{"id":"eb9272ea-4e74-4ad4-898f-53f6125a9d59","arxiv_id":"2501.18389","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Infrared spectroscopy with DFT shows that in RbTi3Bi5 and CsTi3Bi5, bismuth p-states contribute substantially to the low-energy optical response, correlations are strong, and a 150 K phonon plus localization-peak anomaly may mark the onset of bulk electronic nematicity.","lead":"This paper measures how RbTi3Bi5 and CsTi3Bi5, two kagome-lattice metals, absorb infrared light at different temperatures and compares the results with computer band-structure calculations. The data show titanium and bismuth electron states mixing near the Fermi level, strong electron-phonon coupling in the rubidium compound, and a possible bulk electronic ordering transition near 150 K that may be the onset of nematicity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The new Bi2 pz Fermi-surface pockets around the A point are inferred from a uniform rigid Fermi-level shift; if the true renormalization is orbital-selective, this central d-p claim may be an artifact.","rationale":"The reader's weakest-assumption concerned the structural-transition alternative for the 150 K anomaly. That is a reasonable concern, but the paper already hedges the nematicity claim as 'may indicate' and supports it with several null bulk probes. The d-p Fermi-surface pockets are asserted more strongly and are the paper's distinguishing novelty. The uniform Fermi-level shift is a one-parameter fit; attributing new Bi2 pz pockets to it requires applying the same shift to weakly correlated Bi p states as to strongly correlated Ti d states. This is a correctness risk, not merely a disagreement with consensus, and it is not resolved by the existing DFT+U tests. If the orbital-selective-shift test removes the pockets, the central d-p picture is weakened, but the paper would still retain a valid optical study with a conditional 150 K anomaly. The verdict therefore remains CONDITIONAL, matching the reader's verdict, so no change is needed.","tokens_in":16669,"tokens_out":7002,"duration_ms":69521,"concrete_test":"Recompute the band structure and optical conductivity of RbTi3Bi5 with an orbital-resolved self-energy, shifting only the Ti d bands down by 177 meV while leaving the Bi p bands at their PBE positions, and check whether Bi2 pz states still cross the Fermi level around the A point. If the A-point pockets disappear, the uniform rigid shift is the sole source of the central d-p Fermi-surface claim; if they persist, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is the inference of new Bi2 pz Fermi-surface pockets around the A point from a uniform rigid Fermi-level shift of -177 meV (RbTi3Bi5) and -179 meV (CsTi3Bi5), introduced around Fig. 1(b) and Fig. 3(a) of the main text. The low-energy interband fit is used to justify this shift, but the same shift is then applied to all bands, including the weakly correlated Bi pz states. The paper's own correlation argument is orbital-selective: the Ti d flat bands are strongly renormalized, while the Bi pz-derived tilted Dirac bands are the less-correlated partners. A single global shift therefore conflates a d-band self-energy correction with a uniform chemical-potential shift; if the true renormalization is orbital-dependent, the A-point pockets formed by Bi2 pz may be an artifact of the rigid shift rather than a physical Fermi-surface feature. The DFT+U checks in Fig. S10 do not address this, because they shift all Ti d states and fail to reproduce ARPES. No ARPES or quantum-oscillation measurement is cited that directly identifies these A-point pockets, so the central distinguishing claim of the paper—Bi2 pz pockets that separate ATi3Bi5 from AV3Sb5—rests on an untested modeling assumption.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports broadband (50-18000 cm^-1) infrared reflectivity measurements on single crystals of RbTi3Bi5 and CsTi3Bi5, from which the optical conductivity is obtained by Kramers-Kronig analysis. The spectra are decomposed into Drude, localization, interband Lorentzian, and (for RbTi3Bi5) Fano phonon contributions. By comparing with DFT band-structure and optical-conductivity calculations, the authors argue that the low-energy interband response is dominated by transitions between linearly dispersing Ti d bands (bands C and D) and by transitions between tilted Dirac bands involving Bi pz orbitals (bands E and F). To reproduce the experimental interband spectra, they apply a rigid downward shift of the Fermi level by 177 meV (RbTi3Bi5) and 179 meV (CsTi3Bi5); this shift also creates new Bi2 pz Fermi-surface pockets around the A point. Finally, they identify a Fano phonon mode near 200 cm^-1 and a localization peak in RbTi3Bi5, both showing anomalies around 150 K, which they suggest may indicate the onset of bulk nematicity. A correlation-strength estimate from the ratio of experimental to DFT intraband spectral weight gives approximately 0.40 for both compounds, in contrast to CsV3Sb5.","tokens_in":17008,"tokens_out":5522,"duration_ms":56462,"significance":"If the central claims hold, the paper