{"id":"0049b381-808c-4532-bb18-533ea45737fe","arxiv_id":"2412.12703","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A WSe2 monolayer's photoluminescence reveals reproducible switching of AB/BA ferroelectric domains in an adjacent hBN interface.","lead":"This paper uses a neighboring WSe2 monolayer to read out, with light, whether a hexagonal boron nitride interface has switched its electric polarization. That enables non-destructive optical detection of sliding ferroelectric switching, a step toward compact memory devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified","rationale":"The reader's weakest assumption identifies a genuine caveat: the PL ratio's interpretation as a doping probe and the single-gate coupling of field with doping. However, the manuscript provides a control sample without FE domains that exhibits no shift, and the hysteresis is measured at zero readout bias, which isolates the remanent polarization state. The central demonstration—optical detection of FE switching—does not require a quantitative separation of the poling field from the electrostatic doping during the pulse; it only requires that the remanent state after poling is detectable. The sharp, reproducible transitions at +0.065 and -0.098 V/nm and the domain maps after opposite poling are strong evidence. The paper is honest about the imprint and the need for double-gated devices in future work. The claimed novelty is incremental but the evidence is sound. We therefore see no reason to alter the reader's ACCEPT verdict; the concrete test would further strengthen the interpretation but is not necessary for the central claim.","tokens_in":13278,"tokens_out":9528,"duration_ms":89856,"concrete_test":"As an additional check, perform the same PL hysteresis measurement on a dual-gated device with a top gate over the WSe2, holding the WSe2 at a constant carrier density (e.g., at the charge neutrality point) while pulsing the back gate to switch the FE interface. If the abrupt PL-ratio transitions persist at fixed WSe2 doping, the observed hysteresis is unambiguously caused by the FE polarization rather than by electrostatic doping during the poling pulse.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that a remote WSe2 monolayer can optically read the polarization state of a sliding ferroelectric hBN interface—is supported by multiple independent lines of evidence: (i) PL maps correlate with KFM domain images; (ii) the ~200 mV shift in the gate-voltage dependence of the trion onset appears only after applying fields above the expected switching threshold and is absent in a control device without a FE interface (Fig. S4); (iii) hysteresis loops of the PL ratio at zero read-out bias, measured at a fixed spot, show sharp, reproducible transitions at fields consistent with coercive fields of sliding FE hBN; (iv) the domain maps after opposite poling show the expected change in the BA/AB domain areas. The single-gate geometry indeed couples the poling field to electrostatic doping, but the measurement at 0 V and the control sample separate the remanent FE polarization state from an instantaneous field effect. The remaining ambiguity—charge trapping at the FE interface mimicking polarization—is not fully excluded, but the control sample plus the known ferroelectric behavior of twisted hBN make this an unlikely explanation. We therefore find no load-bearing concern that changes the reader's verdict.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports the optical detection of ferroelectric switching in a sliding hBN interface via the photoluminescence (PL) of an adjacent WSe2 monolayer. The authors fabricate a graphite-FE hBN-WSe2 stack, identify AB and BA ferroelectric domains by Kelvin force microscopy (KFM), and show that the PL intensity ratio of the negative trion to the neutral exciton (Xs-/X0) maps the same domains. After applying out-of-plane electric fields above the expected switching threshold, they observe a ~200 mV shift in the gate-voltage dependence of the trion onset, PL maps in which the BA/AB domain areas reverse with poling, and reproducible square hysteresis loops measured at zero bias, with a horizontal imprint of 16.5 mV/nm. A control device without the FE interface shows no such hysteresis. The authors interpret these observations as demonstrating that a remote WSe2 monolayer can serve as a local, non-destructive optical probe of sliding ferroelectric switching in hBN.","tokens_in":13462,"tokens_out":7801,"duration_ms":72444,"significance":"If