{"id":"8c08fffc-5e0b-47d1-bdf0-f6147cff48bb","arxiv_id":"2607.14764","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Calcium silicate hydrate shows a boson peak near 1 THz whose intensity and elastic-heterogeneity parameter vary systematically with Ca/Si, with a coherence maximum near Ca/Si ≈ 1.0.","lead":"This paper measures a vibrational anomaly called the boson peak in calcium silicate hydrate, the binding phase of cement, using two independent methods that agree on its frequency and reveal how it changes with composition. A smart generalist might read it because it offers a new experimental window into nanometer-scale stiffness variations that may govern concrete creep and heat flow.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Calorimetric BP frequency is not independently established: raw kBT/h gives 0.14–0.24 THz, the ×4–5 correction is borrowed, and the Planck fit is explicitly non-unique and THz-anchored.","rationale":"The reader identified the borrowed ×4–5 correction as the weakest assumption. This is on target, but the paper also offers the Planck-like soft-mode fit and a forward DHO model, which partially address the raw conversion. However, on close reading, neither is fully independent: the Planck fit is explicitly non-unique and THz-anchored, and the forward model uses the THz frequency as input. Thus the concern is broader than the ×4–5 factor: the calorimetric dataset alone, taken as an independent probe, does not pin the BP to 1 THz. This supports the reader's CONDITIONAL verdict without changing it. I do not believe the concern is fatal to the THz identification of the BP, because the THz spectral feature is consistent and composition-dependent, but the title-level dual-probe frequency agreement is overstated as it currently stands.","tokens_in":24628,"tokens_out":12496,"duration_ms":110397,"concrete_test":"Recompute the forward DHO prediction of Cp/T3 (Appendix A.1) with ν0 fixed at the raw calorimetric value of 0.2 THz (instead of the THz-fit ~1 THz), rescaling the linewidth to preserve the low-frequency ω^4 quasi-localized asymptote, and compare the predicted Cp/T3 hump to the measured excess within the ±3% experimental uncertainty. If the prediction matches equally well, then the calorimetric data cannot discriminate the 1 THz assignment, confirming that the apparent dual-probe agreement is not independently established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that calorimetry independently locates the BP near 1 THz is not secured. Three routes are offered. (i) The raw conversion f = k_B T_BP/h in Table 4 gives 0.14–0.24 THz, far below 1 THz. (ii) An empirical ×4–5 factor borrowed from Ando et al. [46] for SiO2–Al2O3 and lithium borate glasses is applied in Table 4, but §4.1 itself reports composition-dependent ratios ν0/(k_B T/h) from 4.3 to 7.5, showing the correction is not material-independent. (iii) The Planck-like SPM fit in §3.4 yields νBP = 1.13–1.34 THz, but Appendix A.1 explicitly states the heat capacity does not uniquely select the Planck form — “any asymmetric VDOS sharing the ω4 limit fits comparably” — and that “the absolute frequency is fixed by THz.” The forward DHO→Cp prediction is not an independent determination because it uses the THz-derived DHO parameters as input; it tests consistency, not the frequency. Moreover, Appendix A.1 admits a single VDOS cannot reproduce both THz lineshape and Cp without a frequency-dependent coupling, undermining the direct DHO→Cp mapping. The dual-probe “same frequency” claim therefore rests on one direct THz measurement plus a calorimetric analysis whose frequency assignment is either borrowed or model-prior-dependent.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a systematic THz-TDS and low-temperature calorimetry study of the boson peak in synthetic calcium silicate hydrate (C-S-H) across five Ca/Si ratios (0.5–1.7). After Bruggeman effective-medium correction for Ca(OH)2 and CaCO3 impurities, the authors identify a boson-peak feature in the normalized THz dielectric loss at 0.94–1.05 THz via a damped harmonic oscillator fit, and a Cp/T3 excess hump at 7–12 K. They argue that both probes agree on a ~1 THz characteristic frequency after an empirical correction factor, and that the THz oscillator strength S and calorimetric peak height fall monotonically with Ca/Si while the THz peak height S·ν0/Γ is maximal at Ca/Si ≈ 1.0. This decoupling is attributed to a structural crossover between silicate-chain depolymerization and interlayer calcium filling. From the DHO/CPA analyses the authors extract a medium-range dynamical correlation length of ~0.3–2 nm and a coherent-potential elastic-heterogeneity parameter γ that decreases from 0.98 to 0.48 with increasing Ca/Si. The paper concludes that the boson peak provides a dynamical descriptor of medium-range elastic heterogeneity relevant to creep and thermal transport in C-S-H.","tokens_in":25081,"tokens_out":4362,"duration_ms":41101,"significance":"If the central claims hold, this is a valuable contribution: a first systematic dual-probe characterization of the boson peak in C-S-H, with composition-resolved trends that can be confronted by simulations, a forward DHO-to-Cp consistency check that reproduces the calorimetric excess