{"id":"ff95c8fc-b52d-4262-965d-fbfe5cd72161","arxiv_id":"2608.05059","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"TiSe2's fundamental band gap remains constant from 300 K to 160 K, indicating the charge density wave transition folds bands without opening an electronic gap.","lead":"Using high-resolution photoemission, the authors report that the electronic band gap in TiSe2 stays at about 85 to 100 meV while cooling through the charge density wave transition at 200 K. The result favors a lattice-driven band-folding mechanism over an excitonic gap-opening picture, a long-standing debate in this material.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The normal-phase baseline is not measured by the same method: above TCDW the folded valence band is absent, so the room-temperature gap used for comparison must come from a different k-point/procedure; the claimed constancy across TCDW is therefore not directly established.","rationale":"I read the paper as attempting to settle the excitonic vs lattice-driven CDW debate by showing that the bulk gap is temperature-independent across TCDW. That claim requires a valid normal-phase baseline, and the FVB-based same-EDC method cannot provide it by construction above TCDW. This is not a dispute with the consensus; it is an internal consistency check. The authors do use multiple photon energies and polarizations and acknowledge the 160 K secondary reconstruction, which is good, but they do not explain how the room-temperature EDC at L contains a valence-band edge comparable to the low-temperature FVB. The reader flagged the weak conduction-band feature and fit reliability; I agree with that, but the more logically primary issue is the asymmetry of the measurement across the transition: the observable used below TCDW (FVB-CB) does not exist above it. A feasible fix is to show the normal-phase gap by an explicit two-momentum construction and demonstrate that it matches the low-temperature same-EDC value. Until that is provided, the central claim overreaches its data; however, the concern is addressable and does not require rejecting the paper outright, so the conditional verdict stands.","tokens_in":6928,"tokens_out":8598,"duration_ms":90329,"concrete_test":"Make the 250 K (room-temperature) EDC at L available and specify the feature assigned as the valence-band edge. If no FVB-like valence feature exists in that EDC, recompute the normal-phase gap from the Gamma valence maximum and the L conduction minimum using the same photon energy/geometry and compare it with the low-temperature FVB-CB gap; the constancy claim is supported only if the two values agree within the quoted fit uncertainty. As a cross-check, compare the 160 K FVB binding energy with the 160 K Gamma valence maximum to rule out hybridization shifts of the folded replica.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the comparison of the low-temperature gap with a normal-phase reference, not just the conduction-band fit. The paper states that valence and conduction edges are simultaneously resolved in single EDCs at the same k-point (Figs. 3-4), but the folded valence band (FVB) is a CDW-induced feature: Fig. 2(a) at 250 K shows no FVB at L, and Fig. 4(a) shows it developing only below 200 K. Above TCDW the Gamma-derived valence maximum is not present at L/M, so the room-temperature reference EDC at L cannot supply the same valence-edge reference. If the 250 K gap is instead derived by combining a Gamma (or A) valence band with an L conduction band, the comparison uses a different operational definition of the gap, and the paper's own warning that cross-k comparisons can be unreliable applies to its own normal-phase baseline. The FVB-CB separation below TCDW also need not equal the fundamental indirect Gamma-L gap unless the folded band energy is shown to coincide with the parent Gamma valence maximum; hybridization can shift it. Thus the central constancy claim requires an explicit demonstration that the same band-edge pair is tracked on both sides of TCDW.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports high-resolution ARPES measurements on 1T-TiSe2 across the CDW transition at TCDW ≈ 200 K, using multiple photon energies (119 eV and 95 eV) and polarizations to track the valence and conduction band edges at the L and M points. The authors find that the separation between the folded valence band (FVB) and the conduction band remains in the range 85–100 meV from 300 K down to 160 K, with no abrupt change at TCDW. They interpret this as evidence that the CDW transition does not involve a temperature-driven electronic gap opening, supporting a lattice-driven band-folding scenario of a pre-existing narrow-gap semiconductor.","tokens_in":7163,"tokens_out":5920,"duration_ms":56019,"significance":"If correct, the result is an important experimental constraint on a long-standing debate about the driving mechanism of the CDW transition in TiSe2. The paper’s main strength is the simultaneous resolution of the folded valence band and conduction band in single EDCs at low temperature, which avoids the cross-k referencing problems of earlier works. However, the directness of the central claim is weakened because the normal-phase (room-temperature) gap baseline is not obtained by the same simultaneous-edge method, and the identification of the FVB with the parent