{"id":"e2841087-8dc1-493d-a134-9a3a9628705c","arxiv_id":"1908.01003","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Potassium dosing makes the topmost tetralayer of GaSe behave like a free monolayer, and sulfur alloying rigidly shifts the valence band and photoluminescence energy while preserving the band dispersion.","lead":"Potassium deposited on the layered crystal GaSe changes the electronic band structure of the top layer so it looks like an isolated monolayer, while mixing sulfur into the crystal shifts the band gap without changing the shape of the valence band. The study shows two practical ways to tune a two-dimensional semiconductor's electronic properties and optical color.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"K-decoupling claim is not uniquely supported: at 94 eV the dosed dispersion could be a kz re-sampling artifact from an inner-potential shift, since bulk GaSe already shows a similar bow shape at kz=3.15 Å^-1 and no kz scan was possible after K dosing.","rationale":"I read the paper in good faith and identify the central claim as the K-induced decoupling of the top-most tetralayer. The load-bearing premise is that the observed dispersion change is caused by weaker interlayer coupling rather than the alternatives the paper itself lists. The most concrete and testable version of this concern is the kz ambiguity: the ARPES data for K-dosed GaSe were taken at one photon energy, and the paper's own clean-GaSe data demonstrate that different kz slices already produce either bulk-like or single-layer-like dispersions. Without a kz scan after K dosing, the observed bow-shaped dispersion could be explained by a shift in inner potential or final-state effects rather than by decoupling. This concern is not a rejection of the experimental observation, which is reproducible and well documented; it is a caution about the causal interpretation. The reader's weakest_assumption already identified inner-potential shifts and final-state effects, so I agree with the reader's assessment. The HRSTEM stacking-phase inconsistency in Sec. III A is a serious error, but it pertains to the alloy phase-transition claim rather than to the central K-dosing claim, so I do not make it the primary basis for my concern. The paper's measured 48 meV band inversion differs from the 150 meV of the bulk kz slice, which weakens a pure kz-resampling explanation, but final-state broadening and mixed kz contributions leave the ambiguity unresolved. Therefore a conditional verdict is appropriate, and my analysis does not change the reader's verdict.","tokens_in":12213,"tokens_out":4740,"duration_ms":46622,"concrete_test":"Measure the K-dosed VBM dispersion at multiple photon energies (e.g., 60–120 eV) using fresh K-dosed spots and reduced flux or exposure time to mitigate beam damage. If the bow-shaped dispersion is kz-independent and never coincides with any clean GaSe kz slice, decoupling is confirmed; if it shifts with photon energy and matches a clean kz slice, the observed change is a final-state or inner-potential artifact rather than decoupling.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that potassium decouples the top-most GaSe tetralayer rests on the observed change of the VBM dispersion in Sec. III C (Fig. 4), measured at a single photon energy of 94 eV. The paper's own Fig. 3 shows that bulk GaSe already exhibits an 'inverted sombrero' dispersion resembling a single tetralayer when measured at kz = 3.15 Å^-1 (band inversion ~150 meV), while at kz = 3.54 Å^-1 the dispersion is bulk-like. If K dosing changes the inner potential V0 or the final-state relaxation, the same photon energy would probe a different kz after dosing, so the new bow-shaped dispersion could be a kz re-selection of the bulk band structure rather than a true decoupling. The paper explicitly notes that prolonged exposure prevented photon-energy scans for the K-dosed sample, leaving this ambiguity untested. The alternative mechanisms listed in Sec. III C (surface band bending, chemical reaction, inner-potential shift, final-state effects) are therefore not excluded. The measured 48 meV band inversion differs from the 150 meV bulk kz slice, but final-state broadening or a mixture of kz states could still mimic the observation. A kz-dependent measurement or direct structural evidence for K intercalation is needed to justify the causal 'decouples' language in the abstract and conclusion.