correct, this result provides a new optical readout modality for sliding ferroelectric hBN, with diffraction-limited spatial resolution and simpler device fabrication than electrical or electron-microscopy probes. The claim is supported by several independent pieces of evidence: KFM-to-PL correlation, the appearance of the gate-voltage shift only after fields above the switching threshold, the absence of hysteresis in a control sample without the FE interface, and the sharpness of the switching transitions at fields consistent with literature values. The approach builds on the authors' prior work (ref 35) linking FE domains to doping, and no free parameter is fitted to produce the central observations. The main limitations—single-gate geometry coupling field and doping, incomplete domain switching, and the lack of a full microscopic explanation of the imprint—are explicitly acknowledged and do not undermine the central demonstration.","major_comments":[],"minor_comments":[{"comment":"The interpretation that the optical changes reflect ferroelectric switching would be further strengthened by post-switching KFM images on the same area; the initial KFM image in Fig. 1(c) is taken before any poling, so a direct confirmation that the PL-detected BA/AB assignment after poling matches the stacking would remove residual ambiguity about trapped-charge effects. The authors should at least state this limitation explicitly where the switching maps are presented.","section":"Fig. 2 and Fig. 3"},{"comment":"The 200 mV gate-voltage shift is stated to be comparable to the KFM surface potential contrast, but the line scan in Fig. 1(d) shows only about 115 mV. The authors should clarify whether the relevant potential at the WSe2 position is expected to be larger than the top-surface KFM contrast, or adjust the comparison accordingly.","section":"Fig. 2(c) and Fig. 1(d)"},{"comment":"The notation for the negatively charged trion is inconsistent (XS-, Xs-, X S-); a single symbol should be used consistently throughout the text and figures.","section":"Throughout"},{"comment":"The caption states the PL spectra were measured under 10 µW excitation, while all other measurements in the paper use 1 µW; please specify whether the ratio Xs-/X0 is power-dependent in the relevant regime, or justify the different excitation power.","section":"Fig. 1(e) caption"},{"comment":"The four hysteresis cycles show a slight decrease in the high-ratio level after successive cycles; a brief comment on the reproducibility and possible fatigue or drift would be useful.","section":"Fig. 3(c)"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is a strong candidate for a letter-style publication in a condensed-matter or materials journal. The central claim is well supported by the data and controls, and the limitations are openly discussed. The minor points above are local and should be easy for the authors to address. No concerns about scope or novelty."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is the first optical detection of switching in sliding ferroelectric hBN, and the evidence is solid. The paper deserves a serious referee.\n\nThe new thing is straightforward: the authors use a remote WSe2 monolayer as a local photoluminescence readout of the AB/BA domain state in a ferroelectric hBN interface, and they actively switch the domains with an out-of-plane field. That wasn't in their earlier static work or anyone else's. The demonstration combines KFM domain imaging, PL maps that reverse after opposite poling, sharp and reproducible hysteresis loops measured at zero readout bias, and a control device without the FE interface. That is a convincing package. The hysteresis loops are square, repeat four times, and the transition fields are consistent with known coercive fields for sliding hBN. They also honestly report that switching is incomplete and that there is an imprint of about 16 mV/nm they cannot fully explain. No free parameters are fit to produce the central result.\n\nSoft spots, in proportion: the single-gate geometry couples the poling field to electrostatic doping in the WSe2, so the 200 mV shift in the gate-voltage axis after opposite poling is interpreted as a change in capacitive coupling rather than directly measured. The measurement at zero bias and the control sample largely separate the remanent polarization from the instantaneous field effect, but trapped charge at the interface mimicking polarization is not fully excluded. That is a minor caveat, not a load-bearing flaw. The imprint discussion is honest but open-ended.