valence band maximum is not explicitly verified. The analysis is carefully restricted to T ≥ 160 K to avoid the secondary low-temperature reconstruction, which is a reasonable and clearly stated choice.","major_comments":[{"comment":"The normal-phase gap baseline is not measured by the same method as the CDW-phase gap. In Fig. 2(a), no folded valence band is present at the L point at 250 K, and Fig. 4(a) shows the FVB developing only below 200 K. Therefore the room-temperature EDC at the L point cannot contain both band edges, and the 300 K gap value used in Figs. 4(c,d) must be derived from separate features at different momenta (e.g., the Γ/A valence maximum and the L/M conduction minimum). The paper should explicitly describe how the normal-phase gap is extracted and justify why cross-k referencing is reliable for that baseline, given the paper’s own caution about cross-k comparisons. As written, the claim of a constant gap from the normal phase to 160 K is not directly demonstrated.","section":"Fig. 4(a,c) and 'Gap versus temperature'"},{"comment":"The FVB is assumed to represent the folded parent valence band maximum, so that its separation from the conduction band equals the fundamental indirect Γ-L gap. However, hybridization in the CDW phase can shift folded bands away from their parent energies. The paper does not show a quantitative comparison between the FVB peak position and the Γ/A valence band maximum measured in the same experiment. Without such a check, the measured FVB-CB separation may not equal the fundamental gap. Please report the parent valence band energy and the FVB energy together, or estimate the hybridization-induced shift to justify equating the two.","section":"Fig. 4(a,b) and FVB interpretation"},{"comment":"The conduction band feature is broad and weak, and the reported error bars (standard deviation of fitted peak positions) appear comparable to the total ~15 meV variation in the gap across temperatures. The paper should provide a quantitative detection threshold: what gap opening or enhancement at TCDW would have been observable given the fit uncertainties and the temperature sampling? This would strengthen the claim that the gap is truly constant rather than merely unresolved within the current precision.","section":"Figs. 4(c,d) and error analysis"}],"minor_comments":[{"comment":"The sentence in Section 2 (text accompanying Fig. 4) states that the FVB's binding energy 'remains aligned with the normal phase conduction band'; this appears to be a typo, since the FVB is a valence band and should be aligned with the parent valence band. Please correct.","section":"Introduction/Figure 4"},{"comment":"The paragraph beginning 'Our finding of a nearly constant band gap across TCDW stands in contrast to an earlier ARPES studies...' is repeated almost verbatim in the text near Figure 4 and again in the conclusion section. Remove the duplicate.","section":"Throughout"},{"comment":"There are minor typographical issues: the title has an extra space in 'Charge De nsity Wave', and the abstract uses 'band extreme' where 'band extrema' would be more standard. These should be corrected in the final version.","section":"Title and abstract"}],"recommendation":"major_revision","confidential_remarks":"The paper's central claim depends critically on the normal-phase baseline issue raised in Major Comment 1. If the authors can clearly demonstrate that the room-temperature gap is extracted from the same band-edge pair (or can justify the cross-k approach), the paper could become a solid contribution. The FVB identification issue (Major Comment 2) is also important but may be addressable by a direct comparison with Γ-point data. The paper fits the scope of cond-mat.str-el and the results are potentially of broad interest to the CDW and excitonic insulator communities."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe headline result is that in the CDW phase of TiSe2 the separation between the folded valence band and the conduction band at L/M stays flat at about 85-100 meV from 200 K down to 160 K, and this is measured in single EDCs at the same k-point in two kz planes. That is a meaningful step forward over earlier cross-referenced ARPES, and the authors are careful to avoid the known low-temperature renormalization near 160 K.\n\nThe soft spot is the normal-phase baseline. Above TCDW there is no folded valence band at L/M, so the room-temperature gap cannot come from the same simultaneous-EDC procedure. Figure 2(a) shows no FVB at 250 K, and Figure 4(a) shows it developing only below 200 K. To get a 300 K gap of similar magnitude, they must be combining a valence band measured elsewhere (Gamma or A) with the L/M conduction band, or fitting some weak feature at L/M that is not the same band edge. That is exactly the kind of cross-k comparison the authors argue is unreliable. So the claim that the gap is unchanged 'from the high-temperature normal phase down to 160 K' is not actually demonstrated; what is demonstrated is that the folded-band to conduction-band separation is constant in the CDW phase.\n\nThe second concern is the fit confidence. The conduction band is broad and low-intensity, and the quoted gap variation (85-100 meV) is comparable to the error bars in Figure 4(c,d). The authors do not release raw EDCs, which makes it hard to independently judge the fits. That alone might be fixable with a fuller supplementary. There is also a verbatim duplicated paragraph in the Discussion, which an editor should ask them to fix.