for four of five compositions, explicit falsifiable predictions for creep and SCM reactivity, and a physically plausible mapping between THz spectral parameters and medium-range elastic heterogeneity. The impurity corrections and sensitivity analyses are careful, and the comparison with MD predictions and silicate-glass trends is informative. The main significance is the demonstration that C-S-H behaves like a disordered silicate network with a low-frequency vibrational anomaly near 1 THz, and the suggestion that the CPA parameter γ could serve as a composition-resolved disorder descriptor. However, the strength of the dual-probe 'same frequency' claim and the status of γ as a measured quantity are not yet fully secured, as detailed in the major comments.","major_comments":[{"comment":"The claim that calorimetry independently locates the BP near 1 THz is not secured. The raw conversion f = kB T_BP/h gives 0.14–0.24 THz (Table 4), well below 1 THz. The ×4–5 correction factor is borrowed from ref. [46] for SiO2–Al2O3 and lithium borate glasses, yet §4.1 itself reports composition-dependent ratios ν0/(kB T_BP/h) of 4.3–7.5, showing that the factor is not material-independent. More importantly, Appendix A.1 states that the heat capacity does not uniquely select the Planck form, that 'any asymmetric VDOS sharing the ω4 limit fits comparably,' and that 'the absolute frequency is fixed by THz.' The Planck-like SPM fit is therefore not an independent frequency determination, and the forward DHO→Cp prediction uses the THz-derived DHO parameters as input, testing consistency rather than frequency. Since the abstract and conclusions state that both probes locate the BP near 1 THz","section":"§4.1, Table 4"},{"comment":"The paper presents γ as 'to our knowledge the first experimental measure of this quantity' and as the central composition-resolved descriptor, but γ is a fitted parameter of the CPA/SCBA model applied to the same ε''(ν)/ν spectra used to define the DHO boson peak. The agreement between γ and Γ/ν0 to within ~15% is therefore a correlation between two fits of the same dataset, not an independent validation. This circularity should be addressed explicitly, for example by determining γ from an independent observable (sound velocity, inelastic neutron scattering, or a forward relation with known uncertainty) or by clearly labeling γ as a model-defined effective parameter whose compositional trend, not absolute value, is the claim. As written, the discovery framing in the abstract and conclusions ('coherent-potential elastic-heterogeneity parameter') exceeds what the fitting procedure alone ca","section":"§4.2, Table 5"},{"comment":"The DHO fit parameters are reported without uncertainties, although the CPA fits in Table 5 carry ± errors. The key intensity-decoupling claim — that S decreases monotonically with Ca/Si while the DHO peak height S·ν0/Γ reaches a maximum at Ca/Si = 1.0 — rests on the relative values of S and Γ across five compositions. Without confidence intervals or a sensitivity analysis covering the fit range (0.5–2.5 THz), the background parameterization, and the fixed frequency choice ν ≈ ν0 for the peak height, it is not possible to assess whether the maximum at Ca/Si = 1.0 is statistically significant. Please add bootstrap or covariance-based error bars and test the non-monotonic trend against reasonable variations of the fit window and background model.","section":"Table 2"}],"minor_comments":[{"comment":"The formula n(ν) = 1 + c·Δφ/(2πνd) assumes unwrapped phase. Please state explicitly how phase unwrapping was performed and how its uncertainty propagates into the dielectric loss.","section":"§2.3"},{"comment":"The actual Ca/Si ratios of the C-S-H phase (0.54, 0.90, 1.14, 1.46, 1.57) differ from the nominal values (0.5–1.7). The text often refers to nominal ratios without consistently stating that actual values are used. Please define on which numbers the analysis and figures are based.","section":"Table 1 and throughout"},{"comment":"The forward prediction is described as 'parameter-free' (e.g., 'the most stringent cross-check is parameter-free'), but it uses the THz-derived DHO spectral function as input. The prediction has no adjustable BP frequency, but it is not parameter-free in the literal sense; 'parameter-fixed' or 'no additional adjustable parameters' would be more precise.","section":"§3.4 / Appendix A.1"},{"comment":"Reference [26] appears with inconsistent capitalization ('Li  zhanguo') and a 2026 online date; please verify the citation and standardize the author list.","section":"References"},{"comment":"The three length conventions (ξ1, ξ2, λIR) are presented as 'order-of-magnitude estimates,' which is appropriately cautious. Please explicitly note in the text that the choice of G = 9 GPa without an error bar affects all three lengths, and consider adding a brief statement on how the uncertainty in G affects the correlation-length conclusions.","section":"§4.3, Table 6"}],"recommendation":"major_revision","confidential_remarks":"The experimental work is substantial and the manuscript is generally careful about corrections and sensitivity. The reader's concern about the calorimetric frequency determination is well founded and lands on the central