\n\nThe paper is well-written, the SI provides useful supporting data, and the claims match the evidence. It is an incremental but real advance for the sliding ferroelectric community. I would send it to peer review: a good referee can probe the control sample details and the capacitive model, but the core observation looks solid.","headline":"First optical readout of sliding ferroelectric switching in hBN, with a solid multi-technique evidence package; send to review.","tokens_in":13980,"tokens_out":2009,"would_cite":true,"duration_ms":19596,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The photoluminescence ratio of trions to excitons in a WSe2 monolayer is demonstrated to be an optical readout of ferroelectric switching in an adjacent hBN interface.","keywords":["sliding ferroelectricity","hexagonal boron nitride","van der Waals heterostructure","WSe2 monolayer","photoluminescence","trion","ferroelectric switching","non-volatile memory"],"falsifier":"Measure the same gate-voltage-dependent $X_S^-/X_0$ shift and hysteresis in a double-gated device where the WSe2 monolayer is held at a fixed carrier density while the hBN interface is poled; if the 200 mV shift and the square hysteresis disappear when the doping is fixed, the capacitive-coupling interpretation is wrong. Alternatively, a control stack with an identically prepared hBN interface but no ferroelectric domains should show no shift or hysteresis.","tokens_in":13118,"feed_emoji":"🔬","tokens_out":9011,"duration_ms":73043,"temperature":0.7,"pith_summary":"This paper reports that a single layer of WSe2 placed a few nanometres from a ferroelectric interface in hexagonal boron nitride can act as an optical readout of that interface's polarization state. In a stack of two parallel hBN layers, AB and BA stacking produce opposite out-of-plane electric polarizations, and an applied field can switch between them by sliding the layers. The authors show that the photoluminescence ratio of negative trions to neutral excitons in the WSe2 monolayer follows the local polarization, so optical maps reveal the ferroelectric domains and their switching. This matters because it offers a non-destructive, local, contact-free probe of sliding ferroelectricity, potentially useful for compact non-volatile memory and for studying switching dynamics.","feed_headline":"WSe2 photoluminescence reveals ferroelectric switching in hBN","feed_subtitle":"A trion/exciton PL ratio gives a contact-free local readout of ferroelectric switching in hBN.","key_machinery":"The central object is the photoluminescence ratio of the negatively charged trion (the singlet negative trion $X_S^-$) to the neutral exciton $X_0$ in the WSe2 monolayer; this ratio grows with electron doping. The mechanism that carries the argument is sliding ferroelectricity at the hBN-hBN interface: AB and BA stacking produce opposite out-of-plane polarizations, and an in-plane sliding of one layer by one bond length switches between them. The ferroelectric polarization imprints a local electron-doping modulation on the adjacent WSe2 via capacitive coupling, shifting the gate potential by roughly $\\pm100$ mV for the two domain types. The PL ratio is measured at 4 K under 1.96 eV excitation with a confocal microscope, giving a spatial resolution of about 500 nm.","core_discovery":"The central claim is that the photoluminescence of a remote WSe2 monolayer can be used to detect polarization switching in a sliding ferroelectric hBN interface. Experimentally, after applying an out-of-plane electric field of $+0.11$ V/nm, the gate-voltage dependence of the trion-to-exciton PL ratio is shifted by about 200 mV compared with after applying $-0.11$ V/nm; the shift is attributed to the FE polarization changing the capacitive coupling between the WSe2 and the graphite back gate, modulating the electron density by roughly $3\\times10^{10}$ cm$^{-2}$. Under zero bias, maps of the $X_S^-/X_0$ ratio show two domain populations whose areas change after opposite poling, and the ratio at a fixed point traces reproducible square hysteresis cycles with switching fields of $+0.065\\pm0.001$ V/nm and $-0.098\\pm0.002$ V/nm, with a 16.5 mV/nm imprint. The paper concludes that the WSe2 monolayer is a local optical probe of the AB/BA ferroelectric state.","pith_inferences":["This suggests the same trion-to-exciton PL readout