\n\nThat said, the paper is a serious contribution to a long-running debate. It directly resolves both band edges in the CDW phase where earlier work used inferred gaps, and it gives a plausible explanation for the discrepancy with gap-widening reports. The Fermi-surface maps and MDCs support the claim that the CDW does not significantly reshape the electronic pockets. If the normal-phase baseline were made defensible—say by explicitly measuring the Gamma valence maximum and L/M conduction minimum with the same photon energy and polarization and showing the FVB sits at the Gamma-derived energy—the central claim would be much stronger.\n\nThis paper deserves a serious referee, but my verdict is conditional: the constancy claim across TCDW needs to be reworked or narrowed. It is a good reading-group paper.","headline":"A careful ARPES study that plausibly shows a constant folded-band-to-conduction-band separation below TCDW, but the claim of a constant gap across TCDW overreaches because the normal-phase baseline is not measured by the same method.","tokens_in":7718,"tokens_out":7598,"would_cite":true,"duration_ms":72583,"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":"TiSe2's fundamental band gap stays constant at 85–100 meV from 300 K down to 160 K, across its 200 K charge density wave transition, ruling out a temperature-driven excitonic gap opening.","keywords":["charge density wave","TiSe2","angle-resolved photoemission spectroscopy","band gap","excitonic insulator","band folding","lattice-driven phase transition","periodic lattice distortion"],"falsifier":"A bulk-sensitive measurement that tracks the same conduction and valence band edges across 200 K — for example temperature-dependent optical absorption on the same crystals, or STM tunneling spectroscopy at the L point — would falsify the claim if it showed the gap shifting by more than the fitting uncertainty as the temperature passes through TCDW.","tokens_in":6767,"feed_emoji":"🔬","tokens_out":7824,"duration_ms":69683,"temperature":0.7,"pith_summary":"TiSe2 undergoes a charge density wave (CDW) transition near 200 K, and whether the electronic band gap opens or grows at that transition has been debated for decades. This paper uses high-resolution angle-resolved photoemission spectroscopy to track the bulk valence and conduction band edges in the same spectra, across several Brillouin-zone planes, from 300 K down to 160 K. It reports that the fundamental band gap stays constant at 85–100 meV across the transition, while the CDW appears only as folded valence-band weight and spectral redistribution. On this picture, the transition is a lattice-driven reconstruction of a pre-existing narrow-gap semiconductor, not an excitonic gap opening.","feed_headline":"TiSe2 band gap holds steady through CDW transition","feed_subtitle":"ARPES resolves both band edges at once: the 200 K transition folds bands but never opens the gap.","key_machinery":"The central experimental mechanism is the simultaneous energy-distribution-curve (EDC) measurement of the folded valence band and the conduction band at the same k-point (L and M), achieved by choosing photon energies (119 eV and 95 eV) and polarizations that give both features enough spectral weight. Because both band edges sit in one EDC, the gap is read directly, avoiding the need to cross-reference spectra from different momenta — the step the paper identifies as the likely source of earlier contradictory gap values. The gap magnitude is then extracted by fitting the EDC peak positions, and its invariance across temperature is the load-bearing result.","core_discovery":"The central discovery is that TiSe2's fundamental bulk band gap does not change when the crystal enters its 2x2x2 CDW phase. Selecting photon energies and polarizations that expose both the folded valence band and the Ti 3d conduction band at the same momentum point, the authors resolve both band edges within single energy-distribution curves at L and M. Fits place the gap at 85–100 meV at every temperature from 300 K to 160 K, with run-to-run variations comparable to the fitting uncertainty and no coherent order-parameter-like trend. The transition manifests as an emergent folded valence band whose binding energy lines up with the normal-phase conduction band, while the Fermi-surface contours near L stay nearly identical. The authors conclude that no temperature-driven electronic gap opens at TCDW and that the primary CDW order arises from lattice symmetry-breaking that folds, but does not gap, a pre-existing band insulator.","pith_inferences":["As an editorial extension, temperature-dependent optical absorption or tunneling spectroscopy on the same crystals should show no jump in the absorption edge at 200 K, only spectral-weight redistribution, if the constant-gap claim is right.","The same simultaneous same-k EDC approach could be applied to other candidate excitonic insulators to separate genuine gap opening from band folding without cross-referencing different momenta.","If the gap is set by the normal-state band structure, then perturbations such as strain or doping might shift the CDW transition temperature while leaving the gap essentially