delivers new experimental facts about the ATi3Bi5 family: the low-energy optical response is not solely a Ti d-band story, but involves Bi2 pz states that form tilted Dirac crossings and new Fermi-surface pockets around the A point; and a bulk 150 K anomaly in RbTi3Bi5 may be the nematic transition previously seen only by surface-sensitive probes. The experimental work is careful, including explicit handling of sample air sensitivity, and the band-resolved optical-conductivity calculations provide a concrete microscopic assignment of the low-energy interband features. The quantitative comparison of correlation strength across the 135 family is also a useful contribution. However, the two headline claims rest on modeling assumptions that are not fully tested: the Bi2 pz pockets follow from a rigid Fermi-level shift, and the 150 K nematic interpretation is built on the absence of anomalies in other bulk probes. Both are falsifiable and should be explicitly presented as model-dependent unless additional evidence is supplied.","major_comments":[{"comment":"The new Bi2 pz Fermi-surface pockets around the A point are inferred from a single global Fermi-level shift of -177 meV (Rb) and -179 meV (Cs) applied to all bands. This is load-bearing because the paper itself argues for orbital-selective correlations: the Ti d flat bands are strongly renormalized while the Bi pz-derived tilted Dirac bands are described as the less-correlated partners. A uniform shift therefore conflates a d-band self-energy correction with a chemical-potential shift, and if the true renormalization is orbital-dependent, the A-point pockets may be an artifact of the rigid-shift ansatz. The DFT+U checks in Fig. S10 do not resolve this issue because they shift all Ti d states and are explicitly inconsistent with ARPES. No ARPES or quantum-oscillation data are cited that directly identify these A-point pockets. I recommend either softening the claim to a model-dependent prediction or providing direct experimental confirmation (e.g., ARPES at the A point or quantum oscillations for the new pockets).","section":"Fig. 3(a) and the second paragraph under 'Band structure and optical conductivity calculations'"},{"comment":"The localization peak in RbTi3Bi5 is stated in the Supplemental Material to be so sharp that it cannot be fitted with the displaced-Drude model of Eq. S3, and a simple Lorentzian is used instead. The main text nevertheless uses the temperature dependence of this Lorentzian peak position to claim an electron-phonon coupling anomaly at 150 K. This is internally inconsistent: the physical interpretation of the peak as a boson-induced localization peak, and the significance of the slope change in Fig. 4(c), are both tied to a model that is explicitly not applied to RbTi3Bi5. The paper should either show that the simple Lorentzian reproduces the same peak positions as the displaced-Drude model in a regime where the latter applies, or demonstrate that the peak-position anomaly is robust to the choice of background subtraction and fit function.","section":"Supplemental Material, 'LOCALIZATION PEAK' section and main text Fig. 4(c)"},{"comment":"The interpretation that the 150 K anomaly in RbTi3Bi5 signals the onset of nematicity relies on ruling out a structural transition via the absence of anomalies in interband optical absorption, dc transport, magnetic susceptibility, and specific heat. These are negative observations, and the cited references (Refs. [13,16]) are not displayed in the manuscript, so the sensitivity of those probes in the relevant temperature range cannot be judged. The interband-optical statement is also based on the same decomposed fits used to define the phonon and localization anomalies, so it is not an independent check. A subtle lattice distortion that does not appreciably affect these probes could produce the same Fano and localization-peak changes. The claim should be reframed as 'consistent with nematicity' or supported by a direct bulk structural probe (e.g., thermal expansion or diffuse scattering).","section":"Main text, paragraph beginning 'Below 150 K' and the 'absence of a low-temperature anomaly' argument"}],"minor_comments":[{"comment":"The sentence 'The primary effect of the Fermi level shift is the emergence of Bi2 pz states at the Fermi level around the A point, as illustrated in Fig. 1(a)' appears to refer to the wrong panel; the band structure is in Fig. 1(b) and the Fermi surface in Fig. 3(a).","section":"Main text, second paragraph of the section discussing Fig. 1(b)"},{"comment":"The experimental plasma frequencies carry estimated error bars of 10%, but no corresponding uncertainty is given for the DFT plasma frequencies (4.02 eV and 3.98 eV); a brief statement about the numerical convergence of the DFT values would make the correlation ratio more robust.","section":"Supplemental Material, 'PLASMA FREQUENCY' section"},{"comment":"The phrase 'a similar downward shift of the Fermi level has also been reported in ARPES measurements for the Cs compound' could be clarified: ARPES measures occupied band positions relative to the Fermi level, not a