{"id":"4e17aca5-3ed0-4fd2-9909-466b1b446045","arxiv_id":"2411.15688","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The ~8 eV optical absorption peak in water and ice arises from charge-transfer excitons whose collective intensity increases with the ordering length of proton-ordered water wires, making optical absorption a potential probe of water wires.","lead":"A computational study using GW-BSE theory assigns the main ~8 eV optical absorption peak of water and ice to a charge-transfer exciton between hydrogen-bonded molecules, and shows its intensity grows with the length of ordered water wires. This suggests a standard optical measurement could reveal the ordering of hydrogen-bond chains that are important in biology and ice physics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The intensity-vs-ordering-length claim depends on an unproven proxy: electron-hole overlap is not shown to track BSE oscillator strength with density and H-bond strength held fixed, so the central scaling remains unsupported.","rationale":"The strongest parts of the paper are real: the GW-BSE spectra reproduce experiment after small shifts, and the identification of the ~8 eV peak as a charge-transfer exciton coupled to H-bonds is plausible and supported by the restricted-subspace decomposition and real-space electron/hole densities. The concern is specifically the quantitative scaling claim, which is the paper's main novelty. The authors never compute the BSE oscillator strength as a function of wire ordering length at fixed density and H-bond geometry; instead they use the electron-hole overlap Ω. The Supplemental's Ω(P) fit is based on three phase-averaged points and an explicit assumption, not a derivation. Thus the central claim is underdetermined. The ice Ih vs ice XI comparison is the cleanest part of the evidence, because density and H-bond strength are nearly unchanged while proton order differs, but it is a single pair of points and the two structures differ in temperature treatment (80 K classical vs 0 K relaxed). A controlled l-series calculation would settle whether the scaling is real. Since the paper's qualitative proposal may survive such a test, CONDITIONAL remains the appropriate verdict and the reader's weakest assumption is the correct one. I therefore recommend no change to the verdict.","tokens_in":17553,"tokens_out":6205,"duration_ms":62292,"concrete_test":"Construct a series of ice-Ih supercells (same 64-molecule cell, same volume and SCAN0-relaxed oxygen network) with proton configurations engineered to contain ordered water-wire segments of length l = 1, 2, 3, 4, and a fully ordered (ice-XI-like) chain, keeping all other structural parameters fixed. Run the same GW-BSE calculation and broadening for each configuration, integrate Im ε(ω) over the first absorption peak, and compare the resulting integrated oscillator strengths with the electron-hole overlap Ω. If the integrated oscillator strength does not increase monotonically with l, or if Ω and oscillator strength disagree in ordering, the central scaling claim is refuted; if they agree, the proxy is validated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central, falsifiable prediction is that the ~8 eV absorption peak intensity scales with the ordering length of proton-ordered water wires. For this claim, the paper must show that (i) the electron-hole overlap Ω computed from BSE eigenstates is a quantitative proxy for the integrated oscillator strength of that peak, and (ii) the increase in Ω from liquid water to ice Ih to ice XI is caused by wire ordering length rather than by the simultaneous changes in density, H-bond strength, temperature, and nuclear quantum effects. Neither condition is demonstrated. Fig. 4(d) plots Ω versus wire length, but no BSE oscillator strength versus l is reported. Supplemental Section VIII relates Ω to the average molecular dipole P and fits Ω(P) = 0.005488P^2 − 0.0327788P + 0.0491024 to only three phase-averaged points (liquid 2.95 D, ice Ih 3.09 D, ice XI 3.36 D); the fit has no statistical uncertainty and contains a linear term inconsistent with the claimed Ω ∝ P^2 model. The section explicitly says the relation is 'solely based on the assumption that the local polar field increases when the length of the water wire is longer.' Even if Ω increases with l, Ω is not the optical transition dipole: BSE oscillator strengths are governed by exciton coefficients and interband velocity matrix elements, so integrated Im ε weight need not follow Ω. The claim therefore currently rests on a proxy correlation rather than on the measured observable it predicts.