unchanged — a prediction the paper does not make but its picture implies."],"forward_implications":["This rules out an excitonic-insulator scenario that requires a temperature-driven hybridization gap to open at TCDW.","The primary CDW order in TiSe2 is a lattice symmetry-breaking reconstruction that folds the electronic structure without gapping it, consistent with a pre-existing narrow-gap band insulator.","Earlier reports of gap widening across the transition likely arose from comparing valence-band edges at Γ with conduction minima at M or L, where a strongly renormalized Γ branch can masquerade as a gap change.","The more pronounced electronic reconstruction reported near 20 K, with a sharp V-shaped conduction band and reduced gap, belongs to a distinct low-temperature regime rather than to the primary CDW onset.","Bulk gap values can be reliably read from single EDCs at one k-point when photon energy and polarization are chosen to expose both band edges simultaneously."],"supporting_citations":[{"why":"Earlier ARPES report of notable gap widening across the transition; the present paper contrasts its constant-gap result against this baseline.","marker":"[17]"},{"why":"Earlier ARPES report of apparent gap enhancement in a different temperature regime; another baseline for the discrepancy.","marker":"[18]"},{"why":"High-resolution ARPES showing a sharp V-shaped conduction band and smaller gap near 20 K; used to distinguish the secondary low-temperature reconstruction.","marker":"[19]"},{"why":"Normal-phase ARPES finding of a narrow-gap semiconducting structure consistent with the pre-existing gap picture.","marker":"[21]"},{"why":"Subsequent ARPES reporting a much smaller low-temperature gap deep in the CDW phase.","marker":"[22]"},{"why":"Ab initio analysis of the structural transition and orbital polarization that the authors cite as theoretical support.","marker":"[24]"},{"why":"Recent ARPES revealing additional Γ-point states and orbital-selective renormalization, used to explain why separate-k-point comparisons can mislead.","marker":"[25]"},{"why":"Theory identifying TiSe2 as a symmetry-breaking band insulator; supports the lattice-driven interpretation.","marker":"[26]"}],"fun_headline_variants":["No gap opening in TiSe2 CDW phase","TiSe2 CDW folds bands, leaves gap intact","Constant band gap across TiSe2 CDW transition","TiSe2 transition: lattice folds, gap persists"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on treating the weak, broad spectral feature near the L point as the bulk Ti 3d conduction band and on trusting fits to that feature to locate the true band edge; if it is a surface artifact or the fits are biased by orbital-selective photoemission weights, the constant-gap conclusion collapses.","fun_headline_variants_meta":{"raw":{"variants":["No gap opening in TiSe2 CDW phase","TiSe2 CDW folds bands, leaves gap intact","Constant band gap across TiSe2 CDW transition","TiSe2 transition: lattice folds, gap persists"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000665,"raw_usage":{"total_tokens":3015,"prompt_tokens":905,"completion_tokens":2110,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":521,"completion_tokens_details":{"reasoning_tokens":2046}},"tokens_in":521,"tokens_out":2110,"duration_ms":15212,"temperature":1.0,"reasoning_tokens":2046,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T10:17:27.262554+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A bulk-sensitive measurement that tracks the same conduction and valence band edges across 200 K — for example temperature-dependent optical absorption on the same crystals, or STM tunneling spectroscopy at the L point — would falsify the claim if it showed the gap shifting by more than the fitting uncertainty as the temperature passes through TCDW.","supporting_citations":[{"cited_title":"Chen, Y.-H","cited_arxiv_id":null,"evidence_quote":"Earlier ARPES report of notable gap widening across the transition; the present paper contrasts its constant-gap result against this baseline."},{"cited_title":"Chen, Y.-H","cited_arxiv_id":null,"evidence_quote":"Earlier ARPES report of apparent gap enhancement in a different temperature regime; another baseline for the discrepancy."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"High-resolution ARPES showing a sharp V-shaped conduction band and smaller gap near 20 K; used to distinguish the secondary low-temperature reconstruction."},{"cited_title":"Huber, Y","cited_arxiv_id":null,"evidence_quote":"Normal-phase ARPES finding of a narrow-gap semiconducting structure consistent with the pre-existing gap picture."},{"cited_title":"Yilmaz and E","cited_arxiv_id":null,"evidence_quote":"Subsequent ARPES reporting a much smaller low-temperature gap deep in the CDW phase."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Ab initio analysis of the structural transition and orbital polarization that the authors cite as theoretical support."},{"cited_title":"Yilmaz, Y","cited_arxiv_id":null,"evidence_quote":"Recent ARPES revealing additional Γ-point states and orbital-selective renormalization, used to explain why separate-k-point comparisons can mislead."},{"cited_title":"Pashov, R","cited_arxiv_id":null,"evidence_quote":"Theory identifying TiSe2 as a symmetry-breaking band insulator; supports the lattice-driven interpretation."}],"review_version":1}