change in the chemical potential, so the comparison refers to the inferred renormalization of flat-band energies rather than a direct measurement of a Fermi-level shift.","section":"Main text, section 'Band structure and optical conductivity calculations'"}],"recommendation":"major_revision","confidential_remarks":"The optical data and band-resolved calculations are valuable and likely of interest to the kagome-metals community. My main concern is that the paper's two most distinctive claims—the Bi2 pz A-point pockets and the 150 K bulk nematic onset—are presented as established findings although they depend on a rigid Fermi-level shift and on negative evidence against a structural transition. These are fixable by framing the claims more cautiously or by adding direct confirmation, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The new data are real and useful. This is the first broadband infrared study of RbTi3Bi5 and the first to resolve the low-energy localization peak and correlation strength in CsTi3Bi5; the independent prior IR study of CsTi3Bi5 (ref 51) did not see these features. The band-resolved DFT assignment of the low-energy interband response to Ti d and Bi pz states is careful and mostly convincing. I also credit the transparent handling of the air-sensitive samples and the explicit DFT+U checks that fail, which strengthens the case that a simple rigid shift of the Fermi level is the least-bad way to bring theory and experiment together.\n\nThe soft spots are real but not fatal. The Rb localization peak cannot be fit with the displaced-Drude model, so they use a Lorentzian; they say this in the supplemental, which is honest, but it does mean the peak position and width have a somewhat model-dependent meaning. Bigger concern: the new Bi2 pz Fermi-surface pockets around the A point come only from applying a uniform -177/-179 meV shift to all bands. The paper's own correlation narrative is orbital-selective: Ti d is strongly renormalized, Bi pz is less correlated. A single global shift may therefore be a reasonable effective description of the d-band self-energy but not necessarily a faithful guide to where the pz bands sit. No ARPES or quantum oscillation data directly confirm these A-point pockets, so that central distinguishing claim rests on an untested assumption. It is a good hypothesis, but it should be framed as one.\n\nThe 150 K anomaly in RbTi3Bi5 is handled with the right hedges. The absence of anomalies in transport, susceptibility, specific heat, and interband optics does make a structural transition unlikely, though not impossible. What bothers me more is that the Fano q2 and phonon frequency are plotted without error bars, and the localization peak slope change is visually clear but not quantified. This is a minor issue for a claims-are-conditional paper, but it means the nematicity signature is more suggestive than demonstrative.\n\nOverall: the optical data itself is solid, the DFT comparison is thorough, and the interpretation is clearly labeled as provisional. The paper deserves a serious referee, not because it is revolutionary, but because it is a careful experimental study that adds new facts and one speculatively interesting correlation. I would send it to review and ask the referee to focus on the Fermi-level shift and the error bars on the temperature-dependent parameters. The field (kagome metals, optical spectroscopy) will want this on the record.","headline":"Solid broadband IR study with a credible d-p coupling story, but the new Bi pz pockets and the 150 K nematicity are both less firm than the abstract suggests.","tokens_in":17525,"tokens_out":1825,"would_cite":true,"duration_ms":20592,"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":"Bismuth p-states reshape the low-energy electronic structure of the titanium kagome metals RbTi3Bi5 and CsTi3Bi5, and a 150 K phonon anomaly marks the likely onset of electronic nematicity.","keywords":["kagome metals","flat bands","optical conductivity","nematicity","electron-phonon coupling","Dirac crossings","RbTi3Bi5","CsTi3Bi5"],"falsifier":"A diffraction or pair-distribution-function scan across 150 K looking for a lattice distortion would settle whether the phonon and localization-peak anomalies are electronic, and a search for the predicted Bi2 $p_z$ pockets around the $A$ point by angle-resolved photoemission or quantum oscillations would settle whether the Fermi-level shift picture is correct.","tokens_in":16488,"feed_emoji":"⚛️","tokens_out":11030,"duration_ms":96555,"temperature":0.7,"pith_summary":"The paper uses broadband infrared spectroscopy and density-functional calculations to establish what carries the low-energy electronic response of two titanium-based kagome metals, RbTi3Bi5 and CsTi3Bi5, whose Fermi level sits unusually close to flat bands. It argues that the low-energy optical conductivity is not set by titanium d-states alone: non-kagome bismuth pz states form a tilted Dirac crossing, create new Fermi-surface pockets near the A point, and contribute interband absorption that the d-bands alone cannot explain. It further identifies a change in