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper combines HRSTEM, ARPES, and PL to study GaSe and GaS_xSe_{1-x} alloys. The authors report that in-situ potassium deposition on bulk GaSe transforms the valence-band dispersion from a bulk-like parabola into a bow-shaped ('inverted sombrero') dispersion reminiscent of single-layer GaSe, which they interpret as potassium decoupling the topmost GaSe tetralayer and changing the gap from direct to indirect. They also report, from HRSTEM, a sulfur-concentration-dependent phase transition from AB (ε) stacking in GaSe to AA′ (β) stacking in sulfur-rich alloys, and show from ARPES and PL that alloying rigidly shifts the VBM to higher binding energy with a corresponding blue shift of the photoluminescence, while leaving the dispersion around the VBM largely unchanged.","tokens_in":12387,"tokens_out":4599,"duration_ms":47453,"significance":"If the central claims hold, the work is significant in two respects: it suggests a controllable, surface-based route to monolayer-like electronic structure on a bulk crystal, and it demonstrates that the electronic gap of gallium chalcogenides can be tuned by alloying without substantially altering the VBM dispersion. The paper's strengths include the systematic mapping of the kz-dependent band structure of bulk GaSe, the explicit comparison of measured band inversions against independent DFT and prior ARPES values, and the use of core-level spectroscopy to determine alloy composition. The data are presented with stated energy and momentum resolutions and with error bars on the fitted quantities. However, the causal interpretation of the potassium experiment and the internal consistency of the stacking-phase assignment need attention before the conclusions can be accepted.","major_comments":[{"comment":"The HRSTEM phase assignments are internally contradictory. The text states that Fig. 1(c) is an image of a high-sulfur alloy (x = 0.75) that shows a trigonal lattice with alternating bright and dim columns, 'consistent with the stacking geometry of the ε phase' (i.e., AB stacking). It also states that Fig. 1(g) is pure GaSe, showing a honeycomb mesh with uniform column intensity, 'consistent with AA′ stacking exhibited by the β phase.' These assignments are opposite to the phase boundary stated earlier in the same section, where the ε phase is assigned to x ≤ 0.3 and the β phase to x > 0.3. Pure GaSe should be ε, and a high-sulfur alloy should be β. As written, the evidence contradicts the claimed ε-to-β transition with increasing sulfur content, and this contradiction propagates to the abstract and conclusion. The authors must correct either the sample labels or the phase assignments, and reconcile the interpretation with the known literature phase diagram.","section":"Section III A and Fig. 1"},{"comment":"The central claim that potassium 'decouples the top-most tetra-layer' is not uniquely supported by the data. The K-dosed dispersion is measured at a single photon energy (94 eV), and the paper explicitly notes in Section II that the fully K-dosed sample could not be scanned in photon energy because of beam-induced degradation. Figure 3(c) shows that bulk GaSe already displays a bow-shaped dispersion with a band inversion of (150 ± 10) meV at kz = 3.15 Å^-1, whereas a different kz slice (Fig. 3(d)) is bulk-like. If potassium deposition changes the inner potential V0 (as the authors themselves list in Section III C), the same 94 eV photon energy would probe a different kz after dosing, and the observed bow shape could be a kz re-sampling of the bulk band structure rather than true top-layer decoupling. The alternative mechanisms mentioned in Section III C — surface band bending, chemical reaction, inner-potential shift, and final-state effects — are not excluded. The measured (48 ± 12) meV inversion differs from the bulk kz-slice value, but final-state broadening or a mixture of kz states could still mimic the observation. To support the causal 'decouples' language in the abstract and conclusion, either a kz-resolved measurement of the K-dosed sample or direct structural evidence of potassium intercalation and interlayer separation is needed.","section":"Section III C and Fig. 4"},{"comment":"The statement that the measured (48 ± 12) meV band inversion 'compares well with a value of 80 meV obtained from recent DFT calculations' is an overstatement: the two values differ by approximately 2.7 times the stated experimental uncertainty. The disagreement may be physically meaningful (e.g., due to incomplete decoupling at the surface), but the manuscript should either quantify the comparison more carefully or discuss possible reasons for the discrepancy rather than describing it as good agreement.","section":"Section III C, paragraph on band inversion"}],"minor_comments":[{"comment":"There is a typo in Section I: 'mobilities on the order of of 0.1 cm2V−1s−1' has a duplicated 'of.'","section":"Introduction"},{"comment":"The alloy labels in Fig. 6(a)-(c) are given as 'GaS0.3Se0.7' and 'GaS0.7Se0.3', while the text and Fig. 5 quote compositions of x = 0.27 ± 0.05 and x = 0.61 ± 0.05. The notation should be made consistent, for example by using the fitted values in the figure labels or stating that the labels are nominal values.","section":"Fig. 6 caption and text"},{"comment":"The vertical-axis label 'E_nib' in Fig. 6(c) appears to be a typo for 'E_bin' (or 'Binding energy').","section":"Fig. 6 panel (c)"},{"comment":"The