dual-probe claim; the paper should be revised to distinguish independent frequency determination from consistency checks. The γ descriptor also needs reframing to avoid model-fit circularity. With these revisions, the paper could become acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: this paper gives cement science the first systematic dual-probe characterization of the boson peak in C-S-H, and the THz dataset alone is a real contribution. But the calorimetric half of the 'same frequency' claim does not stand on its own. The raw conversion puts the calorimetric peak at 0.14–0.24 THz, and the leap to ~1 THz depends on a ×4–5 correction borrowed from silicate and borate glasses. The paper itself reports ratios from 4.3 to 7.5 for its own samples, which shows the factor isn't material-independent. The Planck-like fit in Section 3.4 yields 1.13–1.34 THz, but Appendix A.1 admits the heat capacity doesn't uniquely select that form and the absolute frequency is fixed by THz. So the cross-validation is really one direct measurement plus a consistency check, not two independent determinations.\n\nWhat is new and good: five Ca/Si ratios, careful impurity corrections via Bruggeman and TGA subtraction, DHO fits with sensible parameters, a composition-resolved CPA disorder parameter γ that decreases from 0.98 to 0.48, and the Ca/Si≈1 decoupling between spectral weight and peak height. The forward DHO→Cp prediction reproducing four of five compositions is a nice parameter-free check of internal consistency—just not an independent frequency measurement. The comparison with MD trends is fair, and the paper is unusually honest in the appendix about the non-uniqueness of calorimetric models. That honesty earns credit.\n\nSoft spots, in order of severity. First, the 'same frequency' claim should be reworded: calorimetry alone doesn't establish ~1 THz; it establishes an excess with a peak temperature whose conversion to frequency needs a model or an empirical factor. Second, γ is a fit parameter of the CPA model applied to the same spectra that also give Γ/ν0; the agreement between them is two fits to the same data, not independent validation. Third, the choice to use the second Ca/Si=1.7 heat-capacity run is unexplained—state why the first was discarded. Fourth, only γ has error bars; the DHO parameters don't, so significance of the trends is unclear.\n\nWho is this for: anyone working on C-S-H structure, cement creep, or thermal transport, and glass physicists interested in a heavily damped boson peak in a hydrated disordered solid. I'd send it to a serious referee. The core THz observation is solid; the overclaim is in the framing. A revision that fixes the calorimetric conversion language and the Ca/Si=1.7 question would make it a strong paper.","headline":"A solid THz dataset and an overreaching cross-validation claim: the boson peak in C-S-H is real, but calorimetry does not independently place it at 1 THz.","tokens_in":25567,"tokens_out":2352,"would_cite":true,"duration_ms":21218,"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":"This paper establishes that calcium silicate hydrate, the binding phase of concrete, exhibits a boson peak near 1 THz—detected independently by terahertz spectroscopy and low-temperature calorimetry—and that the peak's parameters yield a qu","keywords":["boson peak","calcium silicate hydrate","terahertz time-domain spectroscopy","low-temperature heat capacity","elastic heterogeneity","Ca/Si ratio","medium-range order","vibrational density of states"],"falsifier":"Measure the vibrational density of states of the same five samples directly by inelastic neutron scattering. If the maximum of g(ω)/ω² is not near the terahertz peak frequency for at least four of the five compositions, the paper's central frequency assignment—and the calorimetric shift factor that supports it—fails.","tokens_in":24537,"feed_emoji":"🧱","tokens_out":6318,"duration_ms":61930,"temperature":0.7,"pith_summary":"This paper aims to show that calcium silicate hydrate (C-S-H)—the phase that binds concrete together—carries the same low-frequency vibrational anomaly, the boson peak, that is universal in glasses, and that this anomaly can be measured directly in cement science for the first time. Using terahertz spectroscopy and low-temperature calorimetry on synthetic C-S-H samples with five calcium-to-silicon ratios, the authors locate the peak near 1 THz, extract from it a nanometer-scale correlation length, and derive a parameter γ that quantifies how strongly the stiffness fluctuates from place to place on the medium range. If correct, this gives cement researchers a quantitative handle on the elastic heterogeneity that controls creep and thermal transport, and it opens the same probe to the amorphous supplementary materials used in low-carbon cements.","feed_headline":"Two probes pin cement binder's boson peak near 1 THz","feed_subtitle":"Terahertz spectra and heat-capacity data agree on the vibrational fingerprint of nanometer-scale stiffness disorder in C-S-H.","key_machinery":"The central object is the boson peak, defined as the broad hump in the reduced vibrational density of states g(ω)/ω² (or in the frequency-normalized dielectric loss ε''(ν)/ν) that signals an excess of