electron-phonon coupling inside RbTi3Bi5 near 150 K, namely a Fano-shaped phonon that suddenly turns into an anti-resonance and a localization peak whose position changes slope, and reads this as the bulk onset of the electronic nematic order previously seen only at surfaces. If correct, the result makes nematic order in this family a bulk, purely electronic phenomenon and shows that phonons can act as fingerprints of that transition.","feed_headline":"A 150 K phonon anomaly marks nematic onset in RbTi3Bi5","feed_subtitle":"Broadband infrared spectra reveal bismuth p-states reshaping the Fermi surface of titanium kagome metals","key_machinery":"The load-bearing object is the band-resolved optical conductivity of the kagome 135 lattice, decomposed into a Drude term for itinerant carriers, a localization peak for carriers transiently trapped by bosonic excitations, Lorentzian interband transitions, and a Fano phonon line. The argument turns on a Fermi-level shift: treating the DFT band energies as needing a uniform downward renormalization of 177 meV to 179 meV reproduces the measured interband spectra and moves the Bi2 $p_z$ states onto the Fermi level around the $A$ point without altering the $k_z = 0$ Fermi surface. The Fano resonance, an asymmetric phonon line shape produced by interference between a discrete phonon and a continuum of electronic excitations, supplies the second mechanism: its sharp evolution into an anti-resonance below 150 K, together with a kink in the localization-peak position, is the paper's evidence for a change in electron-phonon coupling as the system enters the nematic state.","core_discovery":"The central discovery is that the low-energy physics of the titanium 135 kagome metals is governed by an interplay between correlated Ti $d$-states and Bi $p$-states, not by the kagome $d$-bands alone. Reproducing the measured optical conductivity requires shifting the DFT Fermi level downward by 177 meV in RbTi3Bi5 and 179 meV in CsTi3Bi5; this shift leaves the $\\Gamma$–M–K–$\\Gamma$ bands essentially untouched but brings Bi2 $p_z$ states to the Fermi level around the $A$ point, producing two new Fermi-surface sheets and making the observed low-energy absorption arise from transitions between linearly dispersing bands, including a tilted Dirac crossing dominated by Bi1 and Bi2 $p_z$ states. The same data show strong electronic correlations: the experimental intraband spectral weight is only about 40 percent of the DFT value, and the DFT+U prescription fails to place the flat bands correctly, indicating orbital-selective correlations beyond a mean-field treatment. In RbTi3Bi5, a Fano-shaped infrared-active $E_{1u}$ phonon near 200 cm$^{-1}$ and the localization peak both change sharply below 150 K, which the paper interprets as a sudden change in electron-phonon coupling accompanying the onset of bulk nematicity.","pith_inferences":["Editorial inference: If the 150 K transition is genuinely nematic, the Fano phonon's sudden change implies that the electron-phonon coupling strength is tied to the nematic order parameter; a temperature-dependent Raman or inelastic x-ray measurement could look for a corresponding phonon anomaly at the same temperature.","Editorial inference: The nearly identical slopes of the CsTi3Bi5 localization peak and of the low-temperature RbTi3Bi5 slope suggest that CsTi3Bi5 may enter a similar state above room temperature; extending optical measurements above 300 K would test this directly.","Editorial inference: The paper's Fermi-level shift is equivalent to a nominal hole doping of roughly 0.8 to 0.9 electrons per formula unit, yet the authors attribute it to band renormalization; angle-resolved photoemission mapping of the Bi2 $p_z$ pockets around the $A$ point would discriminate between a true doping effect and a correlation-driven renormalization.","Editorial inference: If phonons are fingerprints of nematicity, then uniaxial-stress experiments that tune the nematic transition should also tune the Fano parameter and localization-peak slope, providing a mechanical control knob for the purported transition."],"forward_implications":["The 150 K anomaly in RbTi3Bi5 would establish a bulk electronic transition where only surface-sensitive probes had previously suggested nematicity, making the nematic order a property of the crystal rather than of its surface.","The low-energy optical response of the titanium 135 family is dominated by Bi $p_z$ states and tilted Dirac crossings, so theories of these compounds must include non-kagome orbitals rather than treating only the Ti $d$-band manifold.","The failure of DFT+U to reproduce the band structure and optical spectra means the correlations in these materials are orbital-selective and require a treatment beyond static mean-field approaches such as DFT+U.","Phonon and localization-peak anomalies can serve as bulk fingerprints for electronic instabilities in kagome metals, motivating momentum-dependent phonon measurements.","The absence of the 150 K anomaly in CsTi3Bi5, despite a similar low-temperature localization-peak slope, leaves open that nematicity there either