sentence 'The β and ε phases are thought to co-exist for intermediate sulfur content (0.15 < x < 0.35)' is followed by 'our samples do not show experimental evidence for this theory'; the latter phrase is vague and would be more informative if the authors stated, for example, that all measured intermediate-composition samples fell into a single phase or that no phase coexistence was observed within the resolution of the measurements.","section":"Section III A"}],"recommendation":"major_revision","confidential_remarks":"The HRSTEM phase-assignment contradiction is the most immediately actionable issue; if the figure labels are correct, the paper's phase-transition claim may be reversed, and if the text is correct, the figure needs relabeling. Either way, the current version cannot be published as is. The potassium-decoupling claim is also weaker than the abstract suggests; I would encourage the editor to ask for either additional experimental evidence (e.g., kz-dependence or structural measurement) or a significant softening of the causal wording in the abstract and conclusion. The alloy VBM shift and PL correlation are solid and well aligned with the literature."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe thing to know: this paper has two genuinely useful measurements. First, the ARPES data on the kz-dependent band structure of bulk GaSe, and second, the K-dosing evolution showing the VBM dispersion changing from a parabolic bulk-like shape to an inverted sombrero. The alloy part, with a rigid VBM shift tracked against PL, is straightforward and consistent with earlier optical work. The composition analysis from core-level fits is careful, and the comparison of the 48 meV inversion against DFT (80 meV) and a prior monolayer measurement (120–150 meV) is the right way to frame it.\n\nThe soft spots are real but not fatal. The most concrete is in Sec. III A: the text assigns the trigonal (ε-phase) HRSTEM image to the high-sulfur alloy and the honeycomb (β-phase) image to pure GaSe, while the abstract and introduction assert the opposite stacking trend. That is an internal contradiction. It reads like a swapped figure or a typo, but as printed the stacking-phase claim cannot be evaluated. A referee should catch this.\n\nThe bigger conceptual issue is the causal claim about K “decoupling” the topmost tetralayer. The paper’s own Sec. III C lists alternative mechanisms: band bending, chemical reaction, inner-potential shift, final-state effects. The observed dispersion change is consistent with decoupling, and the gradual dosing evolution is suggestive, but there is no direct structural or chemical evidence of where K sits or whether the interlayer distance changes. The stress-test worry about a kz re-sampling artifact is worth stating but I think it is weaker than it sounds: the clean 94 eV data probe a zone center far from the kz=3.15 slice that shows the 150 meV bow, and shifting inner potential by enough to jump between those zones would require an implausibly large V0 change. The 48 meV vs 150 meV difference also argues against a simple re-selection. Still, the authors themselves admit the K-dosed sample prohibited photon-energy scans, so the final-state ambiguity is untested. The conclusion should be softened to match the “consistent with” language used in the results.\n\nWho is this for? Anyone working on III-VI layered chalcogenides, ARPES of 2D materials, or alkali-metal surface doping. The alloy data alone would make it worth a cite. It deserves a real peer review; the HRSTEM error and the over-reach in the conclusion are fixable with revision, but the underlying measurements are valuable.\n\nMy read: conditional accept after revisions. Send it to review.\n\nBest.","headline":"Solid ARPES measurements of GaSe and GaS_xSe_1-x alloys, with a credible alloy gap-shift result, but the headline K-decoupling claim is inferred rather than pinned down and the HRSTEM stacking assignment is internally inconsistent.","tokens_in":13106,"tokens_out":6749,"would_cite":true,"duration_ms":66042,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["79.60.-i","71.20.-b","73.20.At"],"model":"deepseek-v4-flash","headline":"Potassium dosing reshapes bulk GaSe's valence band into a monolayer-like dispersion.","keywords":["gallium selenide","van der Waals materials","angle-resolved photoemission spectroscopy","potassium intercalation","monolayer decoupling","band gap engineering","GaSxSe1-x alloys","stacking polytypes"],"falsifier":"A $k_z$-resolved ARPES scan of the K-dosed surface would settle it: if the top layer is truly decoupled, the inverted-sombrero band should show negligible dispersion along $k_z$, whereas band bending or an inner-potential shift would leave the out-of-plane dispersion essentially intact and only move the bands in energy. Alternatively, cross-sectional scanning transmission electron microscopy of the dosed crystal should reveal potassium atoms sitting