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports GW-BSE calculations of the optical absorption spectra of liquid water, ice Ih, and ice XI, and interprets the main absorption peak near 8 eV as a charge-transfer exciton between hydrogen-bonded water molecules. The authors further propose that in ice this peak is enhanced by a collective excitation occurring on proton-ordered water wires, and that the spectral intensity scales with the ordering length of the water wire. On this basis they argue that optical absorption spectroscopy can serve as a direct probe of water wires and their ordering in hydrogen-bond networks.","tokens_in":17889,"tokens_out":4620,"duration_ms":41859,"significance":"If substantiated, the central claim would be a significant advance: it would turn a standard optical absorption measurement into a structural probe of hydrogen-bond network ordering, with potential applications to confined water, biological systems, and other ice phases. The computational approach is state-of-the-art: the calculations use hybrid-DFT starting points, the GW-BSE method with electron-hole interactions, and the results are compared against experimental spectra for the imaginary and real parts of the dielectric function and for the absorption coefficient. The spectral decomposition into band-to-band contributions and the real-space exciton analysis are carefully presented. However, the load-bearing scaling claim is currently supported by a proxy quantity, the electron-hole overlap, rather than by the directly measured observable, the integrated oscillator strength, and the phase comparison used to establish the trend conflates multiple structural and thermodynamic variables.","major_comments":[{"comment":"The central prediction that the ~8 eV absorption intensity scales with the water-wire ordering length is supported only by the electron-hole overlap Ω computed from BSE eigenstates, not by the BSE oscillator strength or by the integrated Im ε of the peak. No calculation of the oscillator strength as a function of wire length l is reported, and no demonstration that Ω is proportional to the BSE dipole strength is given. Since BSE oscillator strengths are governed by exciton coefficients and interband velocity matrix elements, Ω is not by construction the optical transition dipole. The authors should directly compute and report the integrated Im ε or BSE oscillator strength for excitons on wires of different l, or otherwise establish quantitatively that Ω tracks the oscillator strength for the same exciton states.","section":"Section 'Collective excitation of charge transfer excitons on water wires', Fig. 4(d), and Supplemental Section VIII"},{"comment":"The comparison across liquid water, ice Ih, and ice XI changes not only the wire ordering length but also the density, H-bond strength, temperature, and nuclear quantum effects (the latter are included for liquid water via PI-DPMD but treated classically for ice, as stated in Supplemental Section I). The observed monotonic increase in Ω from liquid to ice Ih to ice XI therefore cannot be uniquely attributed to the ordering length. To support the scaling claim, the authors need to isolate the effect of l from the other variables, for example by comparing ice Ih configurations with different wire-length distributions at fixed density and H-bond strength, or by varying the proton order while holding the oxygen lattice fixed.","section":"Fig. 4(c,d) and the liquid/ice Ih/ice XI comparison"},{"comment":"The statistical basis for the quantitative intensity comparison is thin: the liquid-water spectrum is based on two snapshots, the ice Ih spectrum on eight snapshots, and ice XI on a single optimized structure. No error bars or convergence tests with respect to the number of snapshots are provided. Because the central claim concerns relative peak intensities, which are sensitive to configurational sampling, the authors should either provide more snapshots or show that the reported intensity differences are robust to the number of configurations used.","section":"Supplemental Section I and Fig. 4(c)"},{"comment":"The fitted relation Ω(P) = 0.005488P² − 0.0327788P + 0.0491024 is based on only three phase-averaged points, has no stated uncertainty, and includes a linear term that is inconsistent with the stated model Ω = kP². The section also explicitly states that the relation is 'solely based on the assumption that the local polar field increases when the length of the water wire is longer.' This is a heuristic assumption, not a derivation from the BSE data, and it should not be used as evidence for the scaling law. The authors should either provide a first-principles derivation of Ω(P) or clearly label the relation as a conjecture and remove it from the argument supporting the main claim.","section":"Supplemental Section VIII, Fig. S7"}],"minor_comments":[{"comment":"The sentence 'The theoretical optical spectra of liquid water was generated' should read 'were generated' for