is surface-confined or sets in above room temperature."],"supporting_citations":[{"why":"Supplies the crystal structures used in the DFT calculations and the dc transport, magnetic susceptibility, and specific-heat data that argue against a structural transition at 150 K.","marker":"[13]"},{"why":"First-principles calculations cited for the absence of a charge-density-wave state in ATi3Bi5, anchoring the claim that any low-temperature symmetry change is electronic.","marker":"[17]"},{"why":"STM evidence for superconductivity and nematic order in CsTi3Bi5 without CDW, the surface result the bulk optical probe is meant to corroborate.","marker":"[18]"},{"why":"ARPES and STM study reporting orbital-selective electronic nematicity and nontrivial band topology in a titanium-based kagome superconductor.","marker":"[19]"},{"why":"Reports electronic nematicity without charge density waves in titanium-based kagome metals, giving the prior surface-sensitive evidence for the symmetry lowering.","marker":"[24]"},{"why":"ARPES study of RbTi3Bi5 documenting flat bands and rotational symmetry breaking and used to anchor the empirical Fermi-level position.","marker":"[25]"},{"why":"Identifies a Pomeranchuk instability from electronic correlations in CsTi3Bi5, providing the theoretical mechanism behind the proposed nematic onset.","marker":"[26]"},{"why":"Optical study of RbV3Sb5 that supplies the comparative phonon, localization-peak, and spectral-weight analysis used for the ATi3Bi5 decomposition.","marker":"[31]"},{"why":"Optical study of CsV3Sb5 used as the nearly uncorrelated baseline for the intraband spectral-weight ratio.","marker":"[32]"},{"why":"ARPES study of CsTi3Bi5 in which a similar downward Fermi-level shift was reported, supporting the 177 meV to 179 meV adjustment.","marker":"[36]"}],"fun_headline_variants":["Bi p-states reshape Fermi surface in RbTi3Bi5 and CsTi3Bi5","Ti d-states and Bi p-states couple to drive kagome metal physics","Fermi level shift exposes Bi p-states in titanium kagome metals","Phonon anomaly at 150 K heralds nematicity in RbTi3Bi5"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The interpretation of the 150 K anomaly as the onset of nematicity rests on the assumption that no structural transition occurs, since a subtle lattice distortion that leaves interband optics, dc transport, magnetic susceptibility, and specific heat unchanged could also explain the phonon and localization-peak anomalies.","fun_headline_variants_meta":{"raw":{"variants":["Bi p-states reshape Fermi surface in RbTi3Bi5 and CsTi3Bi5","Ti d-states and Bi p-states couple to drive kagome metal physics","Fermi level shift exposes Bi p-states in titanium kagome metals","Phonon anomaly at 150 K heralds nematicity in RbTi3Bi5"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001155,"raw_usage":{"total_tokens":4820,"prompt_tokens":1016,"completion_tokens":3804,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":632,"completion_tokens_details":{"reasoning_tokens":3714}},"tokens_in":632,"tokens_out":3804,"duration_ms":24769,"temperature":1.0,"reasoning_tokens":3714,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T23:40:14.205700+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A diffraction or pair-distribution-function scan across 150 K looking for a lattice distortion would settle whether the phonon and localization-peak anomalies are electronic, and a search for the predicted Bi2 $p_z$ pockets around the $A$ point by angle-resolved photoemission or quantum oscillations would settle whether the Fermi-level shift picture is correct.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"First-principles calculations cited for the absence of a charge-density-wave state in ATi3Bi5, anchoring the claim that any low-temperature symmetry change is electronic."},{"cited_title":"Liu, M.-Q","cited_arxiv_id":null,"evidence_quote":"STM evidence for superconductivity and nematic order in CsTi3Bi5 without CDW, the surface result the bulk optical probe is meant to corroborate."},{"cited_title":"Pressure-induced double-dome superconductivity in kagome metal CsTi3Bi5","cited_arxiv_id":"2308.10129","evidence_quote":"Reports electronic nematicity without charge density waves in titanium-based kagome metals, giving the prior surface-sensitive evidence for the symmetry lowering."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ARPES study of RbTi3Bi5 documenting flat bands and rotational symmetry breaking and used to anchor the empirical Fermi-level position."},{"cited_title":"Jiang, Z","cited_arxiv_id":null,"evidence_quote":"Identifies a Pomeranchuk instability from electronic correlations in CsTi3Bi5, providing the theoretical mechanism behind the proposed nematic onset."},{"cited_title":"Wenzel, B","cited_arxiv_id":null,"evidence_quote":"Optical study of CsV3Sb5 used as the nearly uncorrelated baseline for the intraband spectral-weight ratio."},{"cited_title":"Fratini and S","cited_arxiv_id":null,"evidence_quote":"ARPES study of CsTi3Bi5 in which a similar downward Fermi-level shift was reported, supporting the 177 meV to 179 meV adjustment."}],"review_version":1}