should work with other monolayer transition metal dichalcogenides, since the coupling is generic to any semiconductor monolayer in capacitive contact with a ferroelectric hBN interface.","Time-resolved or stroboscopic PL measurements could extend this static readout to track domain-wall motion during switching.","The observed 16.5 mV/nm imprint implies an internal field not explained by electrode work-function asymmetry; identifying its microscopic origin could lead to engineered pinning or deterministic switching.","The 200 mV gate-voltage shift is itself a non-volatile electrical signature; combined with optical readout it suggests a memory cell where writing is electrical and reading is all-optical."],"forward_implications":["A non-destructive, local optical readout of ferroelectric domains in hBN is available without contacting the WSe2 layer.","Ferroelectric switching in hBN can be mapped with diffraction-limited resolution, revealing partial switching and pinning at small domains.","The reproducible square hysteresis of the PL ratio offers a route to optically reading non-volatile polarization states in memory devices.","Because the probe is the PL ratio, it must be operated near the WSe2 neutrality point where the FE-induced doping dominates over gate doping.","Double-gated devices should allow independent control of electric field and doping, enabling studies of switching dynamics and domain-wall speed."],"supporting_citations":[{"why":"Establishes stacking-engineered ferroelectricity in bilayer boron nitride and reports the switching field this work uses as the threshold for poling.","marker":"14"},{"why":"Prior demonstration that AB/BA hBN domains n- or p-dope a remote WSe2 monolayer; supplies the trion/exciton readout principle.","marker":"35"},{"why":"Shows the electrostatic moiré potential a twisted hBN interface creates in TMD monolayers, connecting the hBN polarization to WSe2 doping.","marker":"34"},{"why":"Reports interfacial ferroelectricity by van der Waals sliding in hBN and the KFM surface-potential contrast used here to identify domains.","marker":"16"},{"why":"Documents charge-polarized superlattices in marginally twisted hBN, supporting the domain structure and pinning behavior seen in the optical maps.","marker":"15"},{"why":"Operando microscopy of polar domain dynamics in twisted homobilayers; used for comparison of partial switching and pinning at corners.","marker":"21"},{"why":"Theoretical prediction that a nearly parallel hBN interface creates a superlattice potential in a TMD monolayer, motivating the coupled heterostructure.","marker":"33"}],"fun_headline_variants":["Remote WSe2 probe reads hBN ferroelectric state","Optical readout of sliding ferroelectric in hBN","WSe2 glows differently when hBN flips polarity","HBN ferroelectric switching seen through WSe2 light","WSe2 PL ratio tracks hBN ferroelectric domains"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The photoluminescence ratio $X_S^-/X_0$ is an undistorted local measure of electron doping in the WSe2 monolayer, and the doping difference between AB and BA domains comes from the hBN ferroelectric polarization rather than from trapped charges, laser-induced effects, or transfer strain.","fun_headline_variants_meta":{"raw":{"variants":["Remote WSe2 probe reads hBN ferroelectric state","Optical readout of sliding ferroelectric in hBN","WSe2 glows differently when hBN flips polarity","HBN ferroelectric switching seen through WSe2 light","WSe2 PL ratio tracks hBN ferroelectric domains"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000659,"raw_usage":{"total_tokens":3004,"prompt_tokens":921,"completion_tokens":2083,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":537,"completion_tokens_details":{"reasoning_tokens":1997}},"tokens_in":537,"tokens_out":2083,"duration_ms":13305,"temperature":1.0,"reasoning_tokens":1997,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:48:37.084417+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same gate-voltage-dependent $X_S^-/X_0$ shift and hysteresis in a double-gated device where the WSe2 monolayer is held at a fixed carrier density while the hBN interface is poled; if the 200 mV shift and the square hysteresis disappear when the doping is fixed, the capacitive-coupling interpretation is wrong. Alternatively, a control stack with an identically prepared hBN interface but no ferroelectric domains should show no shift or hysteresis.","supporting_citations":[],"review_version":1}