grammatical agreement.","section":"Fig. 1 caption"},{"comment":"The word 'energerically' is a typo and should be 'energetically'.","section":"Results and Discussion, paragraph on ice XI"},{"comment":"The Supplemental Material contains multiple formatting artifacts, for example the '/ba√︂ex' strings in Section III, which interfere with readability and should be cleaned before publication.","section":"Supplemental Material throughout"},{"comment":"The definition of 'water wire' and 'ordering length l' is given through examples rather than a formal definition; a precise definition of what constitutes a wire and how l is counted would improve reproducibility.","section":"Main text, section on water wires"},{"comment":"The three-dimensional plot of the electron-hole overlap as a function of y and l is difficult to read; a two-dimensional plot of the integrated overlap versus l with error bars would more directly support the claimed trend.","section":"Fig. 4(d)"},{"comment":"The statement that in liquid water the wire length is 'l=1 at any temperature' appears to conflict with the earlier use of l=2 for H-bonded pairs; the terminology should be clarified to avoid confusion.","section":"Supplemental Section VI"}],"recommendation":"major_revision","confidential_remarks":"This is a thought-provoking manuscript from an experienced group, and the GW-BSE calculations are of high quality. The main concern is whether the scaling claim can be sustained without a direct calculation of oscillator strengths as a function of wire length. If the authors can supply that analysis and address the confounds in the phase comparison, the paper would be a strong candidate for publication. I have no concerns about the authors' integrity or the novelty disclosure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Plainly: this is a serious GW-BSE study, and the central assignment—that the ~8 eV band is a charge-transfer exciton—is not new, but the paper's specific contribution is the link between that exciton and H-bond geometry. The charge-transfer character was already established by Hahn, Garbuio, Hermann and others (Refs 28-35). What is genuinely new is the correlation of the exciton's electron-hole overlap with the proton-transfer coordinate in liquid water (Fig. 3c), and the collective-exciton mechanism proposed for proton-ordered water wires, where a chain of CT excitons is stabilized by the local polarization field. The spectra are state of the art: GW-BSE on PBE0 starting points, PI-DPMD structures for water, multiple ice configurations, and they match experiment after modest rigid blue shifts (~0.6-0.8 eV) and 0.4 eV broadening. That is credible and useful work.\n\nThe soft spot is the quantitative scaling claim, and it is the load-bearing one. The paper argues that the intensity of the main peak scales with the ordering length l of water wires. Fig. 4(d) plots the electron-hole overlap Ω against l, but no BSE oscillator strength or integrated Im ε is shown as a function of l. Ω is not the optical transition dipole; oscillator strength is governed by exciton coefficients and interband velocity matrix elements, so the link from Ω to measured absorption is unverified. The Supplemental Section VIII tries to connect Ω to the average molecular dipole P with a fit to three phase-averaged points (liquid 2.95 D, ice Ih 3.09 D, ice XI 3.36 D). That fit has no error bars, and the fitted function Ω(P) = 0.005488P² − 0.0327788P + 0.0491024 contains a linear term that is inconsistent with the claimed Ω ∝ P² model. The section itself says the relation is 'solely based on the assumption that the local polar field increases when the length of the water wire is longer.' So the central falsifiable prediction currently rests on a proxy correlation. Add to that: only two snapshots for liquid water, no statistical averaging over ice proton-disorder configurations, and empirical blue shifts that are partly compensating for GW gap underconvergence. None of this kills the paper—the direction of the effect is plausible and the spectra are believable—but it means the headline claim is not yet demonstrated.