in the van der Waals gap or an increased interlayer spacing; if neither is present, the decoupling claim fails.","tokens_in":11909,"feed_emoji":"🧪","tokens_out":4514,"duration_ms":45308,"temperature":0.7,"pith_summary":"This paper asks whether the electronic structure of the layered semiconductor GaSe can be tuned by two knobs: alkali-metal dosing and sulfur alloying. Using angle-resolved photoemission, it argues that depositing potassium on a bulk crystal changes the dispersion around the valence band maximum from the bulk's single parabolic lobe into an inverted-sombrero shape, which is the signature of a single GaSe tetralayer. The authors read this as evidence that potassium decouples the top-most tetralayer from the bulk, so the surface behaves electronically like a monolayer even though the crystal is thick. The same paper shows that alloying with sulfur shifts the valence band rigidly downward and switches the stacking from AB to AA′, which tracks a blue shift in photoluminescence, implying that the gap can be engineered without disturbing the band shape.","feed_headline":"Potassium gives bulk GaSe a monolayer's band shape","feed_subtitle":"A surface tetralayer decouples and the gap flips from direct to indirect; sulfur alloying tunes the gap further.","key_machinery":"The machinery is angle-resolved photoemission spectroscopy combined with photon-energy scans that map the out-of-plane momentum $k_z$, giving the full three-dimensional valence band structure and letting the authors isolate tetralayer-related dispersion. The load-bearing comparison is the inverted-sombrero band inversion at the top of the valence band: its magnitude on K-dosed GaSe, $(48 \\pm 12)$ meV, is benchmarked against a calculated value of 80 meV for the monolayer and a measured value of 120 meV for monolayer GaSe on a substrate. Supporting machinery includes high-resolution scanning transmission electron microscopy to identify stacking phases by atomic-column intensity, core-level photoemission to track potassium-induced chemical changes and alloy composition, and photoluminescence to connect valence-band shifts to the optical gap.","core_discovery":"The central claim is that the valence band maximum of K-dosed bulk GaSe has a dispersion inconsistent with the bulk and consistent with a single tetralayer: instead of one maximum at $\\Gamma$ with effective mass $(1.1 \\pm 0.2)m_0$, the dosed surface has two maxima at $\\pm 0.3$ Å$^{-1}$ with effective mass $(1.7 \\pm 0.2)m_0$ and a local minimum at $\\Gamma$ that sits $(48 \\pm 12)$ meV below them. That inverted-bow dispersion is the shape expected for monolayer GaSe, so the gap should switch from direct in the bulk to indirect in the dosed surface layer. The paper further claims that this change is caused by a strong modification of the top-most GaSe tetralayer, likely through potassium entering the van der Waals gap or reacting with the surface, leading to weaker coupling to the underlying bulk. For the alloys, the paper claims that increasing sulfur content shifts the valence band maximum to higher binding energy while leaving the dispersion and effective mass essentially unchanged, and that the alloy stacking switches from the AB (ε) sequence to the AA′ (β) sequence above roughly 30 percent sulfur.","pith_inferences":["If potassium decouples only the top-most tetralayer, the dosed crystal becomes a platform for studying monolayer physics on a robust bulk substrate, without transfer or encapsulation.","A direct structural test is available: imaging the K-dosed surface with scanning tunneling microscopy or cross-sectional electron microscopy should reveal potassium atoms in the van der Waals gap or an enlarged interlayer spacing, which would confirm the decoupling mechanism the paper infers from the band shape.","The inverted-sombrero criterion could be applied to other alkali metals and other III-VI chalcogenides; if the measured band-inversion energy tracks ionic radius or intercalation propensity, that would independently strengthen the decoupling interpretation.","Because potassium dosing shifts the valence band by only about 0.1 eV without populating the conduction band, an electron-doping strategy that fills the conduction band would be needed to verify the indirect-gap conclusion by directly imaging the conduction-band minimum."],"forward_implications":["Potassium dosing offers a route to monolayer-like electronic structure, including a likely direct-to-indirect gap transition, on a bulk crystal without exfoliating or growing a monolayer.","Sulfur alloying tunes the valence-band edge and the optical gap over a wide compositional range while leaving the dispersion and effective mass around the valence band maximum essentially unchanged.","The stacking phase of GaS$_x$Se$_{1-x}$ alloys can be selected by composition, with the AB-to-AA′ transition occurring near $x \\approx 0.3$.","K-dosed GaSe degrades under prolonged synchrotron exposure, so