low-frequency modes beyond the Debye prediction. The paper extracts it from terahertz spectra using a damped harmonic oscillator plus power-law background, and from heat capacity via the Cp/T³ hump, then interprets the peak through the spatially heterogeneous elasticity picture: a self-consistent Born approximation in which the dimensionless variance γ of shear-modulus fluctuations controls both the peak shape and the phonon scattering rate. The mapping from frequency to length, ξ = vt/(2πν0),","core_discovery":"The paper establishes that C-S-H, the binding phase of hardened cement, displays a boson peak—an excess of low-frequency vibrational states over the Debye prediction—located near 1 THz, detected independently by terahertz time-domain spectroscopy and low-temperature heat-capacity measurements on synthetic samples spanning Ca/Si = 0.5–1.7. After correction for crystalline impurities, both probes place the peak at ~1 THz; the THz peak height and the calorimetric Cp/T3 hump diverge in intensity with composition, revealing that the apparent strongest peak at Ca/Si ≈ 1.0 reflects the most coherent, least damped modes rather than the largest number of excess modes. The peak parameters yield a medi","pith_inferences":["If the link between γ and the energy-barrier distribution holds, terahertz screening of extracted C-S-H could become a faster surrogate for long-term creep testing than mesoscale mechanical methods, since γ is an intrinsic property of the solid phase rather than a packing-scale average.","A direct test the paper makes possible but does not run: measure the boson peak of C-S-H at controlled relative humidities to separate framework heterogeneity from the damping contribution of interlayer water, which could let the method work on saturated pastes.","Because the measurements are on powders, polarized or oriented-sample terahertz experiments could expose the sheet-vs-interlayer anisotropy of γ—a dimension the orientation-averaged values cannot show, and one the paper explicitly leaves open.","The same analysis could be applied to aluminium-substituted C-S-H, where the boson peak might reveal whether aluminium insertion stiffens or disorders the medium-range network—directly relevant to slag and metakaolin blends."],"forward_implications":["If correct, the coherent-potential parameter γ gives the first quantitative, composition-resolved measure of elastic heterogeneity in C-S-H, with γ falling from 0.98 at Ca/Si = 0.5 to 0.48 at Ca/Si = 1.7.","The ~1 nm dynamical correlation length (0.3–2 nm under different conventions) fills the gap between the ~0.16 nm Si–O coordination shell and the ~5 nm colloidal packing unit, giving experiment access to medium-range order in C-S-H.","The Debye-normalized boson-peak frequency (0.15–0.17) places C-S-H in the same family as modifier-rich silicate glasses, so the low-frequency dynamics of cement's binding phase are governed by the same physics as conventional glass formers.","The intensity decoupling at Ca/Si ≈ 1.0—peak height maximum with falling spectral weight—marks a structural crossover between silicate-chain depolymerization and interlayer calcium filling, a composition window relevant to low-carbon blended cements.","The dual-probe strategy is transferable to other amorphous solids, including fly ash, slag, and calcined clays used in low-carbon cements, where the boson peak could serve as a structure-based descriptor."],"fun_headline_variants":["Two probes agree: cement binder's boson peak at 1 THz","Calorimetry and THz pin cement's boson peak near 1 THz","Boson peak in C-S-H pinned at 1 THz by two probes","THz + calorimetry locate cement's boson peak at 1 THz","Boson peak in cement binder: agreement at 1 THz"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The entire dual-probe agreement rests on the assumption that the factor of 4–5 used to convert the calorimetric peak temperature into a frequency—borrowed from melt-quenched silicate glasses—also applies to the heavily damped, water-bearing C-S-H; if that factor is different, calorimetry alone puts the peak at 0.14–0.24 THz rather than ~1 THz.","fun_headline_variants_meta":{"raw":{"variants":["Two probes agree: cement binder's boson peak at 1 THz","Calorimetry and THz pin cement's boson peak near 1 THz","Boson peak in C-S-H pinned at 1 THz by two probes","THz + calorimetry locate cement's boson peak at 1 THz","Boson peak in cement binder: agreement at 1 THz"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002533,"raw_usage":{"total_tokens":9618,"prompt_tokens":894,"completion_tokens":8724,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":638,"completion_tokens_details":{"reasoning_tokens":8621}},"tokens_in":638,"tokens_out":8724,"duration_ms":52346,"temperature":1.0,"reasoning_tokens":8621,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T01:05:13.931660+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the vibrational density of states of the same five samples directly by inelastic neutron scattering. If the maximum of g(ω)/ω² is not near the terahertz peak frequency for at least four of the five compositions, the paper's central frequency assignment—and the calorimetric shift factor that supports it—fails.","supporting_citations":[],"review_version":1}