\n\nWho is this for? People working on the electronic structure of water, excitons in disordered media, and optical probes of H-bond networks. It deserves a serious referee: the calculations are heavy and the proposal is testable. A referee should ask for direct oscillator-strength versus wire-length data, more snapshots, and error bars on the scaling. My own verdict would be conditional, not accept.","headline":"Strong GW-BSE spectra with a plausible but under-supported wire-length scaling claim; worth refereeing, not yet conclusive.","tokens_in":18408,"tokens_out":3319,"would_cite":true,"duration_ms":28910,"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":"The main optical absorption peak of water and ice is a charge-transfer exciton whose intensity in ice is set by the length of proton-ordered water wires.","keywords":["optical absorption spectroscopy","water wires","charge transfer exciton","hydrogen-bond network","proton ordering","ice XI","electron-hole interaction","GW-BSE"],"falsifier":"Measure the 8 eV absorption peak in a series of ices whose proton-ordering correlation length is independently varied — for example KOH-doped ice annealed to different degrees — and check whether the peak intensity rises monotonically with the independently measured wire-ordering length; the paper's mechanism predicts monotonic growth, while a flat or saturated curve would falsify the length-scaling claim.","tokens_in":1567,"feed_emoji":"💧","tokens_out":1862,"duration_ms":70900,"temperature":0.7,"pith_summary":"The paper argues that the main optical absorption peak of water and ice — the one near 8 eV — is a charge-transfer exciton stretched across hydrogen bonds, and that in ice its strength is boosted by collective excitations on proton-ordered water wires. It claims that the intensity of that peak grows with the ordering length of such wires, reaching a maximum in ice XI, the proton-ordered phase of ice. If correct, optical absorption spectroscopy becomes a direct probe of water-wire presence and length, which matters because water wires are thought to carry protons, energy, and information in biological and confined environments but have never been directly detected. The argument is carried through first-principles many-body calculations on molecular dynamics snapshots, benchmarked against measured spectra.","feed_headline":"8 eV water peak reveals proton-ordered wires","feed_subtitle":"The peak is a charge-transfer exciton whose intensity grows with wire length, maxing out in ice XI.","key_machinery":"The central object is the charge-transfer exciton built from a $1b_1$ lone-pair hole on one water molecule and a $4a_1$ antibonding electron on its hydrogen-bonded neighbor, quantified through the electron-hole overlap density $\\rho_h \\times \\rho_e$ computed from the Bethe-Salpeter eigenstates. The mechanism that carries the argument is the proton-ordered water wire: a chain of hydrogen bonds along which all molecular dipoles point the same way, producing a local polarization field that stabilizes a collective relay of charge-transfer excitons and increases the transition strength. The paper reports that the electron-hole overlap — and with it the oscillator strength and binding energy — increases monotonically with ordering length $l$ and peaks at $l=\\infty$ in ice XI, which is what links a standard optical spectrum to a hidden structural motif.","core_discovery":"In both liquid water and ice, the \\sim 8 eV absorption peak is a bound exciton of charge-transfer character: an electron is excited from the lone-pair $1b_1$ orbital of a hydrogen-bond acceptor molecule onto the antibonding $4a_1$ orbital centered on the protons of the donor molecule, so the electron-hole pair is stabilized along the hydrogen-bond direction. In ice, where hydrogen bonds are intact and connected into chains, the paper finds that a proton-ordered water wire creates a local polarization field from aligned molecular dipoles; this field stabilizes a collective relay of charge-transfer excitons along the wire, suppresses recombination, and enhances oscillator strength, electron-hole overlap, and exciton binding energy as the ordering length $l$ grows. The binding energy rises from about 2.2 eV in liquid water to about 3.4 eV in ice Ih and about 3.6 eV in ice XI, and the electron-hole overlap density at the first peak increases monotonically from the short-wire limit in water to the infinite-wire limit in ice XI. The paper therefore proposes the spectral intensity of the main peak as a measure of water-wire ordering length.","pith_inferences":["A direct extension of the paper's mechanism is that time-resolved optical absorption could watch proton ordering propagate during the ice Ih-to-XI transition, with the 8 eV peak height tracking the growing correlation length.","Confined water in protein channels or nanotubes, where transient wires are proposed, would be a natural testbed: the model predicts peak enhancement whenever a wire of length $l\\ge 3$ forms.","The supplement's scaling relation between electron-hole overlap and molecular dipole moment suggests a quantitative route to infer local polarization from spectral intensity, though the paper only establishes the relation in the systems studied.","A still-unstated consequence of the assignment is that standard optical calculations without electron-hole interactions would miss the wire-length signal entirely, so experimental comparisons must be made against excitonic