studies of the dosed surface require gentler probes or lower photon flux than the full photon-energy scans used on pristine GaSe.","The same ARPES strategy can be applied to other layered monochalcogenides to test whether alkali dosing generically produces a decoupled top tetralayer."],"supporting_citations":[{"why":"Supplies the theoretical orbital character and tetralayer dispersion picture used to interpret the VBM and its interlayer coupling.","marker":"[13]"},{"why":"Provides the earlier ARPES data on GaSe that the measured bulk dispersion is stated to be fully consistent with.","marker":"[34]"},{"why":"Supplies the theoretical prediction that bulk GaSe has its conduction band minimum at Γ, supporting the direct-gap assignment.","marker":"[33]"},{"why":"Gives the DFT band-inversion value of 80 meV used to benchmark the K-dosed monolayer-like dispersion.","marker":"[40]"},{"why":"Provides the measured 120 meV band inversion for single-layer GaSe on a substrate, used as a monolayer reference point.","marker":"[41]"},{"why":"Supplies optical absorption data on GaS$_x$Se$_{1-x}$ that the alloy gap trend is consistent with.","marker":"[22]"},{"why":"Cited as a precedent for potassium intercalating into the van der Waals gap of metallic transition-metal dichalcogenides.","marker":"[42]"},{"why":"Cited as a precedent for potassium intercalation in a semiconducting transition-metal dichalcogenide, supporting the proposed decoupling mechanism.","marker":"[43]"}],"fun_headline_variants":["K dosing on GaSe yields monolayer-like indirect band shape","Potassium decouples GaSe surface, inverting band dispersion","GaSe surface after potassium: monolayer band, indirect gap","Sulfur alloying shifts GaSe gap but keeps band dispersion","K surface turns bulk GaSe into a single tetralayer band"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the new dispersion seen after potassium dosing is caused by the top-most GaSe tetralayer becoming electronically decoupled from the bulk, rather than by the alternative mechanisms the paper itself lists, such as surface band bending, chemical reaction, a shift of the inner potential, or altered final-state effects.","fun_headline_variants_meta":{"raw":{"variants":["K dosing on GaSe yields monolayer-like indirect band shape","Potassium decouples GaSe surface, inverting band dispersion","GaSe surface after potassium: monolayer band, indirect gap","Sulfur alloying shifts GaSe gap but keeps band dispersion","K surface turns bulk GaSe into a single tetralayer band"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000264,"raw_usage":{"total_tokens":1646,"prompt_tokens":1030,"completion_tokens":616,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":646,"completion_tokens_details":{"reasoning_tokens":530}},"tokens_in":646,"tokens_out":616,"duration_ms":6495,"temperature":1.0,"reasoning_tokens":530,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:26:14.894119+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A $k_z$-resolved ARPES scan of the K-dosed surface would settle it: if the top layer is truly decoupled, the inverted-sombrero band should show negligible dispersion along $k_z$, whereas band bending or an inner-potential shift would leave the out-of-plane dispersion essentially intact and only move the bands in energy. Alternatively, cross-sectional scanning transmission electron microscopy of the dosed crystal should reveal potassium atoms sitting in the van der Waals gap or an increased interlayer spacing; if neither is present, the decoupling claim fails.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the theoretical orbital character and tetralayer dispersion picture used to interpret the VBM and its interlayer coupling."},{"cited_title":"Plucinski , author R","cited_arxiv_id":null,"evidence_quote":"Provides the earlier ARPES data on GaSe that the measured bulk dispersion is stated to be fully consistent with."},{"cited_title":"Olgu \\'i n , author A","cited_arxiv_id":null,"evidence_quote":"Supplies the theoretical prediction that bulk GaSe has its conduction band minimum at Γ, supporting the direct-gap assignment."},{"cited_title":"Ben Aziza , author V","cited_arxiv_id":null,"evidence_quote":"Gives the DFT band-inversion value of 80 meV used to benchmark the K-dosed monolayer-like dispersion."},{"cited_title":"\\ Chen , author H","cited_arxiv_id":null,"evidence_quote":"Provides the measured 120 meV band inversion for single-layer GaSe on a substrate, used as a monolayer reference point."},{"cited_title":"Rossnagel , author E","cited_arxiv_id":null,"evidence_quote":"Cited as a precedent for potassium intercalating into the van der Waals gap of metallic transition-metal dichalcogenides."},{"cited_title":"Eknapakul , author P","cited_arxiv_id":null,"evidence_quote":"Cited as a precedent for potassium intercalation in a semiconducting transition-metal dichalcogenide, supporting the proposed decoupling mechanism."}],"review_version":1}