sign and strength."},{"cited_title":"Buˇ cko, J","cited_arxiv_id":null,"evidence_quote":"It supplies the van der Waals dispersion corrections used for the layered heterostructures."}],"review_version":1}