theory."],"forward_implications":["The intensity of the \\sim 8 eV absorption peak in ice becomes a readout of the presence and ordering length of proton-ordered water wires.","The same assignment separates phases of water by wire length: short wires ($l\\simeq 2$ hydrogen-bonded pairs) in liquid water, disorder-limited finite wires in ice Ih, and infinite wires in ice XI.","Because the exciton is coupled to hydrogen-bond strength, the spectral shape of the main peak can also report the distribution of hydrogen-bond strengths in disordered water.","The predicted near-independence of the liquid-water main peak with temperature follows from wires being dominated by single hydrogen-bonded pairs, so changes in the peak height signal ordering rather than thermal population.","The approach extends naturally to any ice phase or confined aqueous environment where proton-ordered chains are proposed, offering a spectroscopic route to detect them without scattering-based structural probes."],"supporting_citations":[{"why":"Supplies the GW quasiparticle-energy framework on which the excitonic calculation is built.","marker":"[39]"},{"why":"Supplies the Bethe-Salpeter equation treatment of electron-hole interactions used to obtain the spectra.","marker":"[40]"},{"why":"Provides the ab initio many-body implementation that makes the GW-BSE spectra computationally feasible.","marker":"[41]"},{"why":"Provides the machine-learned molecular dynamics model used to generate liquid-water configurations.","marker":"[42]"},{"why":"Prior identification of the main absorption peak as a charge-transfer exciton, which the paper extends to water wires.","marker":"[29]"},{"why":"Experimental optical absorption spectrum of liquid water against which the theory is benchmarked.","marker":"[57]"},{"why":"Experimental optical absorption spectrum of ice used for comparison.","marker":"[58]"},{"why":"Establishes ice XI as the proton-ordered ferroelectric phase whose infinite wire defines the long-range-ordering limit.","marker":"[63]"}],"fun_headline_variants":["8 eV exciton peak exposes water wires","Water wire ordering measured by 8 eV absorption intensity","Ice XI's 8 eV peak signals proton-ordered water wires","Charge-transfer exciton tracks water-wire length in ice","8 eV absorption reveals water-wire ordering"],"cache_read_input_tokens":20480,"weakest_assumption_plain":"The load-bearing premise is that the calculated electron-hole overlap, averaged over a handful of structural snapshots, tracks the experimental oscillator strength well enough that differences in the 8 eV peak across water, ice Ih, and ice XI can be attributed to water-wire ordering length rather than to other structural differences such as hydrogen-bond strength and density.","fun_headline_variants_meta":{"raw":{"variants":["8 eV exciton peak exposes water wires","Water wire ordering measured by 8 eV absorption intensity","Ice XI's 8 eV peak signals proton-ordered water wires","Charge-transfer exciton tracks water-wire length in ice","8 eV absorption reveals water-wire ordering"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000511,"raw_usage":{"total_tokens":2509,"prompt_tokens":995,"completion_tokens":1514,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":611,"completion_tokens_details":{"reasoning_tokens":1438}},"tokens_in":611,"tokens_out":1514,"duration_ms":10848,"temperature":1.0,"reasoning_tokens":1438,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:00:33.519700+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the 8 eV absorption peak in a series of ices whose proton-ordering correlation length is independently varied — for example KOH-doped ice annealed to different degrees — and check whether the peak intensity rises monotonically with the independently measured wire-ordering length; the paper's mechanism predicts monotonic growth, while a flat or saturated curve would falsify the length-scaling claim.","supporting_citations":[{"cited_title":"Zhang, F","cited_arxiv_id":null,"evidence_quote":"Provides the machine-learned molecular dynamics model used to generate liquid-water configurations."},{"cited_title":"Garbuio, M","cited_arxiv_id":null,"evidence_quote":"Prior identification of the main absorption peak as a charge-transfer exciton, which the paper extends to water wires."},{"cited_title":"Hayashi and N","cited_arxiv_id":null,"evidence_quote":"Experimental optical absorption spectrum of liquid water against which the theory is benchmarked."},{"cited_title":"Kobayashi, J","cited_arxiv_id":null,"evidence_quote":"Experimental optical absorption spectrum of ice used for comparison."},{"cited_title":"Tajima, T","cited_arxiv_id":null,"evidence_quote":"Establishes ice XI as the proton-ordered ferroelectric phase whose infinite wire defines the long-range-ordering limit."}],"review_version":1}